Cuaderno De Documentacion

SECRETARIA DE ESTADO DE ECONOMÍA,

MINISTERIO SECRETARÍA GENERAL DE POLÍTICA ECONÓMICA DE ECONOMÍA Y ECONOMÍA INTERNACIONAL Y HACIENDA SUBDIRECCIÓN GENERAL DE ECONOMÍA INTERNACIONAL

CUADERNO DE DOCUMENTACION

Número 86-16º (alcance)

Alvaro Espina Vocal Asesor 29 Octubre de 2008 (20 horas)

CD 86-16º de alcance 29 de Octubre de 2008 “Beijing responded favourably to Korea's proposal for a regional bail-out fund, but Tokyo deferred, fearing that this would be dominated by China, given that country's immense dollar reserves. Tokyo then proposed funneling Asian reserves through the IMF, but China deferred, fearing that this initiative would be dominated by Japan, which has long participated in IMF deliberations. Financial diplomacy is evidently more difficult than in 1944...... , move quickly from the leaders' meeting, which is largely about the photo-ops, to the meeting of finance ministers, where the real business will occur. And once there, focus on pragmatic reforms. Clamp down on regulatory arbitrage. Raise capital requirements. Make the regulatory regime less procyclical. Use taxes and regulation to drive transactions in credit default swaps and other derivative instruments into an organised exchange. This modest approach will not be hailed as a New World Financial Order. But it will be a useful first step toward making the world a safer financial place. And it will minimise the danger that the new Bretton Woods conference will go down in history as a failure.” Barry Eichengreen: “New world pragmatism” Soluciones baratas, limpias, rápidas, arbitrajistas y, a ser posible, sin coste para el contribuyente. Pragmatismo. Eso es lo que necesitamos para salir de la crisis. El CD 86-15º proporcionaba dos fórmulas de este tipo para hacer frente al Armagedón de los hedge funds y de los mercados de divisas (anticipándose al desplazamiento del epicentro de la crisis hacia las quiebras de los deudores soberanos, tras extenderse, dentro de EEUU, a las tarjetas de crédito y a las grandes empresas de automoción y tras la amenaza que ya se cierne sobre la Europa del Este): el “hedgy- corralito”, y el fondo de estabilización gestionado por un fiduciario internacional por cuenta de los países aportantes (al que denominaremos Cristina-proof-IMF). Este último parece gustarle también a Japón, aunque China lo condicionaría, obviamente, a tener algo que decir en ese foro, lo que señala la buena dirección a seguir en la cumbre de Noviembre. (En

1 cualquier caso, la admisión de China en cualquier nuevo foro económico internacional debería condicionarse al respeto por su parte de los derechos laborales fundamentales, supervisado por la OIT. Mientras esto no ocurra, sólo debería participar en iniciativas parciales, como la del consejo de administración del fondo de estabilización, pero no en el del FMI). Una cumbre que para Barry Eichengreenn debe dar paso enseguida a las discusiones técnicas, que es de donde saldrá algo práctico (frente a las elevadas espectativas que ponen en la cumbre-cumbre Paul Rogers y Bernard-Henri Levy: este último considera que hasta entonces el Apocalipsis se encuentra en suspenso). Eichengreen presenta también un paper ante la NBER en la que estudia los riesgos de un boom seguido de un pinchazo en Polonia, a la vista de lo ocurrido con la etapa pre-euro y post- euro en Chipre, Grecia, Irlanda, Malta, Portugal y España. Luigi Zingales –tras hacer la contabilidad astronómica del rosario de planes de rescate que amenazan con asfixiarnos- aporta en Enonomists’ Voice otras dos soluciones, también muy baratitas y capaces de conjugar reestructuración e interés de los partícipes (haciéndoles participar tambien un poquito del riesgo en que han incurrido). Como Zingales es uno de los mayores expertos mundiales en economía y finanzas de la quiebra, sus soluciones son de ese tipo y sólo tienen costo para los reguladores (que deberían actuar deprisa, par que cuando sus normas entren en vigor quede algo por reparar, especialmente en el caso de los hogares hipotecados en las zonas de mayor desigualdad): 1.- Una nueva figura estandarizada de restructuración de deuda hipotecaria, bajo protección judicial (obligatoria para el prestamista, pero voluntaria para el prestatario insolvente): Se reduciría el monto total de la hipoteca en la misma proporción en que han caído los precios de las casas en el correspondiente distrito desde el momento en que se suscribió la hipoteca (utilizando para ello el índice Case-Schiller de precios de la vivienda por distrito postal). En contrapartida, el prestamista adquiriría un derecho al 50% de la diferencia entre el valor de la hipoteca revisada y el precio de venta de la vivienda en el momento en que el prestatario se desprenda de ella, durante el período de vida de la hipoteca. (En caso de cancelación anticipada, con venta posterior, la distribución de la

2 diferencia debería ser proporcional al peso de la hipoteca pendiente de pago en el momento de la decisión judicial de reestructuración, respecto al valor reestimado de la vivienda en esa decisión).

2.- Nueva figura de reestructuración bancaria instantánea (pre-packaged) bajo supervisión judicial, alternativa a la entrada del Tesoro en el capital de los bancos. En caso de grave insolvencia bancaria, se ofrece a los accionistas la posibilidad de abonar la parte alícuota de la deuda pendiente (excluida la deuda a corto plazo). De rechazarse, en todo o en parte, el accionista perdería esos derechos y el juez (con facultades Cramdown) los transferiría a los titulares de la deuda. Un saneamiento de este tipo se exigiría a todos los bancos en dificultades para seguir acudiendo a la ventanilla de redescuento de la Fed. Si los titulares de la deuda son instituciones que no pueden invertir en acciones, se les concede un plazo de dos años para desprenderse de ellas. Esta figura no daría lugar a la activación de los seguros contra quiebra (CDS). Y, siguiendo con propuestas sin coste para el contribuyente, Stephen Roach reaparece en escena y sugiere desde FT, sin alzar demasiado la voz, que el Congreso debería redefinir el mandato político de la Fed, añadiendo a las cuatro palabras que lo han definido hasta ahora (full employment & prices stability) dos más: financial stability. Aunque con gran discrección, el desterrado Stephen Roach (antes llamado el avinagrado, convertido ahora en el que le canta las verdades a América) sugiere cosas muy sencillas para corregir el “enfoque repudiable” de una Fed que acabó especializándose en la “negligencia regulatoria” y en la “patochada sistemática”: Sustituir la convicción ideológica por el sentido común1; la política reactiva por políticas preventivas; la confianza en las verdades ideológicas por la medición objetiva de los fenómenos financieros. Aprender a ir contra la corriente, en lugar de echar leña al fuego de los mercados de activos; a exigir requerimientos marginales para el crédito durante el boom, en lugar de hacer la vista gorda a la aparición de productos financieros exóticos a tipos de interés nulo. Todo ello, sin olvidar que la Fed, como toda institución pública, tiene también una misión de persuasión moral sobre los comportamientos de los agentes financieros, contrapuesta a la manga ancha ante el juego

1 Larry Kudlow piensa que el sentido común actual consiste en aplicar sin dogmatismos una mezcla de políticas: Reagan+Friedman+Keynes.

3 de moral hazard que se practicaba abiertamente bajo sus ojos, haciendo pasar por innovación la peor basura de la ingeniería financiera2. Todo eso cabe sólo en dos palabras, añadidas al mandato de la Fed: estabilidad financiera. Decir esto ahora no tiene mérito, cuando todo el mundo rivaliza en saldar cuentas con la Fed (y con algunos más: véase el rosario de responsabilidades que Steve H. Hanke enumera en Forbes, aunque desde las posiciones fundamentalistas defendidas por CATO). El mérito estaba en decirlo apoyándose exclusivamente en la teoría económica comúnmente admitida y en la evidencia más fácilmente observable, pero cuando nadie quería escucharlo y cuando el que lo decía se arriesgaba al destierro a Asia desde uno de los puestos más brillantes para un economista en la capital financiera del mundo. Porque la crisis no sólo está permitiendo depurar los mercados, sino también las profesiones que se ocupan de ellos, y especialmente la de los economistas (por no hablar de los amos del universo, que no salen de su dorado asombro, retirados a su refugio de Greenwich, todavía conmocionados por la bomba atómica financiera que ellos mismos arrojaron, pero sin el menor asomo de arrepentimiento: eso si, buscando ahora verdades más sólidas...., en la parroquia del lugar). Y entre todos los economistas que se han dejado ver durante los roaring nineties y los no-tan-alegres Twenty-first century beginnings, Roach es probablemente el que ha salido deontológicamente más incólume: se la jugó y pareció salir derrotado, hasta que la debacle vino a colocar a cada uno en su sitio. Sin esa calidad de roca incorruptible no se puede dirigir la nueva Fed. Stephen Roach es el hombre. Si, como dice Natalia Redesberg lo que se ha llevado por delante a Wall Street ha sido la desconfianza, ésta no volverá mientras no se sitúe al frente de la supervisión a alguien de la máxima confianza y experiencia en los mercados. Pero para eso debe haber un cambio en Washington y seguramente lo habrá si

2 A título de ejemplo, véase “La Anatomía de un colapso ficticio de Fondo de alto riesgo”, publicada en 2005 por Rob Kirby (incluida en este CD).

4 no se produce un “desastre para el derecho de voto” el próximo9 día 5, como se teme Christopher Edley Jr, decano de Berkeley. Y ¿qué decir de los hedgies? Sobre la necesidad de una regulación estricta hay mucho papel en este CD. Para David Weidner, mientras no se regule su actuación no habrá verdadera calma en los mercados (porque es bien sabido que son una bomba de relojería, pero nadie sabe de cuánta potencia ni cuando estallará: véase la narración ficticia de Rob Kirby). Sólo estos últimos meses la Fed ha empezado a hacer público lo que sabe sobre los procelosos mundos de los derivados y los fondos de alto riesgo, pero se trata de un problema global que afecta también muy seriamente a Europa, como señala David Oakley. Pero una de las características más sobresalientes de la crisis actual es la puesta en cuestión de la aportación al bienestar general, al crecimiento y a la eficiencia económica (tanto para la asignación de factores como para la distribución de los recursos) de áreas enteras de la actividad financiera dominante durante los últimos decenios, como el “fallo” de los fondos de inversión en empresas no cotizadas (private equity: Steven M. Davidoff). Pero lo más contundente es la lógica analítica y la evidencia recogida por Geetesh Bhardwaj, Gary B. Gorton, y K. Geert Rouwenhorst, según la cual, de unos beneficios “producidos” por la actividad específica de los hedge funds (en particular, por los Commodity Trading Advisors: CTA) equivalentes al 5,4% del negocio antes de aplicar comisiones entre 1994 y 2007, los clientes “aconsejados” sólo han recibido 85 pb. O sea, una cuantía insignificantemente distinta de cero, de modo que casi les habría dado igual orientar sus inversiones por criterios automáticos o “ciegos”, porque los CTA sólo generan verdadera riqueza “alfa” para ellos mismos, fruto de la opacidad y la falta de transparencia con la que vienen funcionando,3 que se traduce, además en una aceleración de los ciclos de ducha escocesa financiera que causa gravísimos daños en las economías emergentes, como se deduce del estudio de Enrique G. Mendoza contenido en este CD, daños que Soros considera imprescindible reparar mediante un rescate internacional.

3 “Fooling Some of the People All of the Time: The Inefficient Performance and Persistence of Commodity Trading Advisors”

5 BACKGROUND PAPERS:

1. Even as Dow Soars 11%, Skeptics Lurk, by Vikas Bajaj....9 2. A Detroit Bankruptcy Beats a Bailout, by Steven Pearlstein....12 3. Consumers Feel the Next Crisis: Credit Cards, by Eric Dash....14 4. America must lead a rescue of emerging economies, by George Soros....17 5. Add ‘financial stability’ to the Fed’s mandate, by Stephen Roach...19 6. Plunging markets are far from irrational. There is a simple explanation for falls in share prices - investors fear a depression is coming, by Oliver Kamm....21 7. Greenspan vs. Buffett, by Rick Newman....23 8. “How far back" are we?, by Barry Ritholtz....25 9. Loss Souls. A Day of Reckoning Dawns for Denizens of Hedge- Fund Heaven, by David Segal....28 10. Distrust Toppled Investment Banking Industry, by Natalia Redenbaugh....32 11. Regulate hedge funds now. Commentary: The looming threat is too great to let funds run free, by David Weidner's....36 12. White House Explores Aid for Auto Deal , by Edmund L. Andrews and Bill Vlasic....38 13. Hardships Past Haunt Europe’s Search for Financial Safety , by Katrin Bennhold....41 14. In Praise of Bankruptcy, by Henry Thompson....43 15. A Voting Rights Disaster?, by Christopher Edley Jr....45 16. The Behavioral Revolution , by David Brooks....47 17. The Age of Prosperity Is Over. This administration and Congress will be remembered like Herbert Hoover, by Arthur B. Laffer....49

6 18. We Need Reagan + Friedman + Keynes, by Larry Kudlow....52 19. A Suspended Apocalypse, by Bernard-Henri Levy....54 20. Look who pays for the bailout, by Shawn Tully with Joan Caplin....56 21. But Have We Learned Enough?, by N. Gregory Mankiw....60 22. The Next New Deal. The huge opportunities—and huge risks—of a possible Obama administration., by John Heilemann....62 23. Russia's CDS Spreads Spike: How Much Is Russia's Default Risk Rising?, RGE....64 24. Welfare for Detroit, W....68 25. Shelter From The Storm, by Steve H. Hanke....69 26. New world pragmatism, by Barry Eichengreen....71 27. Doing Business in Spain, by Jaime Pozuelo-Monfort....73 28. Foreclosure prevention efforts and market stability, by Governor Elizabeth A. Duke....75 29. Over-the-counter derivatives, by Patrick M. Parkinson....77 30. OTC to become UTC, taking ICAP and TP with’em?, Posted by Paul Murphy....82 31. Is this the real reason for today’s equity market gains?, Posted by Neil Hume....82 32. CDS report: European credit derivatives at record wides, by David Oakley....83 33. The Anatomy of a Fictional Hedge Fund Collapse, by Rob Kirby....87 34. LTCM Revisited - A Forensic Account, by Rob Kirby 35. Sudden Stops, Financial Crises and Leverage: A Fisherian Deflation of Tobin's Q, by Enrique G. Mendoza....90 36. A crisis-opportunity moment, by Paul Rogers....124

7 37. Is Poland at Risk of a Boom-And-Bust Cycle in The Run-Up to Euro Adoption?, by Barry Eichengreen and Katharina Steiner....128 38. Fooling some of the People all of the Time: The Inefficient Performance and Persistence of Commodity Trading Advisors, by Geetesh Bhardwaj, Gary B. Gorton and K. Geert Rouwenhorst....179 39. The Financial Crisis and Sustainable Security, by Paul Rogers....216 40. Credit Derivatives and Bank Credit Supply, by Beverly Hirtle....220 41. The Failure Of Private Equity, by Steven M. Davidoff....264 42. Urban Inequality, by Edward L. Glaeser, Matthew G. Resseger, Kristina Tobio....326 43. Plan B, by Luigi Zingales....389

Addenda Release Date: October 29, 2008 For immediate release The Federal Open Market Committee decided today to lower its target for the federal funds rate 50 basis points to 1 percent. The pace of economic activity appears to have slowed markedly, owing importantly to a decline in consumer expenditures. Business equipment spending and industrial production have weakened in recent months, and slowing economic activity in many foreign economies is damping the prospects for U.S. exports. Moreover, the intensification of financial market turmoil is likely to exert additional restraint on spending, partly by further reducing the ability of households and businesses to obtain credit. In light of the declines in the prices of energy and other commodities and the weaker prospects for economic activity, the Committee expects inflation to moderate in coming quarters to levels consistent with price stability. Recent policy actions, including today’s rate reduction, coordinated interest rate cuts by central banks, extraordinary liquidity measures, and official steps to strengthen financial systems, should help over time to improve credit conditions and promote a return to moderate economic growth. Nevertheless, downside risks to growth remain. The Committee will monitor economic and financial developments carefully and will act as needed to promote sustainable economic growth and price stability. Voting for the FOMC monetary policy action were: Ben S. Bernanke, Chairman; Timothy F. Geithner, Vice Chairman; Elizabeth A. Duke; Richard W. Fisher; Donald L. Kohn; Randall S. Kroszner; Sandra Pianalto; Charles I. Plosser; Gary H. Stern; and Kevin M. Warsh. In a related action, the Board of Governors unanimously approved a 50-basis-point decrease in the discount rate to 1- 1/4 percent. In taking this action, the Board approved the requests submitted by the Boards of Directors of the Federal Reserve Banks of Boston, New York, Cleveland, and San Francisco.

8 Business

October 29, 2008 Even as Dow Soars 11%, Skeptics Lurk By VIKAS BAJAJ A tug of war is in full swing on Wall Street and those pulling for stocks came out way ahead on Tuesday for the first time in a while.

After four mostly miserable weeks, a powerful afternoon rally left traders wondering if it was time to buy again. Shares, the bulls argued, have become too cheap to resist, despite signs of trouble in the economy. Many other investors, however, remained unpersuaded. At about 2 p.m., the market exploded into one of its biggest rallies since World War II, with the Dow Jones industrial average closing up 889.35 points, or 10.9 percent, to 9,065.12. In the last 69 years, the Dow has gained that much on only one other day, and that was two weeks ago, on Oct. 13. There was no single catalyst for the surge, and market specialists said investors seemed to be coming around to the idea that stocks were worth buying, given that the Dow had plunged 32 percent since the end of August. By some measures, stocks are cheaper than they have been in decades. Investors also may have also been looking ahead to a Wednesday meeting at the Federal Reserve, at which policy makers are expected to cut interest rates again. “Circle today as one of those days that the fundamental issues trumped panic and fear,” said Robert J. Froehlich, vice chairman and chief investment strategist with DWS Investments. But, he added, he was not ready to declare that stocks would not fall below the closing level on Monday.

9 The big question on the minds of investors on Wall Street and Main Street, however, remains this: Have stocks fallen enough to reflect the steep declines in profits that are sure to accompany a potentially long global recession? While some prominent investors like Warren E. Buffett and R. Jeremy Grantham, who had been bearish in the past, have in recent days said that they think stock prices had fallen far enough for them to start buying, others remain unpersuaded. “It really depends on how negative earnings are over the next couple of quarters,” said Barry Ritholtz, chief executive of Fusion IQ, an investment and research firm, who is not ready to declare that stocks have hit their low mark but says he thinks that point may be nearing. “If earnings are worse than expected, we will run into some problems.” The weakness in the economy was evident even as stocks were rising on Tuesday. Early in the day, the Conference Board reported that consumer confidence fell to the lowest level ever registered in the 41-year history of the firm’s survey. Another report showed that home prices in 20 major markets fell 16.6 percent in August, a slightly bigger decline than the month before. The credit markets remained under stress. Interest rates on mortgages and investment-grade bonds rose, which means consumers and businesses will pay more to borrow money. The average rate for 30-year fixed-rate mortgages was 6.56 percent, up from 6.38 percent a week earlier, according to HSH Associates, a publisher of financial data. But investors appeared to give short shrift to those reports, which simply confirmed what had been clear for months: American consumers are in a deep funk, and the housing market is in for a painful and protracted decline. “For me, the best part about today is that the market went up in the wake of what was some really discouraging economic news,” said Stuart Schweitzer, global markets strategist at J.P. Morgan Private Bank. “When the markets go up on bad news, it holds out hope that the bad news has been digested.” Rather than those bleak economic numbers, some analysts were looking ahead to the meeting of the Fed’s rate-setting committee. Investors are betting that the Fed will cut its benchmark fed funds rate to 1 percent, from 1.5 percent. Earlier on Tuesday, the Fed said that lending in a short-term debt market surged after the central bank started a program under which it lends directly to corporations by purchasing their commercial paper. The amount of such debt issued by companies jumped nearly 46 percent on Monday, from Friday. The Fed is acquiring three-month commercial paper, a category that jumped to $67 billion on Monday, up from less than $8 billion a day for the last several weeks. Later in the day, GMAC, the troubled finance division of General Motors, said it was able to borrow from the Fed program through commercial paper backed by loans, a development that should help the company. The rally on Tuesday may partly reflect a growing confidence among investors that the recent moves by the Fed and Treasury Department will prevent more cataclysmic failures in the financial system, like the bankruptcy of Lehman Brothers, analysts said. “People are feeling much more comfortable that the financial system is stabilizing, and that allows them to focus on fundamental valuations of stocks,” said Todd Steinberg, head of equities and commodity derivatives at BNP Paribas-Americas. “The caveat to that is there is still a lot of economic issues to come.”

10 It is hard to know whether stock prices reflect a realistic assessment of the coming economic pain. As a group, stock analysts have been slow to reduce their expectations of future corporate profits. (History shows that Wall Street is typically slow to make adjustments to its forecasts at turning points.) Analysts as a whole are expecting only a 6 percent decline for 2008 in the profits of the companies that make up the Standard & Poor’s 500-stock index, and a 16.9 percent increase in earnings for 2009, according to a report by Steven C. Wieting, an economist at Citigroup. By contrast, Mr. Wieting and his colleagues are forecasting a 12.4 percent decline in profit for 2008, and a 13.5 percent drop in 2009. Still, Mr. Wieting said he thought stocks had fallen by more than earnings would through the end of next year, suggesting the market might have fallen too far in recent weeks. There are also technical reasons to expect stocks to rally, according to Mr. Ritholtz, who writes the popular economic and financial blog the Big Picture. Stocks had fallen so far, so quickly, that many shares were trading below where they should have been, given their prospects, he said. Many stocks were being dumped by hedge funds and other investors being forced to sell to pay back their investors and their lenders. “This is the most expected rally in the world,” he said. “People knew it was coming, but nobody knew when it was going to show up.” Some investors were forced to buy shares to cover their bets that stocks would fall in price. Among the worst hit were investors who had bet against the price of Volkswagen, the German automaker. Porsche, a rival carmaker, said on Tuesday that it would increase its stake in Volkswagen to 75 percent, forcing hedge funds to buy the shares to cover short sales. The price of Volkswagen shares nearly doubled on Tuesday, closing at 918 euros. The company briefly became the world’s most valuable. Following are the results of Tuesday’s Treasury auction of four-week bills and two-year notes: Michael M. Grynbaum contributed reporting.

At 2:24 PM ET 939.11–1.40–0.15%

A Detroit Bankruptcy Beats a Bailout By Steven Pearlstein Wednesday, October 29, 2008; D01 Not content with $25 billion in government loans to retool factories for fuel-efficient cars, the auto industry is already back at the trough, this time angling for a taxpayer investment in the firm that would result from a merger of General Motors and Chrysler. You can just imagine the pitch from the populists of the Michigan congressional delegation: If the government is willing to invest $250 billion to bail out pinstriped bankers, then the least it could do is throw an extra $10 billion to rescue the domestic auto industry and the millions of workers and retirees who depend on it. There's only one difference: The government will make money on its bank investment, while the GM-Chrysler deal is a lemon. As reported by Reuters, GM and Chrysler would have the Treasury invest $3 billion directly in the newly merged automaker in exchange for preferred shares with warrants, as with the banks. The government would take over $3 billion of the company's pension obligation. To deal with the industry's short-term liquidity problem, the government would also commit to buying $4 billion in commercial paper issued by the new company. All that for two companies whose market values today are each less than $4 billion. The rationale for this scheme is pretty simple: If nothing is done, the financial situation of both of these companies is so dire that one or both of them will be forced to file for bankruptcy protection in the next several months. And a bankruptcy filing, we are told, will send an already weakened economy over the cliff, wiping out 2 million jobs, shifting tens of billions of dollars in pension obligations to the government and lopping two percentage points off the nation's gross domestic product. Although somewhat exaggerated, there is a kernel of truth in this doomsday scenario: This would be a particularly bad time for the U.S. economy to have GM or Chrysler go under. It would have devastating effects on many communities in the Midwest, deal another blow to consumer and investor confidence, and put further strain on already shaky government budgets. But even with a government-financed merger, the companies are going to have to shrink by at least 25 percent to reflect the realities of a shrinking market and much-reduced market shares. That translates into the direct loss of an additional 40,000 jobs and the indirect loss of several hundred thousand more. There is simply no way to avoid this pain without making the company a permanent ward of the federal government. The real flaw in the government-financed merger proposal is that it spares the companies from bankruptcy reorganization, the very process they need to get their costs and structure in line with market realities. Only a bankruptcy court can reduce the burden of pension and health benefits to 600,000 retirees that are slated to cost the companies $90 billion over the next decade. Only a bankruptcy court can override the state laws that make it difficult and expensive for Chrysler and GM to pare back a combined network of 10,000 dealerships, about 10 times more than Toyota has in the United States.

12 And only a bankruptcy court can impose on members of the United Auto Workers pay and benefit packages comparable to those paid at the nonunionized plants of foreign manufacturers that have been stealing market share from the Big Three for decades. If the Treasury were to commit government funds without getting this kind of long- overdue restructuring, it would simply be throwing good money after bad. But that's not all. Taxpayers should also demand that the Treasury take the same hard line in negotiating a rescue for the automakers that it took in structuring the rescues of Fannie Mae, Freddie Mac, AIG and Bear Stearns. Equity holders of both auto companies -- including Cerberus Capital Management, the hedge fund that purchased Chrysler from Daimler with very little of its own money -- should be wiped out, or at most given a small stake in the new company. Creditors should get only 30 or 40 cents on the dollar owed -- about what the debt is selling for now -- plus an equity stake in the new company. And top management of both companies should be shown the door, along with most of the directors, in recognition of their failure to deliver for shareholders and creditors. All of these terms -- the cost-cutting, the dealer restructuring, the haircuts for shareholders and creditors, the management changes -- can be negotiated upfront and presented as a done deal to the bankruptcy court. Such a "prepackaged" bankruptcy would allow GM-Chrysler to run through the reorganization process in a matter of a few weeks or months without missing a payroll or a day of production. It would save as many jobs as can reasonably be saved and preserve what value is left in the companies, while giving taxpayers a reasonable chance of earning a return on their investment. Honestly, I don't know if the GM-Chrysler merger makes a lot of business sense. The promised savings of $9 billion a year are overblown -- as far as I can tell, most of that represents reduction in capacity that could just as easily be accomplished by the companies separately as together. At the same time, in an industry that has gone from producing 17 million vehicles a year to 12 million, some consolidation is probably inevitable. What I do know, however, is that's it's time for Michigan's senators and congressmen and governors to stop shilling for Wall Street creditors and shareholders and defending managers and unions that stubbornly ignore market realities. For years, these politicians did the bidding of the Big Three by pushing a protectionist agenda and fighting off attempts to impose reasonable fuel-efficiency standards -- and the result is that all three companies now stand on the brink of financial collapse. This time, they owe it to their constituents and the country to back a painful but credible strategy to save the industry rather than one that simply bails out the industry from another mess of its own creation.

13 Business

October 29, 2008 Consumers Feel the Next Crisis: Credit Cards By ERIC DASH First came the mortgage crisis. Now comes the credit card crisis. After years of flooding Americans with credit card offers and sky-high credit lines, lenders are sharply curtailing both, just as an eroding economy squeezes consumers.

The pullback is affecting even creditworthy consumers and threatens an already beleaguered banking industry with another wave of heavy losses after an era in which it reaped near record gains from the business of easy credit that it helped create. Lenders wrote off an estimated $21 billion in bad credit card loans in the first half of 2008 as more borrowers defaulted on their payments. With companies laying off tens of thousands of workers, the industry stands to lose at least another $55 billion over the next year and a half, analysts say. Currently, the total losses amount to 5.5 percent of credit card debt outstanding, and could surpass the 7.9 percent level reached after the technology bubble burst in 2001. “If unemployment continues to increase, credit card net charge-offs could exceed historical norms,” Gary L. Crittenden, Citigroup’s chief financial officer, said. Faced with sobering conditions, companies that issue MasterCard, Visa and other cards are rushing to stanch the bleeding, even as options once easily tapped by borrowers to pay off credit card obligations, like home equity lines or the ability to transfer balances to a new card, dry up.

14 Big lenders — like American Express, Bank of America, Citigroup and even the retailer Target — have begun tightening standards for applicants and are culling their portfolios of the riskiest customers. Capital One, another big issuer, for example, has aggressively shut down inactive accounts and reduced customer credit lines by 4.5 percent in the second quarter from the previous period, according to regulatory filings. Lenders are shunning consumers already in debt and cutting credit limits for existing cardholders, especially those who live in areas ravaged by the housing crisis or who work in troubled industries. In some cases, lenders are even reining in credit lines after monitoring cardholders who shop at the same stores as other risky borrowers or who have mortgages from certain companies. While such changes protect lenders, some can come back to haunt consumers. The result can be a lower credit score, which forces a borrower to pay higher interest rates and makes it harder to obtain loans. A reduced line of credit can also make it harder for consumers to manage their budgets, because lenders have 30 days to notify their customers, and they often wait to do so after taking action. The depth of the financial crisis has shocked a credit-hooked nation into rethinking its habits. Many families once content to buy now and pay later are eager to trim their reliance on credit cards. The Treasury Department, which is spending billions of dollars in taxpayer money to clean up an economic mess brought on in part by all sorts of easy credit, recently started an advertising campaign inviting consumers to check into the “Bad Credit Hotel,” an online game that teaches the basics of maintaining good credit. At the same time, the fear factor among lenders has deepened just as the crisis makes it harder for some financially stretched consumers to wean themselves from credit cards for even basic needs, like gas and food. “We are not going to say, ‘Yahoo, this is over,’ and extend credit like we did without fear,” Jamie Dimon, JPMorgan Chase’s chief executive, said in a recent conference call. “If you’re not fearful, you’re crazy.” Even those with good credit ratings are not excepted. American Express, which traditionally catered to more upscale cardholders, said it would be increasing effective interest rates by 2 or 3 percentage points for some of its credit card holders — a move that could, for example, push a 15 percent rate up to 18 percent. “We think it’s prudent given the nature of those products and the economic environment we face,” Daniel Henry, its chief financial officer, said in a recent conference call. Some reward programs have also gotten stingier as lenders cut corners to save money. Card companies, for example, have taken to substituting cheaper brands for a Sony big-screen television as a way of lowering the cost of their redemption prizes. For less creditworthy customers, issuers are pulling back on promotional offers that allowed borrowers to pay no interest for months as they try to get ahead of stiffer lending rules that have been proposed by federal banking regulators and Congress. The regulations, while beneficial to consumers, will curb profits on card issuers’ riskiest customers. JPMorgan said that it was withdrawing some teaser-rate loans that were only marginally profitable. Discover Financial shortened the duration of its zero-balance offers. And lenders, over all, are slowing the flood of mail offers to a trickle with moves that would translate for the average American household into about 13 fewer pieces of credit card junk mail a year than its peak in 2005. Mail offers to new and existing customers are on pace to

15 drop below 8.4 billion pieces, the lowest level since 2004, according to Mintel Comperemedia, a direct marketing research firm. Online credit card applications have fallen for the first time in five quarters, in part because customers are receiving fewer mail offers that drive them to the Web, according to data from comScore, an Internet marketing research firm. “We used to get a couple of offers a week, but I haven’t seen a credit card offer in over a year,” said Brett Barry, who owns a real estate agency outside Phoenix and described his credit record as strong. “What blows me away is these companies are in the business of extending credit, but they don’t want to do it for me.” Mr. Barry said that, without any notice, American Express had reduced the credit limit on his business and personal credit card at least four times in the last year, which he said had lowered his credit score. The moves have also made it difficult for him to manage his payroll and budget, he said. “Credit card issuers have realized their market is shrinking and that there is no room for extra credit cards, so they have to scale back,” said Lisa Hronek, a research analyst at Mintel. “People are completely maxed out with mortgages, home equity lines and credit card debt.” At the same time, credit card profit margins have been narrowing, largely because lenders’ own financing costs remain elevated as investors spurn credit card bonds, just as they did mortgages. Another factor is that the interest rates banks charge even creditworthy borrowers have come down after the emergency actions taken by the Federal Reserve to ease the credit crisis. Meanwhile, bank executives say consumers are starting to curb their spending, to an extent that may become clearer Wednesday when Visa reports its third-quarter results. In previous downturns, banks could make up the missing profits by raising fees. This time, there may be less room to maneuver. “The last time credit costs spiked, the late fees were much lower, so card issuers could turn to that and reprice more nimbly,” a Morgan Stanley analyst, Betsy Graseck, said. “There is just more scrutiny now, and coming after the subprime mortgage crisis, the world is more sensitive to the way lenders behave.”

America must lead a rescue of emerging economies George Soros Published: October 28 2008 22:05 | Last updated: October 28 2008 22:05 The global financial system as it is currently constituted is characterised by a pernicious asymmetry. The financial authorities of the developed countries are in charge and they will do whatever it takes to prevent the system from collapsing. They are, however, less concerned with the fate of countries at the periphery. As a result, the system provides less stability and protection for those countries than for the countries at the centre. This asymmetry – which is enshrined in the veto rights the US enjoys in the International Monetary Fund, explains why the US has been able to run up an ever-increasing current account deficit over the past quarter of a century. The so-called Washington consensus imposed strict market discipline on other countries but the US was exempt from it. The emerging market crisis of 1997 devastated the periphery such as Indonesia, Brazil, Korea and Russia but left America unscathed. Subsequently, many peripheral countries followed sound macroeconomic policies, once again attracting large capital inflows, and in recent years have enjoyed fast economic growth. Then came the financial crisis, which originated in the US. Until recently peripheral countries such as Brazil remained largely unaffected; indeed, they benefited from the commodity boom. But after the bankruptcy of Lehman Brothers, the financial system suffered a temporary cardiac arrest and the authorities in the US and Europe resorted to desperate measures to resuscitate it. In effect, they resolved that no other big financial institution would be allowed to default and also they guaranteed depositors against losses. This had unintended adverse consequences for the peripheral countries and the authorities have been caught unawares. In recent days there has been a general flight for safety from the periphery back to the centre. Currencies have dropped against the dollar and the yen, some precipitously. Interest rates and credit default premiums have soared and stock markets crashed. Margin calls have proliferated and spread to stock markets in the US and Europe, raising the spectre of renewed panic. The IMF is discussing a new credit facility for countries at the periphery, in contrast to the conditional credit lines that were never used because the conditions attached to them were too onerous. The new facility would carry no conditions and no stigma for countries following sound macroeconomic policies. In addition, the IMF stands ready to extend conditional credit to countries that are less well qualified. Iceland and Ukraine have already signed and Hungary is next. The approach is right but it will be too little, too late. The maximum that could be drawn under this facility would be five times a country’s quota. In the case of Brazil, for example, this would amount to $15bn, a pittance when compared with Brazil’s own foreign currency reserves of more than $200bn. A much larger and more flexible package is needed to reassure markets. The central banks at the centre should open large swap lines with the central banks of qualifying countries at the periphery and countries with large foreign currency reserves, notably China, Japan, Abu Dhabi and Saudi Arabia, ought to put up a supplemental fund that

17 could be dispersed more flexibly. There is also an urgent need for short-term and longer-term credit to enable countries with sound fiscal positions to engage in Keynesian counter-cyclical policies. Only by stimulating domestic demand can the spectre of a world-wide depression be removed. Unfortunately the authorities are always lagging behind events; that is why the financial crisis is spinning out of control. Already it has enveloped the Gulf countries, and Saudi Arabia and Abu Dhabi may be too concerned with their own region to contribute to a global fund. It is time to start thinking about creating special drawing rights or some other form of international reserves on a large scale, but that is subject to American veto. President George W. Bush has convened a G20 summit for November 15 but there is not much point in holding such a meeting unless the US is serious about supporting a global rescue effort. The US must show the way in protecting the peripheral countries against a storm that has originated in the US, if it does not want to forfeit its claim to the leadership position. Even if Mr Bush does not share this point of view, it is to be hoped the next president will – but by then the damage will be much greater. The writer is chairman of Soros Fund Management and author of ‘The New Paradigm for Financial Markets’

18 COMMENT & ANALYSIS Add ‘financial stability’ to the Fed’s mandate By Stephen Roach Published: October 27 2008 19:46 | Last updated: October 27 2008 19:46 A regulatory backlash is under way as the US body politic comes to grips with the financial crisis. Wall Street – or what is left of it – is first in the line of fire. But the era of excess was as much about policy blunders and regulatory negligence as about mistakes by financial institutions. As Washington creates a new system, it must also redefine the role of the Federal Reserve. Specifically, the US Congress needs to alter the Fed’s policy mandate to include an explicit reference to financial stability. The addition of those two words would force the Fed not only to aim at tempering the damage from asset bubbles but also to use its regulatory authority to promote sounder risk management practices. Such reforms are critical for a post-bubble, crisis-torn US economy. This is not the first time the US Congress has needed to refine the Fed’s mandate. After the great inflation of the 1970s, the so-called Humphrey-Hawkins Act of 1978 was enacted. That required the Fed to add price stability to its original post-second world war policy target of full employment. In the late 1970s, Congress felt the Fed needed the full force of the law to tackle a corrosive inflation problem. This legislative change empowered Paul Volcker, a later Fed chairman, in his courageous assault on double-digit inflation. By focusing on financial stability, the Fed will need to adjust its tactics in two ways. Firstly, monetary policy will need to shift from the Greenspan-Bernanke reactive, post-bubble clean-up approach towards pre-emptive bubble avoidance. Second, the bank will need to be tougher in its neglected regulatory oversight capacity. By adding “financial stability” to the Fed’s policy mandate, I am mindful of the pitfalls of multiple policy targets. However, single-dimensional policy targeting does not cut it in a complex world. As such, the Fed will need to be creative in achieving its mandated goals – using monetary policy, regulatory oversight and enforcement and moral persuasion. Just as the Fed has been reasonably successful in its twin quests for price stability and full employment, I am confident it can rise to the occasion with the addition of financial stability to its mandate. I am not suggesting the Fed develops numerical targets for asset markets. It should have discretion as to how it interprets the new mandate. Yes, it is tricky to judge when an asset class is in danger of forming a bubble. But hindsight offers little doubt of the bubbles that developed over the past decade – equities,

19 residential property, credit and other risky assets. The Fed wrongly dismissed these developments, harbouring the illusion it could clean up any mess later. Today’s problems are a repudiation of that approach. There is no room in a new financial stability mandate for bubble denialists such as Alan Greenspan, the former Fed chairman. He argued that equities were surging because of a new economy; that housing forms local not national bubbles and that the credit explosion was a by-product of the American genius of financial innovation. In retrospect, while there was a kernel of truth to all of those observations, they should not have been decisive in shaping Fed policy. Under a financial stability mandate, the Fed will need to replace its ideological convictions with common sense. When investors buy assets in anticipation of future price increases the Fed will need to err on the side of caution and presume that a bubble is forming that could threaten financial stability. The new mandate would also encourage the Fed to deal with excesses by striking the right balance between deploying its policy interest rate and other tools. In times of asset-market froth, I favour the “leaning against the wind” approach with regard to interest rates – pushing the Federal funds rate higher than a narrow inflation target might suggest. But there are other Fed tools that can be directed at financial excesses – margin requirements for equity lending as well as controls on the issuance of exotic mortgage instruments (zero-interest rate products come to mind). In addition, the Fed should not be bashful in using the bully pulpit of moral persuasion to warn against the impending dangers of asset bubbles. Of equal importance is the need for the Fed to develop a clearer understanding of the linkage between financial stability and the open-ended explosion of derivatives and structured products. Over the past decade, an ideologically-driven Fed failed to make the distinction between financial engineering and innovation. It understood neither the products nor their scale, even as the notional value of global derivatives hit $516,000bn in mid-2007, the eve of the subprime crisis – up 2.3 times over the preceding three years to a level that was 10 times the size of world gross domestic product. The view in US central banking circles was that an innovations-based explosion of new financial instruments was a huge plus for market efficiency. Driven by its ideological convictions, the Fed flew blind on the derivatives front. On the one hand, this was hardly surprising as these are largely private, over-the-counter transactions. What is surprising is that the authorities failed to develop metrics that would have helped them understand the breadth, depth and complexity of the derivatives explosion. This trust in ideology over objective metrics was a fatal mistake. Like all crises, this one is a wake-up call. The Fed made policy blunders of historic proportions that must be avoided in the future. Adding financial stability to its mandate is vital to preventing such errors again. The writer is chairman of Morgan Stanley, Asia

October 28, 2008 Plunging markets are far from irrational There is a simple explanation for falls in share prices - investors fear a depression is coming Oliver Kamm In the month to date, stock markets in the US and Europe have declined by more than a quarter. In Asia they have fallen by slightly more. And with these declines has come a surge in stock market volatility. Why is it happening? Does it matter to anyone except the traders shown slumped or shouting on television? And is this any rational way to run an economy? Perhaps surprisingly, the answers are fairly straightforward. And the main lesson to be drawn is not that capital markets are inherently dangerous and destabilising. It is rather that they can help to protect people from future risks that were disastrously overlooked in the long build-up to today's financial crisis. Stock markets have been collapsing for three main reasons. First, in more stable economic times, investors often employed a practice known as the “carry trade”. This meant borrowing in a currency with a very low interest rate - typically the yen - and investing the money in higher-yielding markets (Australia and New Zealand are good examples). As interest rates have been slashed throughout the developed world, this trade has become much less attractive. Secondly, the crisis has exacted a heavy toll on hedge funds. These are like upmarket mutual funds for very wealthy private investors. They differ from, say, pension funds or unit trusts in two main ways. They can borrow to invest (known as leverage); and they can sell stocks that they do not own, in the hope that the price will fall and they can buy the stock back at a profit (known as short-selling). This means that hedge funds are often better placed to take an opposite position to conventional investment fashion than funds with a more conventional mandate. Yet with neither justice nor logic, hedge funds have suddenly assumed the role of villain, according to tabloid mythology of the credit crisis. More seriously, regulators have curtailed their capacity to sell short. This is misguided. It is contributing to a flood of redemptions from hedge funds - which requires the funds to sell their assets. Thirdly, and most important, investors are considering the economic outlook and concluding that there are immense risks to their future wealth. The value of any financial asset depends on how much cash it will generate for an investor in the future. In the case of a bond, the cash flows are the stream of future interest payments. In the case of a stock, they are the stream of future dividends paid out of corporate earnings. To work out what those cash flows are worth now, you have to apply a discount rate to your expectations for the future payments. A dividend payable in 2010 is less valuable to an investor now than a dividend paid in 2008, because it is less certain - so you have to apply a higher discount rate it to work out its value today. Quite suddenly, the risk to the world economic outlook has ratcheted up. Investors see bad data and the collapse of leading banks, and worry that a cyclical economic downturn

21 might become a depression. The risk of a seriously nasty outcome means that companies' future earnings are less certain. So sharp falls in stock prices, and the volatility of stock markets, are far from irrational. They are a signal - not a cause - of the underlying stresses in the world economy. How far does this matter to the ordinary consumer and saver? It matters because most of us have money tied up in pension funds, which are invested, to a greater or lesser extent, in the stock market. The only sensible course to take is to regard investing as a long-term discipline. Investment returns become less volatile on average - and hence less risky - over longer time horizons. And yes, financial markets are a valuable feature of a modern economy. They put companies who need money, and can use it productively, in touch with people who have money, and hope to boost their financial returns by investing it. Is this not a trite conclusion, given all that has happened? No, it is not. The ructions that have engulfed the Western financial system come down to many things - but they started in the US housing bubble. The housing market in the UK has also frozen up. Houses are a huge investment for most of us - yet are terribly difficult to trade. Robert Shiller, a Yale economist, has recently proposed a remedy in an outstanding short primer, The Subprime Meltdown. He argues that creating derivatives markets in house prices would help to tame the cycle of boom and bust. Sceptical investors could express that view in more immediate ways than selling their own homes. Housebuilders would note the expected price declines signalled by the market, and scale back their building activities. Shiller's idea is plausible. It illustrates how financial markets need not be the casino or bearpit of popular criticism. Investment is above all about efficiently managing risk. Investors can spread their risk by diversifying their portfolios across stocks and asset classes. Financial markets also enable businesses to hedge against risks. These are valuable disciplines, and they can help us in future. As Shiller envisages, they might even correct the most damaging distortion in our economy: the fascination with bricks and mortar, to the detriment of making things and providing services. Oliver Kamm is a Times leader writer and former investment banker

22 Tuesday, October 28, 2008

Greenspan vs. Buffett October 27, 2008 02:17 PM ET | Rick Newman| If Alan Greenspan lived on a flood plain, would he buy insurance? When the former Federal Reserve chairman testified before Congress recently, he kicked off his remarks by announcing that "we are in the midst of a once-in-a-century credit tsunami." Those once-in-a-century analogies are usually used to explain away something that's so rare you can't possibly be blamed if you fail to prepare for it. Plan for every hundred-year disaster and there's little time or money left to invest in the good life. Greenspan's shrug-off brought to mind Warren Buffett, who's made news lately by snapping up big chunks of Goldman Sachs and General Electric at depressed prices and by generally being the only living person expressing any optimism at all about stock markets. There's also a new book about Buffett, The Snowball by Alice Schroeder, who spent hundreds of hours talking with the Oracle of Omaha. Buffett is famous for his ability to calculate risk—and avoid it—which is the very thing that banks and consumers catastrophically failed to do over the past couple of years. "He always thinks through what's the worst possible thing that could happen," Schroeder told me during an interview. "What we're seeing now is a lot of people who said, 'This kind of calamity has never happened before, so it probably won't happen to me.' But that doesn't mean the calamity will never happen." Contrast that with Greenspan's Panglossian perspective. By way of explaining the housing boom and subprime lending explosion he presided over—which caused the housing bust and recession we're enduring now—Greenspan described the "best insights of mathematicians and finance experts" whose job was to make sure that that credit tsunami didn't happen. But instead, he testified, "the whole intellectual edifice...collapsed in the summer of last year because the data inputted into the risk management models covered only the past two decades, a period of euphoria." Bad data. Bad outcome. Tsunami. The solution, Greenspan pointed out, would have been mathematical models "fitted more appropriately to historic periods of stress." In other words, the long view. But there was at least one finance expert whose risk-management models predicted a chance of catastrophe. In his Berkshire Hathaway shareholder letter from 2002, Buffett wrote that derivatives—such as credit-default swaps, now causing tremors in world markets—were "toxic" investments, "time bombs" that could wreck the financial system. In his 2003 shareholder letter, Buffett famously called derivatives "financial weapons of mass destruction." At around the same time, Greenspan was lowering interest rates to historically low levels, fueling a lending binge, a housing bubble, and the mass securitization of mortgages good and bad. He was also extolling the virtues of derivatives, arguing that the benefits outweighed the costs and that more regulation was unnecessary.

23 The Snowball also details how Greenspan and Buffett both played a role in cataclysms that might not have been once-a-century events but were certainly dramatic contretemps during Greenspan's "period of euphoria." One of them was the 1991 bond-trading scandal at Salomon Brothers, which mushroomed into a kind of bank run that nearly sank one of Wall Street's mightiest investment banks. Buffett, a big Solomon shareholder, saw the risks of collateral damage up close and engineered a rescue effort that helped prevent turmoil at other banks. Greenspan, as Fed chairman, also had a front-row seat, and called Buffett at one point to offer his encouragement. Then, in 1998, the huge hedge fund Long-Term Capital Management nearly collapsed, threatening another global run on the financial system. The fund's proprietors lobbied Buffett for a big capital infusion, but he demurred. Instead, it was Greenspan who engineered an unprecedented bailout of the private firm, lest its problems infect dozens of other institutions. Buffett has since preached about the lessons of those incidents, while more euphoric investors apparently forgot them. Buffett's message of prudence was unwelcome during the housing boom, when he seemed like a financial fuddy-duddy. Newsweek even called him "the alarmist of Omaha." But after the first domino fell in the current crisis—when the government bailed out a rump Bear Stearns earlier this year—Buffett put the pieces together in an interview with Schroeder. "It's a version of what I went through at Salomon," he said, "where you were just inches away all the time from, in effect, an electronic run on the bank." Greenspan was there too, but he seems to have drawn different lessons. If it wasn't a hundred- year calamity, it didn't seem to count. Who needs insurance when you have exuberance?

“How far back" are we? Barry Ritholtz Posted by Barry Ritholtz on Tuesday, October 28, 2008 | 12:30 PM

This chart is from Dr. Artur Adib, a physicist interested in finance. He wanted to attempt to see how bad things are in the stock market, so he generated the chart below. It compares current pricews (top) with the Dow price in terms of the number of years since the first occurrence of that price (bottom, red) I liked the way this puts the current historical price chart into a somewhat different perspective.

25 Consumer Confidence Plunges to Record Low Posted by Barry Ritholtz on Tuesday, October 28, 2008 | 10:45 AM

The Conference Board reported Consumer Confidence Index for October hit a record low of 38. Is there any connection between weak consumer confidence readings and bullish reversals in the stock market? To decide, have a look at the following quick-and-dirty overview (which includes all data going back to the Feb-67 start of the confidence series):

(*Note: Interim low - market was already in a secular upswing.) (#Note: Consumer confidence readings that were closest to the latest reading.)

Table via Mike Panzner From the Conference Board "The Conference Board Consumer Confidence Index™, which had improved moderately in September, fell to an all-time low in October. The Index now stands at 38.0 (1985=100), down from 61.4 in September. The Present Situation Index decreased to 41.9 from 61.1 last month.

26 The Expectations Index declined to 35.5 from 61.5 in September. The Consumer Confidence Survey is based on a representative sample of 5,000 U.S. households. Says Lynn Franco, Director of The Conference Board Consumer Research Center: "The impact of the financial crisis over the last several weeks has clearly taken a toll on consumers' confidence. The decline in the Index (-23.4 points) is the third largest in the history of the series, and the lowest reading on record. In assessing current conditions, consumers rated the labor market and business conditions much less favorably, suggesting that the fourth quarter is off to a weaker start than the third quarter. Consumers' appraisal of current conditions deteriorated sharply in October. Those saying business conditions are "bad" increased to 38.3 percent from 33.4 percent, while those claiming business conditions are "good" declined to 9.2 percent from 12.8 percent. Consumers' assessment of the labor market was also much more negative. The percentage of consumers saying jobs are "hard to get" rose to 37.2 percent from 32.2 percent in September, while those claiming jobs are "plentiful" decreased to 8.9 percent from 12.6 percent." FYI: This sort of sentiment always favors the challenger, not the incumbent party, in the White House race.

Loss Souls A Day of Reckoning Dawns for Denizens of Hedge-Fund Heaven By David Segal Washington Post Staff Writer Tuesday, October 28, 2008; C01 GREENWICH, Conn. It isn't easy to get the super-rich to discuss their money woes, and the Rev. Chuck Davis, who runs the Stanwich Church here in the hedge-fund capital of the world, has tried. His flock includes the mandarins of finance who've lost fortunes in the stock market debacle. In a recent sermon, he urged parishioners to simply admit that they're enduring a terrible ordeal. "C'mon, let's us talk about it, right now," he said from the pulpit. "There's fear. I've never seen this kind of fear in people. There's concern. Our world has been rocked in some ways. I think we've come to realize that we've lived an illusion for a little while, haven't we?" This isn't an "Amen!" kind of place, but listeners were rapt. "And we've been shocked," he continued, "even though it was an illusion and it wasn't a reality, coming back to reality, we still want to know that God's in the midst of it. Because darkness would seem to drown out our hope and sense of well-being." Uplifting words, and delivered to a group that you have to assume could use a hug. But if there is darkness here in Greenwich, it's not visible, at least not yet. The impeccably kept home to scores of suddenly unemployed and just barely employed investment bankers, Greenwich is the most famous of the money towns of Fairfield County -- along with Darien, New Canaan and others -- where some 28,000 millionaires resided the last time anyone counted. The area has just suffered its version of a Category 5 hurricane. But this disaster has unfolded If there was a FEMA for portfolio devastation, Greenwich would be swarming with feds.on the QT. To find the stress points and hints of agony, you have to look around. A total of five homes have gone to foreclosure of late, though just two were worth more than $4 million -- mid-priced for this area. The town voted to eliminate 15 jobs last week, the first government layoffs anyone can recall. Just one family at the all-girls Greenwich Academy, where upper-school tuition costs $31,000 a year, has asked for financial aid. "I hasten to add that, as with all independent schools, financial aid is need-based," says Molly King, head of the school. "There is paperwork to fill out and a vetting process." At L'Escale, one of the pricier local restaurants, some specials were just announced. "The first drink is on us!" read a news release. Guests can choose from "sublimely balanced" cocktails said to be inspired by the seven virtues, one of which, ironically enough, is prudence. "We're also doing a prix fixe menu, which we've never done before," Anshu Vidyarthi, L'Escale's manager, says on the phone. "I think the mood here is, 'We've had a great party, now let's make sure we can continue the party, so let's be smart about what we're doing.' " It might well be too early in Meltdown '08 for the residents of Greenwich to betray outward symptoms of financial distress. For some, this could be a matter of pride. Others who've lost money still have a ton left; they could be reeling and living large at the same time.

28 So we decided to poll the religious leaders of this town and spend a little time in the pews. Maybe the losses haven't yet caused a lot of overt damage. But how about interior struggles? Are people in Greenwich searching their souls? How about rethinking their priorities? Is anyone here -- and this might be too much to ask -- talking about regrets? Not many, it turns out. The Rev. James Lemler of Christ Church says he has counseled a few members who are looking for a spiritual life after years of focusing almost exclusively on the material. "It's as though the market has let them down," he says, "and now they would like something a little firmer to believe in." At Temple Sholom, a mere parking lot away, attendance is up a bit, and there has been some talk, from former Wall Street mega-millionaires, about a search for deeper meaning. But Rabbi Mitch Hurvitz says he hasn't exactly witnessed a Great Awakening. "I have seen a few of what I call recovering secularists," he says. "People who say, 'Wait, there must be something more out there,' and that's a good thing. But we're not talking about a deluge of people." Actually, it's fair to say that here in the suburban symbol for all that has gone haywire in our financial system, it's pretty hard to find signs of contrition. That's right. If you were holding your breath for an apology from someone in the private sector -- "Dear Nation, Our bad. Love, the Suits" -- you should exhale. If Greenwich is any indication, the finance titans have little on their minds except winning back their losses. "The time I hear soul-searching here is when a guy's wife leaves him and he says, 'I should have spent more time with her,' " says Laurence Lorefice, a psychiatrist in Greenwich. "Which suggests these people are capable of soul-searching. But this is about what the stock market is doing to their careers. So what I hear now is, 'I should have shorted the market, I should have gotten out sooner, I should have seen this coming.' Not one person has said, 'I rue the day that I got into this because it's empty, meaningless and vacuous.' Nobody here is talking about guilt." * * * Can we just say it? If the upper tax brackets of Greenwich were writhing in private, nobody would weep. We're not talking about nuclear-bomb-scale writhing, obviously. But a few nervous breakdowns, some personal bankruptcies, perhaps a few moving vans -- who would mind? Yes, mortgage-backed securities, stocks, bonds and every hybrid of financial instrument are traded all over the world. But nowhere else has the cachet-turned-baggage of Greenwich, long a home to blue-blood families (this is where the Bush family dwelled for years) and more recently a haven for Wall Street's new money, such as hedge-fund katrillionaire Steve Cohen. The place looks the part. There are two car dealerships here that sell Rolls-Royces, and there is a strip of high-end retail, along Greenwich Avenue, that is New England's answer to Rodeo Drive. Among the stores are Saks Fifth Avenue, Theory and Christopher Fischer. But for all that, Greenwich is a realm of stinking riches where everyone is pretending to ignore the smell. Ostentation is frowned upon. Last week, a golden-toned Mercedes-Benz with the license plate MADXS was seen darting around town, but it was striking for how out-of-place it looked. The goal here is to live well without ever seeming extravagant -- especially now. "A lot of parties have been scaled back, or canceled," says Morgan Mitchell, a former managing director of J.P. Morgan, "because it's embarrassing to throw a big party when people are hurting."

29 Mitchell is a co-founder of one of the rare gathering groups that needs more chairs when it convenes: the Greenwich Leadership Forum, a group of executives who meet once a month to discuss how Christian values can mesh with the values of the executive suite. The meetings are held at 7 a.m. in a banquet room at the Indian Harbor Yacht Club and open with a little ceremony: A forum leader takes a copy of the Wall Street Journal and shoves it between the pages of the New Testament. "It's symbolic," says Richard Murphy, the other founder and a management consultant by trade. "It's the two bibles, coming together, the bible of business and the Bible. We know the former well, but we don't always know enough about the latter. And we want to show that these two bibles can coexist." About 90 people turned up for the last meeting to hear John Murphy of OppenheimerFunds talk about how his training in the Jesuit tradition has guided his career. That's about 30 more people than the usual crowd that has turned up since the forum started five years ago, though Mitchell isn't sure if that's because of the quality of the headliner or because the credit crisis spurred some deep thoughts. He's certain, however, that at the heart of our current economic problems is a pretty basic erosion of ethics. "Too many people felt comfortable pushing the envelope," he says. "Because frankly, it's exciting. I know from being on a trading desk. You make a billion-dollar trade, it's exciting. You help a billion-dollar company get bought or merged, it's exciting. And you can get caught up in that. But at some point in this country, we started taking risks that aren't ethical. I think it's because the penalties for getting caught weren't all that great, and the golden rule got lost." Mitchell says he had a discussion recently with an investment banker who lamented that he'd lost three years of wealth in the stock swoon, time he could have spent with his children. But that guy is a forum regular. Ask for an example of a master of the universe whose worldview was upended in the past month, and Mitchell is stumped. So is Murphy. "Right now, we're on the periphery," he says, "and I don't know if, in my lifetime, we'll see a huge impact on this great force out there called money." * * * For a look at what they're up against, we shopped for real estate. Specifically, we stopped at an open house for a newly built, four-bedroom, five-bathroom home in a section of Greenwich called "downtown," although there isn't much downtown about it. The exterior of this building, which seems a bit too large for its plot of land, looks like a Mediterranean villa reconfigured for an industrial office park. Actually, it's two 6,000- square-foot "townhomes," attached to each other like twin whales. Inside each, there are heated floors, tons of marble, an elevator, a wine cellar, a sauna and lots of extras, including a $21,000 TV projector in a room in the basement. "The walls in this room cost $25,000," says Anthony Pavia, who installed the sound system and a total of six TV sets in the house. He was putting final touches on the wiring as he described the impact of the looming recession on his business. It's down if you look only at new construction in Greenwich, he says, but he's made up any slack through an increase in the number of security systems he's been hired to put in residences here and elsewhere. "I heard this story on the radio about how sales of safes are going up because rich people don't trust the banks, and maybe that explains it," he says. "But whatever it is, I'm installing a lot of alarms."

30 As he talked, a handful of prospective buyers walked through the house, pitched by a very thin, very blond real estate agent. You had to wonder who would buy this place, which frankly now seems like a bit of an anachronism, something conceived in the era of Dow 14,000 and born 6,000 points too late. The price tag for each townhome is $5.9 million. "The potential buyers will be people in McMansions in Mid-Country or Back-Country" -- two of the more coveted sections of town -- "who are being forced into a sale, or who are being smart and want to consolidate and have a little cash in reserve," says Bryan Dinkelacker, who owns the local Engel & Volkers real estate brokerage, which is selling the house. Wait, this is for people who are downsizing? "Oh yeah," Dinkelacker says. "These will be people saying, 'Instead of a $20 million, 20,000- square-foot home, I can move into a smaller house and maintain my quality of life.' They can have the same life, but in a smaller package." * * * Both townhomes have languished on the market since mid-July. But if the boom years are truly over for the Greenwich elite, they won't suffer alone. The town, which includes a surprisingly large number of blue-collar neighborhoods, produces gobs of money for the state budget of Connecticut -- about $600 million in personal income taxes in 2006, roughly four times the combined totals from Norwich, Hartford, Bridgeport, New Haven and Waterbury, according to the state's Office of Policy and Management. Already, there is talk of shortfalls. Politicians in the capital are bracing for the state's most reliable jackpot machine to shut down. That won't wipe the smirk off anyone's face in the rest of the country, but the inevitable tax hikes and cutbacks will surely dial back the schadenfreude in the Nutmeg State. Maybe a real reckoning is coming. Some in the church- and temple-going community think so, and they believe it could be exactly the figurative knock upside the head that the money- obsessed here have long needed. That includes Davis of Stanwich Church, who spent eight years in Mali, one of the poorest countries in Africa, before moving to Greenwich. "We've had amazing wealth here, obnoxious wealth," he says, "and yet so many people in Greenwich have a poverty mentality. It's always, 'Do I have enough? I need more, I need more.' " Davis has been heartened by the reaction of regulars who heard his sermons on the town's economic travails, but he sounds a little surprised by the small number of new faces he's seen on Sundays. He thinks more will come as people really start to suffer. Perhaps, but there are those who think it's pretty horrible in Greenwich already. On a recent Sunday, one Christ Church congregant -- an investment banker who asked not to be identified -- likened the quiet of Greenwich to Hiroshima right after it had been turned into radioactive rubble. "The people in Hiroshima were so shocked that at first, there wasn't any noise," he says after services. "That's what's happening in Greenwich. Everyone is in a state of total shock right after the bomb."

October 28, 2008 Distrust Toppled Investment Banking Industry By Natalia Redenbaugh The widespread economic and market panic has trickled down through the economy in such a way that even our housekeeper is affected by it. Like many, she is gripped by fear and a lack of understanding. To help her understand what's transpired in an easy-to-understand way, I had to explain why highly competent investors (handicapped by their own knowledge) failed to protect themselves and their clients. The cause was simple: Widespread, pervasive DISTRUST. As a business anthropologist in the investment community (partner with Kairos Capital Advisors) I am always watching people, particularly those networks of people that make up businesses and the market. To understand more about what happened in the market, I first explained how people act with trust or distrust. Have you ever had someone lie to you? Maybe not even lie, but a stretching of the truth? Have you noticed how the lack of trust spreads? They may have stretched the truth about their competence in one area, but knowing they exaggerated about one thing leads you to distrust them on unrelated issues. More obvious is when someone steals; say a housekeeper (not my housekeeper) pocketed a couple of dollars she found while doing laundry. You now begin to suspect that she has been stealing from you all along and this two-dollar incident leads you to believe she may have stolen hundreds of dollars. Perhaps you misplaced your earrings but you are now sure that the housekeeper must have stolen them. It takes a long time to build trust. Unfortunately it can be destroyed in an instant. Now what if your housekeeper were to deny the incident, claim she never took the two dollars and you must have misplaced it? Distrust grows and there is no hope of rebuilding it. The first step to rebuilding trust is to accept responsibility and offer to take action over time to rebuild what's been eviscerated. If you choose to salvage the relationship, you will give the person a chance to earn your trust back, though it takes time and repeated, observable actions to win what was lost back. Distrust is an insidious mood; it can spread beyond the person who broke it in the first place. Perhaps you never hired a housekeeper before. This was your first experience. Would you not consider that perhaps ALL housekeepers steal and therefore you won’t hire another? At the very least you may step back, reassess, perhaps talk with your neighbors to be sure this is not a widespread practice by housekeepers in general before you hire another. Let’s now take it one step further. Your housekeeper hasn’t actually stolen two dollars. You did indeed misplace it. However, you have been bombarded by friends’ tales of household help stealing from them left and right. You could possibly begin to wonder whether or not

32 your housekeeper, like those of your friends, may be stealing and you just haven’t caught her yet. With this background of distrust you begin to watch your housekeeper with new eyes. Now the two dollars disappear and you think “aha”, she is just like the others and jump to the conclusion that she is a thief even though she is not. This distrust may not be limited to your housekeeper. Now you begin to watch the gardener, the contractor and the pool man, anyone who could have the access and opportunity to steal from you. Distrust is dangerous because like a metastasized cancer, it spreads everywhere quickly. It destroys relationships, sometimes unnecessarily. It is this spread of distrust, much of it caused by rumors that experienced investors knew to be false, which caused many to short or sell stocks. Throughout the history of the stock market, rumors always arise, though an experienced investor knows that false rumors usually are revealed to be false and normalcy is promptly restored. In fact these rumors (or rather invalid assessments) created buying opportunities for investors who understood that all would be righted in a short time; the drop in stock value temporary. But before I could show Marcia, our housekeeper, how distrust toppled the investment banking industry, I needed to first show how businesses built trust and value. I started out explaining in an uncomplicated way how the market generally works. As our readers already know (but wasn’t as clear to my Brazilian housekeeper), public companies usually report earnings each quarter and set new targets for the next quarter at the same time. People buy or sell stocks in these businesses based on whether or not they trust that these companies will reach the targets, and whether or not those targets will improve the valuation enough to warrant a higher stock price. Companies that achieve their targets over time build more trust. It takes many quarters and years to build a company’s reputation. Now there are many “speed bumps” that can hinder a company’s race to their quarterly goal. These “speed bumps” can also make it more likely that some will reach their goal while others may not. Importantly, there are external factors that could hinder a company’s growth. Rising gas prices would affect the firms, such as airlines and transportation companies, whose operating costs would rise. Trade regulations would affect those companies who rely on open markets to access customers in other parts of the world. Tax increases could make the cost of doing business increase and reduce profitability. Healthcare requirements could increase the cost of doing business and again affect profitability. Many of these “speed bumps” are the result of policy changes, but sometimes they materialize due to internal mistakes. A failed product, such as an unsafe or unreliable air bag in a car, could lead to a recall jeopardizing future sales. Management could change, which would call into question the competence of the new team's ability to achieve a target. Suffice it to say, investors are always assessing events of the internal and external variety in hopes of assessing their impact on company operations, not to mention what it will mean for the firm's future share price. These “speed bumps” and a company’s failure to achieve a target generally result in a lack of confidence. But a lack of confidence is different than distrust. It is generally contained to a specific set of circumstance, and doesn’t affect other domains. The path toward rebuilding confidence is usually clear. What's happened in these uncertain markets of late has been the result of a growing lack of trust. This is not a “crisis of confidence”, but something much more powerful.

33 Distrust is the story here, and it's been a paucity of trust so great that it obliterated the century long business of investment banking. How did this come about? Before elaborating it's useful to remind reader that policy emanating from Washington drives individual behavior. It makes some assets more valuable than others. And in 2007 there arose two policies that combined with mistrust to unravel the market. All of us remember the Enron and WorldCom scandals. Those were examples of earnings opacity that led to policy changes meant to rebuild confidence in firm balances sheets. First the SEC made the management of those companies responsible for their actions. Secondly, regulators sought to enhance reporting rules in hopes of preventing the Enron/Worldcom mistakes from happening again. Unfortunately, when politicians and regulators seek to fix yesterday's problems, they often overreact. In short, there were many regulations put into affect, and not all of them were good. Notably, the Financial Accounting Standards Board (FASB) Statement No. 157 became effective after November 15, 2007. What is commonly referred to as mark-to-market accounting changed how assets on balance sheets were valued. This required that public companies adjust their asset values based on the market value, as opposed to the book value of assets. In the past, performing loans (an asset) would be valued at book value. If a compnay had a $300,000 mortgage, which had been in good standing for 10 years, it could be booked at its accrued value. FASB 157 required that its value be adjusted to reflect the current market value of the mortgage. In a time of distrust, when investors were selling these mortgages at fire sale prices, FASB 157 required that banks reduce the value of similar assets, even if performing, to the amount they were sold at by the latest distressed seller. As more mortgages were sold at distressed prices, driving their value down with each desperate sale, banks had to repeatedly reduce the value of those assets on their balance sheets even if they planned to hold them for the life of the loan. Compounding the banks’ problems were requirements to maintain reserves to offset these assets. As their assets fell in value, their capital requirements rose. This is how Fannie Mae and Freddie Mac got into trouble. The second policy change that was the nail in the coffin for the investment banks, banks in general and AIG was the SEC's alteration of short selling rules on the books since 1938. This allowed investors to sell securities they didn’t own even if the share price was falling. This made it much easier to "short" shares, thereby opening the market to short selling on a massive scale. The above policies, fueled by massive distrust, led to an avalanche of falling valuations. It all happened so quickly that even the most experienced investors were caught in disbelief. In simple terms, here is how it happened. Many want to blame a few investors who allegedly conspired to topple the investment industry. Everyone looks to parcel out blame in an environment of distrust. It may turn out that a few violated the law with regard to short selling, but it's more realistic to assume that a lack of trust sparked a forest fire that couldn’t be contained. And thanks to mark-to-market accounting, banks were forced to continuously write down their assets and raise more capital. After Paulson perhaps foolishly let Lehman fail, distrust rose and banks began to experience a run on their capital at the very time they needed to raise reserves. Congress didn’t contain the damage until the $700B bailout was passed, and the FDIC raised the government’s guarantee on deposits to $250,000.

34 Meanwhile, people began to close their money market accounts and liquidate their portfolios. Unlike commercial banks, investment banks were less secure because they didn’t have the safety net of government guarantees. Short sellers, again fueled by distrust, began shorting bank stocks. Rumors fomented the latter, the kind of rumors that experienced investors knew to be exaggerated and perhaps untrue. Many of them bet against the gossip since this approach worked in the past. Unfortunately, it is much harder to regain trust than it is to destroy it, so there was no time to invalidate the rumors before people liquidated their accounts and shorted the banks. To understand this, a non-banking scenario is perhaps useful. Suppose you had a group of friends. Everyone begins gossiping about one of your girlfriends. They say she is nearly bankrupt, her husband is about to lose her job, and they’ve been living beyond their means. Would you trust giving her a loan? You might give her a gift if you were able to, but it's unlikely she would rate a loan. Even if the rumor turns out to be untrue, it's unlikely that she'll merit a loan until it is soberly observed over time that the rumors are false and she remains solvent. The above is how distrust works. This is what happened with the banking industry in general, but the investment banks more so because they lacked the government guarantees when it came to solvency. Policymakers perhaps should have foreseen the consequences of the 2007 policy alterations, but there was little time. Alan Greenspan testified that he didn’t see it coming. Even Ace Greenberg, the wise investor who headed up Bear Stearns for as long as I’ve been alive, couldn’t see it and certainly couldn’t stop it. Personal phone calls by him to other investors couldn’t reverse the rumors. This man had more experience and integrity navigating markets than nearly anyone, but even he didn’t have the clout to set the rumors straight in time to save Bear Stearns. The pervasive mood of distrust, and the policy changes that allowed people to act on it, destroyed trillions of dollars of wealth in just weeks. Experienced investors, ones that for years were able to make deals based on a handshake and their honorable reputations, were engulfed in the losses because all that they knew and had experienced to be true about rumors vs. integrity led them to bet against rumormongering that was terrifying the markets. Now the investment banks no longer exist. After 100 years they are gone, possibly never to return. This business model failed. Even the giants like Goldman Sachs could not combat the widespread impact of eroding trust. Sadly, it will take years to rebuild trust lost, and which led to the disappearance of some venerable institutions. But like all problems economic, this too shall pass, and when it does one can hope for the ushering in of new policies that enhance, rather than inhibit, the essential process whereby public companies develop trust with those possessing capital. Natalia Redenbaugh is a partner at Kairos Capital Advisors.

DAVID WEIDNER'S WRITING ON THE WALL Regulate hedge funds now Commentary: The looming threat is too great to let funds run free Last update: 12:01 a.m. EDT Oct. 28, 2008 NEW YORK (MarketWatch) -- In an about-face, the very private Kenneth Griffin made a very public call last week to investors, the media and anyone who wanted to listen in about his darling hedge fund company, Citadel Investment Group. This is so ironic that it's almost unbelievable. Griffin was forced to a do a conference call about the financial stability of Citadel after rumors suggested Citadel was in trouble. Griffin said the firm was well capitalized and was working to reduce risk on its balance sheet. Yes, he said, the fund is down 35%, but they couldn't see it coming. The problem is that Citadel's formula, and the trading strategies of all the other funds out there, only work if there are certain givens in the market. Maybe it's that money-market funds won't break the dollar or the Lehman Aggregate Bond Index won't go below 90, or Lehman Brothers itself won't disappear -- things that almost never happen, until this year. "I've never seen a market as full of panic as we've seen in the last seven or eight weeks," Griffin said, appealing for calm, and really trying to convince lenders and investors alike to keep their money in Citadel. Coming from a 40-year-old wunderkind of the hedge fund world, you have to wonder what kind of market experience he really has. Sure, he traded from his college dorm room back in 1987, but did that give him the understanding to handle this environment, a climate that has baffled experts around the world with its unpredictability?

The answer, of course, is that it doesn't matter. The $2 trillion hedge fund industry has become so big, so much a part of the financial apparatus, that it must be regulated now. The funds are too leveraged, too intertwined, and the markets they play in are so dangerous that it's a wonder that we haven't felt the full brunt of a hedge fund chain-reaction so far. Or maybe we have. The industry has lost $210 billion during the last three months, according to hedge fund research. Overall returns are down more than 10%. And many believe that hedge funds were directly responsible for the rumors and selling that drove major financial institutions such as Bear Stearns Cos. and Lehman Brothers Holdings Inc. out of business. Rumors of insolvency also swirled around Fannie Mae and Freddie Mac and banks such as Morgan Stanley and Wachovia Corp. Those failures and the deep losses at other financial firms have cleaved 40% of the stock market's value and forced government seizure of the banking system, and it may have been triggered by the same kind of game Griffin's call suggests is being played against Citadel now: the nasty rumor.

36 No oversight The thing about rumors is that they only hurt if one of two situations exists: they are true, or are not obviously untrue. Hedge funds are at risk even more than traditional financial firms because they are completely opaque. Hedge funds don't report trading positions, or holdings. They don't say how much they owe or who they owe it to. We don't know how many out there are about to collapse or are collapsing. We don't even know for sure how many are out there. Hedge funds don't have to register with the Securities and Exchange Commission after Bulldog Investors' Phillip Goldstein challenged the SEC's oversight. See full story. This lack of information has led to some panic investigation by the Federal Reserve in recent days. The Fed is trying to find out just how much is owed and to whom at Citadel and another fund, Sankaty Advisers LLC, according to reports. See full story.

The stakes are high should the government continue to keep the hands- off approach favored not only by industry insiders such as Goldstein, but former regulators such as former Fed Chairman Alan Greenspan. Citadel has diversified into financial services. It acts as a market maker to other industry players. Citadel handles 30% of all option trades and 8% of all American stock trades. Some say those businesses would be fine if the fund was forced to close, but who can be certain? Reformers should also remember it's not just the hedge funds, but the markets they play in. For instance, the $50 trillion credit derivatives market is over-the-counter, which means there's no clearing of trades, and that means "it is difficult to know how much insurance exists on each borrower, or to know who has insured whom, and for how much," according to an article by Stanford University professor Darrell Duffie. To shed light on the market the Federal Reserve is pressing the market players for a clearing agency or exchange. See report on credit derivatives. Suing banks The question facing Wall Street and Washington is whether regulation will come fast enough to stave off disaster, or will a big fund like Citadel have to collapse before anyone takes notice? Already there is some pushback from the industry that underscores how firms have abandoned ethics in their pursuit of returns. Greenwich Financial and Braddock Financial reportedly have threatened to sue banks for renegotiating mortgages because they have bet against those loans getting repaid. All of this is further sullying an already tarnished reputation for the hedgies. Even on Wall Street, hedge funds by virtue of their unorthodox trading strategies and outsized fees and compensation, are viewed with scorn. There is a jealousy factor at work too. With a net worth estimated at $3 billion, Griffin is one of a select few of hedge fund managers such as Steven Cohen at SAC Capital, John Paulson at Paulson & Co., Eric Mindich at Eton Park, Raj Rajaratnam of Galleon Group and other boy wonders who were been paid billions at a young age. Those glory days may be over. They can succumb to regulation in the light of transparency or bow to market rumors in the privacy of darkness.

37 Business

October 28, 2008 White House Explores Aid for Auto Deal By EDMUND L. ANDREWS and BILL VLASIC WASHINGTON — The Bush administration is examining a range of options for providing emergency financial help to spur a merger between General Motors and Chrysler, according to government officials. People familiar with the discussions said the administration wanted to provide financial assistance to the deeply troubled Big Three Detroit automakers, possibly by using the Treasury Department’s wide-ranging authority under the $700 billion bailout program that Congress approved this month. Another option under consideration is to tap a $25 billion loan program that Congress just created to help the auto companies modernize their plants. A third option would involve going back to Congress, immediately after the Nov. 4 election, for authority to spend funds aimed specifically at the auto industry. But officials have not yet decided how much assistance to provide or how to structure any aid program. G.M. and the parent of Chrysler, Cerberus Capital Management, are in talks to possibly merge the two companies, which are losing sales and hemorrhaging cash. People close to the talks said G.M. needs between $5 billion and $10 billion in assistance, mainly to cover G.M.’s own needs between now and the time of the merger. Any financial help from the government could help provide a level of confidence to investors in such a deal, and possibly cover some of the revamping costs of a merger, which would be substantial. The government’s bailout program was originally created to rescue banks and other financial institutions, but the Treasury Department decided last week to allow some insurance companies to participate as well. A bailout for carmakers would be the latest in a series of government-financed rescue efforts for banks, Wall Street firms and an insurance conglomerate. While few experts dispute the car industry’s troubles, rescuing them would also increase political pressure to help ailing industries like airlines and steel producers. The automobile industry and lawmakers from Michigan are now arguing that the car companies should be included, because their financing subsidiaries, which have been starved for credit, represent an important channel for consumers to obtain loans to buy cars. Any federal help for the financing units could be used to provide car loans, which is seen as crucial to increasing car sales. Many dealers have had trouble closing sales with car buyers because of tighter lending standards. On Monday, White House officials said the car companies might well be eligible for some sort of help under the broader financial rescue program, known as the Troubled Assets Relief Program, or TARP. “It’s clear that the automakers are dealing with a very serious situation, they have for some time,” Dana Perino, the White House press secretary, told reporters on Monday.

38 “Automakers do have financing arms — many of them do — and it’s possible that some of those financing arms could be a part of the rescue package,” she continued. “We’re trying to work with them as much as we can. There are some things we may or may not be able to do.” A spokesman for G.M., Greg Martin, said Monday that the company had been asking the Treasury Department to extend aid to automakers as it had to other troubled industries. “We believe the federal government should consider using all the tools available to it, including some recently enacted, to support industries that are in distress and that are essential to the U.S. economy,” Mr. Martin said. A spokeswoman for GMAC Financial Services, Gina Proia, said the auto finance operation was also seeking assistance from Treasury. “At this point we are working with the government officials to understand the application process of TARP and other programs to determine any potential participation by GMAC,” she said. All three of the major American car companies were already struggling with slumping car sales, soaring gasoline prices and huge losses. But their financial conditions became much worse in the last two months as the credit markets became frozen, unemployment jumped sharply and the economy reached the brink of a recession. With only a week left before the presidential election, the political and economic stakes have increased. The Republican nominee, Senator John McCain, several weeks ago abandoned hope of winning in Michigan, which has the highest unemployment rate in the country. But the collapse of a major car manufacturer would send shock waves through Indiana, Ohio and potentially other crucial states with large auto plants or suppliers. The Treasury Department, which oversees the rescue program, warned on Monday that the car companies would not be eligible for the capital injections that the government was offering to banks and some insurance companies. Under that program, the government hopes to invest $250 billion in banks and would receive nonvoting preferred shares in exchange. “The capital purchase program is available only to federally regulated banks and savings institutions,” said Michele Davis, a spokesperson for the Treasury. But Treasury officials did not rule out other forms of assistance. Under the law that Congress passed in early October, the Treasury Department has almost unlimited discretion to buy up any kind of assets from any kind of financial institution. Ford and Chrysler both have financing units, mainly for the purpose of providing car loans, and those subsidiaries might qualify as financial institutions. General Motors is in a more complicated situation, however, because it spun off 49 percent of its financing unit, the General Motors Acceptance Corporation, to Cerberus, the same private- equity firm that acquired Chrysler after the breakup of DaimlerChrysler last year. Both G.M. and Chrysler are in desperate need of cash to stave off possible bankruptcy filings. G.M., the nation’s largest automaker, lost $18.8 billion in the first six months of the year, and is burning through more than $1 billion in cash each month. The company had $21 billion in cash as of June, but is rapidly depleting its reserves to offset declining revenues. Analysts said that at its current cash-burn rate, G.M. would fall below its minimum operating requirements by sometime next year. A merger with Chrysler would give G.M. access to about $11.7 billion in cash that was on Chrysler’s books as of June.

39 Talks between G.M. and Cerberus are continuing, and the sides are said to be committed to reaching a deal soon, according to people close to the discussions. The failure of General Motors, the Ford Motor Company or Chrysler would have broad consequences for the economy. The companies combined employ more than 200,000 people in the United States, and indirectly support jobs for millions more Americans through their suppliers and dealerships. If any of the three companies were to go bankrupt, it would probably leave behind huge liabilities for federal and state governments. Shortfalls in their pension plans would become the responsibility of the Pension Benefit Guaranty Corporation, and its reserves are already stretched. G.M. has been in merger talks since September with Chrysler’s majority owner, the private equity firm Cerberus Capital Management. But the two sides have been unable so far to secure new financing from banks and other lenders, according to people with knowledge of the talks. Edmund L. Andrews reported from Washington, and Bill Vlasic from Detroit.

October 28, 2008

40 World Business

October 28, 2008 Hardships Past Haunt Europe’s Search for Financial Safety By KATRIN BENNHOLD “I haven’t forgotten history,” says Gert Heinz, a tax adviser in Munich. “If you depend on paper money you can lose everything. We’ve learned that the hard way after two world wars.” So when Chancellor Angela Merkel went on television recently to tell Germans that their bank accounts were safe, Mr. Heinz, who at 68 still remembers the rows of canned food that his mother hoarded in the attic, decided he would rather be safe than sorry. He converted a chunk of his savings into gold, as he had done before, and stocked up on a six- month supply of rice, sugar, flour and a special brand of powdered milk that lasts 50 years. Mr. Heinz may be an extreme example, but he is not alone. As Europeans grasp for security in the face of a financial storm that — at least so far — has affected them much less directly than it has many Americans, they are reflecting the history of their tortured continent that has weathered wars, revolutions and financial crises over the centuries. In America, wealth and retirement savings are much more tied up in the stock market, with a majority of people owning at least a modest stake. By contrast, only 13 percent of German households and 24 percent of French ones own shares, according to 2006 figures compiled by the European Savings Institute. Pensions are still largely state-dependent, not 401(k)-style investment accounts. Even when it comes to the British, who look more like Americans in terms of credit card debt and mortgage exposure, only about 20 percent of them have invested in the market. But history teaches Europeans that situations can go from bad to perilous in no time, and there is a long tradition of burying the family’s treasures in troubled times. Today, tales of people taking money out of banks to keep it somewhere else are circulating in Paris bistros and London pubs. Companies selling or renting safes are reporting an increase in demand, though hard numbers are difficult to establish. “History matters. In times of crisis you really get to know a country and its people,” said Toni Pierenkemper, a professor of economic history at the University of Cologne. “Traumatic events are seared into the collective consciousness and often survive into the next generations.” Germany, where many people lost their savings twice in the 20th century, is one of the richest laboratories of European historical scars — welts that help explain the country’s fears of inflation and its interest in maintaining a complex public-private banking system. The network of 446 publicly owned savings banks, which have been guaranteeing their deposits for three decades, have proved popular in recent weeks as nervous Germans moved part of their savings onto their books.

41 Even more striking, an overwhelming surge in the demand for gold has forced several Internet sites and gold vendors to temporarily shut down their sales operations. “I’ve never seen anything like this. We’re basically sold out until the end of the year,” said Robert Hartmann, co-founder of ProAurum, a gold vendor based in Munich. Despite the shortages, about 200 customers line up at his counter every day, he said, asking for coins and bars, not certificates. According to his estimates, “only about 5 percent of German investors” are interested in converting a small part of their savings to gold, “but it’s enough to put suppliers under strain.” Mr. Heinz, the financial adviser, started diversifying his investments in the run-up to the 1992 recession. He has bought land and gradually built up a stock of gold, which he buys in small units, suitable as emergency payment, and keeps in a safe at a specialized company. He still remembers the stories his grandfather told of a suitcase full of bank notes buying no more than a loaf of bread during hyperinflation in 1923 Germany. He also remembers the currency changes of 1948 that again wiped out savings. Some of his clients share his fears. He tells of one who dropped a vacuum-sealed sack filled with gold and silver to the bed of a lake, in case the government were to levy an additional tax on wealth or try to ban households from owning gold. Another has buried his riches in a forest. All of them think that it is just a matter of time until another wave of inflation erodes the value of money. In France, gold sales have also increased, but people are more trusting of the government. The entire French banking system was state-owned as recently as 1987, and the state has played a prominent role in the economy since the days of Louis XIV. In contrast to the unease and suspicion among most Americans about the government’s taking stakes in distressed financial institutions, many people in France feel comforted — even vindicated — in their belief that the state has a responsibility to look after the economy. “We nationalize, we denationalize and we re-nationalize. That’s the way it goes in France,” quipped Bernard Candiard, director-general of Crédit Municipal de Paris, a 231-year-old pawnshop that is doing brisk business these days and is a state monopoly. “Today we are again in the process of nationalizing banks,” Mr. Candiard said. “It’s a sign to the people. The government wants to give them confidence. And confidence in France is the state. We have a different tradition from America.” In Eastern Europe, where people were hardened by modest living standards under communism, the trust in government is more limited. In Bulgaria, for example, more people are looking at moving their money to Western European banks that offer an unlimited deposit guarantee, said Adriana Alexandrova, a financial executive at the TV2 television station in Sofia. “There is a great lack of confidence in the banks and in the government,” said Ms. Alexandrova, whose parents had to close their pharmacy business after several Bulgarian banks collapsed in the mid-1990s. “Several of my friends have moved their money to safes.” The phrase “he likes gardening” has a double meaning in Bulgaria, she added. “Under communism it mainly meant people preparing jars of pickled vegetables to get them through the winter. These days, it mainly means someone is hiding money in their garden or under their pillow.”

42 In Praise of Bankruptcy Daily Article by Henry Thompson | Posted on 10/28/2008 In one word, the market approach to the financial problem is bankruptcy. Firms go bankrupt when they do not have enough revenue to pay their bills. Banks make money by borrowing from lenders at a low interest rate and lending to borrowers at a higher interest rate. If banks make bad loans and borrowers quit repaying, banks go bankrupt. Insurance firms help people avoid risk, collecting premiums to pay those who suffer bad luck. If the premiums collected by an insurance firm are less than what it has to pay, it goes bankrupt. AIG sold insurance policies to stockholders that banks and other firms would not go bankrupt and could not pay the policies when that happened. Bankruptcy is a normal part of economic life, covered by laws that guarantee stockholders will be compensated as much as possible. More efficient firms move in to take over what is left of bankrupt firms, buying what can be put to productive use. There is no crime in bankruptcy and, if handled quickly, little economic harm. When the largest US energy company Enron went bankrupt a few years ago, there was not even a ripple in the energy markets, much less the economy. Bankruptcy is not criminal and should not be a surprise, but it can be unnerving if large, well-known firms go bankrupt. Banks and insurance firms are careful when lending or selling policies because they want to ensure their revenue will pay their bills. Government involvement, however, provides a cushion for failure and allows banks and insurance firms to be careless. This carelessness occurred with the government-sponsored mortgage bank, the Federal National Mortgage Association. Fannie Mae provides backing to mortgage banks, more or less encouraging them to make bad loans. Fannie Mae makes subsidized loans to mortgage companies when they are short of cash. Freddie Mac is a government mortgage bank that sells mortgages without the usual worry of making a profit, given its taxpayer backing. The government has taken over these two losing mortgage banks, and losses will be paid by taxpayers. The government provides subsidized mortgage insurance in case home buyers cannot pay. This insurance lets commercial mortgage banks relax and make loans to people who might not be able to pay. Government support for people wanting to buy a house elevated demand for houses and pushed up prices. Rising prices made home buyers confident they could buy a house they could not afford and sell it soon for a profit, counting on a "greater fool" to come

43 along. Realistically, people should only buy a house when they plan to live in it and can actually pay for it. Greater fools do not always come along. The result of government meddling in the mortgage market is that people have bought houses they cannot afford. When prices quit going up, people were left owing more on their house than it was worth in the market. With their subsidized mortgage insurance and little penalty, people defaulted on their mortgages. The mortgage banks are left without income. This mortgage mess is the root cause of the present financial crisis. One part of the evolving financial bailout is the government using taxpayer money to help people who have not been able to pay their mortgage. The government is taxing those who have paid their mortgages and transferring the money to those who have not. It is not a good idea to reward inefficiency. The government is also giving money to select financial and insurance firms, rewarding their poor performance with taxpayer money. Better advice is, "Don't throw good money after bad." The failed firms should go bankrupt. Another part of the bailout plan is that the Treasury will actually buy houses with defaulted mortgages that the failing banks are holding - the overpriced mortgages that people quit paying. The Treasury has become a realty speculator, hoping to sell these overpriced houses sometime in the future for an even higher price. It is much more likely that taxpayers will pay the losses. The bailout money will purchase 6% of the houses in the United States - not such a large amount and only a very small part of the total real-estate market. The bailout money, as large as it is, will have little effect on the aggregate housing market. As another part of the bailout, the Federal Reserve will make short-term loans to troubled banks and insurance companies to meet their payroll or other bills. The Fed's job is to make loans to banks and buy or sell bonds to control the money supply. Certainly the bankrupt firms will be first in line to borrow such short-term funds. These loans are likely to go unpaid and be written off at taxpayer expense. It is easy for the Fed to make loans since it is in charge of the money supply. In the bailout, the Treasury also plans to buy a stake in the failed firms, using taxpayer money to become part owner of second-rate mortgage banks and insurance firms - your tax dollars at work. The underlying goal of the financial bailout is not to keep the economy "healthy" but to keep a few Wall Street firms, mortgage banks, and insurance firms in business. Never mind that most mortgage and insurance firms in the country are profitable; the government wants to support the inefficient, large, high-profile firms. If these firms were allowed to go bankrupt, the economy would recover quickly. Other firms, not necessarily with an address on Wall Street, would step in and buy them out. Wall Street is much less important now than in the past, due to national and global financial competition. Profit motives in business are clear, but governments have no profit motive and are able to collect taxes, print money, and borrow against future taxpayer money to pay their bills. Mortgage and other financial-market firms will wait to see what the government agencies do in the market and then generally do the opposite, playing against taxpayer money. The rules are changing with more government involvement, but competition will continue. The situation would be like the government making delivery of packages less than 5 pounds illegal except by the US Post Office. The present financial problems would disappear quickly if the government let the markets operate and let inefficient firms go bankrupt. The irony is that the government is stepping in to solve the problems it created. The solution might "work," but the underlying disincentives in the mortgage and insurance markets will persist. Increased government meddling in the financial markets will only make the financial problems linger.

A Voting Rights Disaster? By Christopher Edley Jr. Tuesday, October 28, 2008; A17 Suppose in your neighborhood there are 600 registered voters per machine, while across town there are only 120 per machine. (That's a 5 to 1 disparity, which is what exists in some places in Virginia today.) On Election Day, your line wraps around the block and looks to be a four- hour wait, while in other areas lines are nonexistent. This ought to be a crime. It amounts to a "time-tax" on your right to vote, and some of your neighbors will undoubtedly give up and go home. This scenario raises three questions: Nationwide, will it discourage tens of thousands, or untold millions? Which presidential candidate and down-ballot candidates might benefit from this "tax"? And what can be done in the next few days? Voting rights advocates, watching this slow-motion train wreck that could disenfranchise so many minority voters, have filed emergency litigation in Virginia and Pennsylvania demanding that, at the very least, officials be prepared with plenty of paper ballots and reserves of competent poll workers. More litigation may follow elsewhere. Judges can hold official feet to the fire, but they shouldn't have to. Assigning blame -- whether the fingers are being pointed at Congress or the Justice Department, county registrars or state legislators -- isn't crucial this week. Neither is this the time to focus on the reasons for failure -- whether indifference, incompetence, indolence or animus. What's crucial is that state and local officials nationwide salvage the situation by implementing second-best strategies: For starters, redistribute machines on the basis of voter registration, instead of assuming that minorities won't show up. Stockpile paper ballots, under lock and key, and offer a paper ballot voting option if wait times reach 45 minutes. Train platoons of reserve poll workers and stand by to shuttle them where they are needed. Commit right now to holding the polls open late if necessary. Advertise what you're prepared to do. For heaven's sake, a lot of people bled for this opportunity. In 2001, former presidents Jimmy Carter and Gerald Ford led a commission, of which I was a member, to dissect the previous year's voting fiasco in Florida. Many of our recommendations found their way into the Help America Vote Act of 2002. Disappointingly, Congress failed to create an explicit and easily enforceable prohibition against grossly disproportionate resource allocations between polling places in the same state or even the same county -- the level of government at which, preposterously, we typically finance and administer elections. This localism means that the infrastructure of democracy vies for resources with potholes, parks, sheriffs and firefighting. It also means that locally powerful communities get better service on something that -- above all else -- is supposed to be scrupulously equal in this country. Even without a new statute, there are enough plausible legal theories on this to boggle the mind. Voting is a fundamental right, but as I saw on the Carter-Ford commission and again as a member of the U.S. Civil Rights Commission, Election Day resource disparities have enormously different racial and class impacts that are based on the dynamics of power and poverty. In election cycle after cycle, registrars act surprised when problems crop up disproportionately in poor neighborhoods. If there isn't enough money to run decent elections everywhere, Americans should share the pain equally.

45 We have the equal protection clause of the U.S. Constitution, with a rich history of helpful Supreme Court rulings, including even Bush v. Gore's solicitude in 2000 for Florida voters being treated differently and arbitrarily in the administration of elections. We have the Voting Rights Act of 1965 and Title VI of the Civil Rights Act of 1964. Individual states have their own constitutional provisions guaranteeing "equal protection" and "due process," so state attorneys general can and should add pressure. Laws aside, if all else fails, there's common sense. Next Tuesday is certain to be the mother of all turnout stories. The signs pointing to this include registration data, turnout in the primaries, polling data on public enthusiasm, early-voting volume, and Barack Obama's ground game. Polling sites should prepare to be overwhelmed, not underused. This election features the first African American major-party presidential nominee, 40 years after the murder of the voting rights leader Martin Luther King Jr., running a race almost beyond race, with Obama, Hillary Clinton, Bill Clinton, Sarah Palin and George W. Bush generating the kind of excitement that happens about twice in a century. We must not let managerial failures threaten the hopefulness of newly engaged communities and young voters. Do we need election monitors from Canada or South Africa? Perhaps France? The world is watching. When we awake Nov. 5, no failure of administration should have tarnished our outsize pride in our democracy. Christopher Edley Jr. is dean of the Berkeley Law School and co-founder of Berkeley's Chief Justice Warren Institute for Race, Ethnicity and Diversity. He was a member of the National Commission on Federal Election Reform and is an unpaid adviser to the Obama campaign.

46 Opinion

October 28, 2008 OP-ED COLUMNIST The Behavioral Revolution By DAVID BROOKS Roughly speaking, there are four steps to every decision. First, you perceive a situation. Then you think of possible courses of action. Then you calculate which course is in your best interest. Then you take the action. Over the past few centuries, public policy analysts have assumed that step three is the most important. Economic models and entire social science disciplines are premised on the assumption that people are mostly engaged in rationally calculating and maximizing their self-interest. But during this financial crisis, that way of thinking has failed spectacularly. As Alan Greenspan noted in his Congressional testimony last week, he was “shocked” that markets did not work as anticipated. “I made a mistake in presuming that the self-interests of organizations, specifically banks and others, were such as that they were best capable of protecting their own shareholders and their equity in the firms.” So perhaps this will be the moment when we alter our view of decision-making. Perhaps this will be the moment when we shift our focus from step three, rational calculation, to step one, perception. Perceiving a situation seems, at first glimpse, like a remarkably simple operation. You just look and see what’s around. But the operation that seems most simple is actually the most complex, it’s just that most of the action takes place below the level of awareness. Looking at and perceiving the world is an active process of meaning-making that shapes and biases the rest of the decision-making chain. Economists and psychologists have been exploring our perceptual biases for four decades now, with the work of Amos Tversky and Daniel Kahneman, and also with work by people like Richard Thaler, Robert Shiller, John Bargh and Dan Ariely. My sense is that this financial crisis is going to amount to a coming-out party for behavioral economists and others who are bringing sophisticated psychology to the realm of public policy. At least these folks have plausible explanations for why so many people could have been so gigantically wrong about the risks they were taking. Nassim Nicholas Taleb has been deeply influenced by this stream of research. Taleb not only has an explanation for what’s happening, he saw it coming. His popular books “Fooled by Randomness” and “The Back Swan” were broadsides at the risk-management models used in the financial world and beyond. In “The Black Swan,” Taleb wrote, “The government-sponsored institution Fannie Mae, when I look at its risks, seems to be sitting on a barrel of dynamite, vulnerable to the slightest hiccup.” Globalization, he noted, “creates interlocking fragility.” He warned that while the growth of giant banks gives the appearance of stability, in reality, it raises the risk of a systemic collapse — “when one fails, they all fail.”

47 Taleb believes that our brains evolved to suit a world much simpler than the one we now face. His writing is idiosyncratic, but he does touch on many of the perceptual biases that distort our thinking: our tendency to see data that confirm our prejudices more vividly than data that contradict them; our tendency to overvalue recent events when anticipating future possibilities; our tendency to spin concurring facts into a single causal narrative; our tendency to applaud our own supposed skill in circumstances when we’ve actually benefited from dumb luck. And looking at the financial crisis, it is easy to see dozens of errors of perception. Traders misperceived the possibility of rare events. They got caught in social contagions and reinforced each other’s risk assessments. They failed to perceive how tightly linked global networks can transform small events into big disasters. Taleb is characteristically vituperative about the quantitative risk models, which try to model something that defies modelization. He subscribes to what he calls the tragic vision of humankind, which “believes in the existence of inherent limitations and flaws in the way we think and act and requires an acknowledgement of this fact as a basis for any individual and collective action.” If recent events don’t underline this worldview, nothing will. If you start thinking about our faulty perceptions, the first thing you realize is that markets are not perfectly efficient, people are not always good guardians of their own self-interest and there might be limited circumstances when government could usefully slant the decision- making architecture (see “Nudge” by Thaler and Cass Sunstein for proposals). But the second thing you realize is that government officials are probably going to be even worse perceivers of reality than private business types. Their information feedback mechanism is more limited, and, being deeply politicized, they’re even more likely to filter inconvenient facts. This meltdown is not just a financial event, but also a cultural one. It’s a big, whopping reminder that the human mind is continually trying to perceive things that aren’t true, and not perceiving them takes enormous effort.

48 OPINION OCTOBER 27, 2008 The Age of Prosperity Is Over This administration and Congress will be remembered like Herbert Hoover. By ARTHUR B. LAFFER

About a year ago Stephen Moore, Peter Tanous and I set about writing a book about our vision for the future entitled "The End of Prosperity." Little did we know then how appropriate its release would be earlier this month. Financial panics, if left alone, rarely cause much damage to the real economy, output, employment or production. Asset values fall sharply and wipe out those who borrowed and lent too much, thereby redistributing wealth from the foolish to the prudent. This process is the topic of Nassim Nicholas Taleb's book "Fooled by Randomness."

David Gothard When markets are free, asset values are supposed to go up and down, and competition opens up opportunities for profits and losses. Profits and stock appreciation are not rights, but rewards for insight mixed with a willingness to take risk. People who buy homes and the banks who give them mortgages are no different, in principle, than investors in the stock market, commodity speculators or shop owners. Good decisions should be rewarded and bad decisions should be punished. The market does just that with its profits and losses. No one likes to see people lose their homes when housing prices fall and they can't afford to pay their mortgages; nor does any one of us enjoy watching banks go belly-up for making subprime loans without enough equity. But the taxpayers had nothing to do with either side of the mortgage transaction. If the house's value had appreciated, believe you me the

49 overleveraged homeowner and the overly aggressive bank would never have shared their gain with taxpayers. Housing price declines and their consequences are signals to the market to stop building so many houses, pure and simple. But here's the rub. Now enter the government and the prospects of a kinder and gentler economy. To alleviate the obvious hardships to both homeowners and banks, the government commits to buy mortgages and inject capital into banks, which on the face of it seems like a very nice thing to do. But unfortunately in this world there is no tooth fairy. And the government doesn't create anything; it just redistributes. Whenever the government bails someone out of trouble, they always put someone into trouble, plus of course a toll for the troll. Every $100 billion in bailout requires at least $130 billion in taxes, where the $30 billion extra is the cost of getting government involved. If you don't believe me, just watch how Congress and Barney Frank run the banks. If you thought they did a bad job running the post office, Amtrak, Fannie Mae, Freddie Mac and the military, just wait till you see what they'll do with Wall Street. Some 14 months ago, the projected deficit for the 2008 fiscal year was about 0.6% of GDP. With the $170 billion stimulus package last March, the add-ons to housing and agriculture bills, and the slowdown in tax receipts, the deficit for 2008 actually came in at 3.2% of GDP, with the 2009 deficit projected at 3.8% of GDP. And this is just the beginning. The net national debt in 2001 was at a 20-year low of about 35% of GDP, and today it stands at 50% of GDP. But this 50% number makes no allowance for anything resulting from the over $5.2 trillion guarantee of Fannie Mae and Freddie Mac assets, or the $700 billion Troubled Assets Relief Program (TARP). Nor does the 50% number include any of the asset swaps done by the Federal Reserve when they bailed out Bear Stearns, AIG and others. But the government isn't finished. House Speaker Nancy Pelosi and Senate Majority Leader Harry Reid -- and yes, even Fed Chairman Ben Bernanke -- are preparing for a new $300 billion stimulus package in the next Congress. Each of these actions separately increases the tax burden on the economy and does nothing to encourage economic growth. Giving more money to people when they fail and taking more money away from people when they work doesn't increase work. And the stock market knows it. The stock market is forward looking, reflecting the current value of future expected after-tax profits. An improving economy carries with it the prospects of enhanced profitability as well as higher employment, higher wages, more productivity and more output. Just look at the era beginning with President Reagan's tax cuts, Paul Volcker's sound money, and all the other pro-growth, supply-side policies. Bill Clinton and Alan Greenspan added their efforts to strengthen what had begun under President Reagan. President Clinton signed into law welfare reform, so people actually have to look for a job before being eligible for welfare. He ended the "retirement test" for Social Security benefits (a huge tax cut for elderly workers), pushed the North American Free Trade Agreement through Congress against his union supporters and many of his own party members, signed the largest capital gains tax cut ever (which exempted owner-occupied homes from capital gains taxes), and finally reduced government spending as a share of GDP by an amazing three percentage points (more than the next four best presidents combined). The stock market loved Mr. Clinton as it had loved Reagan, and for good reasons. The stock market is obviously no fan of second-term George W. Bush, Nancy Pelosi, Harry Reid, Ben Bernanke, Barack Obama or John McCain, and again for good reasons.

50 These issues aren't Republican or Democrat, left or right, liberal or conservative. They are simply economics, and wish as you might, bad economics will sink any economy no matter how much they believe this time things are different. They aren't. I was on the White House staff as George Shultz's economist in the Office of Management and Budget when Richard Nixon imposed wage and price controls, the dollar was taken off gold, import surcharges were implemented, and other similar measures were enacted from a panicked decision made in August of 1971 at Camp David. I witnessed, like everyone else, the consequences of another panicked decision to cover up the Watergate break-in. I saw up close and personal Presidents Gerald Ford and George H.W. Bush succumb to panicked decisions to raise taxes, as well as Jimmy Carter's emergency energy plan, which included wellhead price controls, excess profits taxes on oil companies, and gasoline price controls at the pump. The consequences of these actions were disastrous. Just look at the stock market from the post-Kennedy high in early 1966 to the pre-Reagan low in August of 1982. The average annual real return for U.S. assets compounded annually was -6% per year for 16 years. That, ladies and gentlemen, is a bear market. And it is something that you may well experience again. Yikes! Then we have this administration's panicked Sarbanes-Oxley legislation, and of course the deer-in-the-headlights Mr. Bernanke in his bungling of monetary policy. There are many more examples, but none hold a candle to what's happening right now. Twenty-five years down the line, what this administration and Congress have done will be viewed in much the same light as what Herbert Hoover did in the years 1929 through 1932. Whenever people make decisions when they are panicked, the consequences are rarely pretty. We are now witnessing the end of prosperity. Mr. Laffer is chairman of Laffer Associates and co-author of "The End of Prosperity: How Higher Taxes Will Doom the Economy -- If We Let it Happen," just out by Threshold.

October 27, 2008 We Need Reagan + Friedman + Keynes By Larry Kudlow Back in early 1981, when I went to Washington to work for President Reagan, one of the architects of supply-side economics, Columbia University’s Robert Mundell, visited my OMB budget-bureau office inside the White House complex. At the time we were suffering from double-digit inflation, sky-high interest rates, a long economic downturn, and a near 15-year bear market in stocks. So I asked Prof. Mundell, who later won a Nobel Prize in economics, if President Reagan’s supply-side tax cuts would be sufficient to cure the economy. The professor answered that during periods of crisis, sometimes you have to be a supply-sider (tax rates), sometimes a monetarist (Fed money supply), and sometimes a Keynesian (federal deficits). I’ve never forgotten that advice. Mundell was saying: Choose the best policies as put forth by the great economic philosophers without being too rigid. Of course, John Maynard Keynes was a deficit spender during the Depression. Milton Friedman warned of printing too much or too little money. And Mundell, along with Art Laffer, Jack Kemp, and others, revived the importance of reducing high marginal tax rates to reward work, investment, and risk. The idea was to make each of these activities pay more after tax, and in the process boost asset values across-the-board. This incentive model of economic growth was used effectively by President John F. Kennedy and the great 1920s Treasury man, Andrew Mellon. During the 1980s Reagan enacted Mundell’s three-legged approach. He slashed tax rates on the supply-side and was not afraid to run budget deficits in the Keynesian mold. At the same time, Reagan gave Paul Volcker carte blanche to practice the tough-minded monetarism that curbed excess money and vanquished inflation. This eclectic policy mix reignited economic growth, and it ushered in a three-decade prosperity boom that revived free-market capitalism. Today, however, the economic naysayers are ignoring the advice of Prof. Mundell. Looking at our financial crisis, with its deflationary sweep from stock markets to home prices to energy, they want to lurch leftward to a big-government tax-and-spend regulatory approach. Instead, we need to put all three legs of the Mundell hypothesis in place. And we’re already two-thirds of the way there. Treasury man Henry Paulson is using a $700 billion rescue package to prop up banks with new capital, purchase distressed assets, and backstop inter-bank lending. Keynesian deficits will finance it. But it’s working. While ankle biters on the left and right have dissed Paulson’s plan, important credit-market spreads have declined significantly in the last two weeks.

52 Fed head Ben Bernanke, meanwhile, is combating deflation with a Friedmanite monetarist approach -- the second leg of the Mundell mix. Over the past two months the Fed has doubled its balance sheet and spurred a major increase in the basic money supply in order to meet the enormous liquidity demands that always accompany deflation. The Fed should keep this up in the coming months until stocks, commodities, and credit show life-signs of recovery. But what’s missing is Mundell’s third policy leg: supply-side tax cuts. And here we find the partisan debate of the closing days of the presidential and congressional elections. Democrats want to tax the rich, redistribute the wealth, and spend our way out of the economic doldrums. It won’t work. Senators Barack Obama and Harry Reid, along with Speaker Nancy Pelosi, disdain supply-side tax incentives. But Sen. John McCain wants to reemploy them as a recovery tool. McCain is right, and now is the time for the Republican party to call for sweeping tax cuts that would reduce marginal rates by half for businesses, individuals, and investors. Yes, it would be bold. But no bolder than Reagan in the 1980s, Kennedy in the 1960s, or Mellon in the 1920s. The corporate tax rate should be slashed from 35 percent to less than 25 percent, including capital-gains. (Corporations, let’s not forget, don’t pay taxes. Only individuals do, since business costs are passed along to consumers.) The top individual rate should similarly be lowered, with fewer income brackets to clutter up the tax code. And investment taxes on capital-gains and dividends should be cut from 15 percent to 7.5 percent to revive the dormant animal spirits of investors. These tax cuts would mean all three legs of Robert Mundell’s pragmatic approach to policy are in place. Use the money supply to combat deflation (inflation is not the problem), employ deficits to rescue and stabilize the banking and credit system, and slash tax rates to reignite economic growth. In effect, a successful rescue plan requires a drawdown of all the major economic schools of thought. Given the current economic emergency, we need all the help we can get. For a change, how about a little pragmatism in the policy mix? That just might do the trick.

A Suspended Apocalypse Does anyone in the financial world actually know how to prevent the coming disaster? Is it too late? Bernard-Henri Levy, The New Republic Published: Monday, October 27, 2008

We are living in an extraordinary time. The world has been badly shaken. In the space of a few days a system that we thought was as secure and assured as the air we breathe lost all its landmarks, its clarity, and was seemingly swallowed up by a black hole. Money--essential to the spirit of peace- -congealed, like blood in veins. Credit--this fine word is also expressive of people's faith in others--like a machine that jammed, and then stopped. Confidence--the famous "confidence" that is also integral to the pact among citizens and the reasons it must be perpetuated--like a spell that is evaporating. The situation brings to mind the words of Thomas Hobbes, whose attempts to shed light on the enigma of the institutions of society were only taken half-seriously. One recalls "Leviathan," Rousseau's "Social Contract," de la Boetie's "Discourse on Voluntary Servitude": theories that had almost fallen out of view but in fact described what is taking place now in plain view, during a worldwide crisis unprecedented in the history of our various capitalisms. What is a social bond and how is it broken? Voila. Here we are. This debacle, this shipwreck, is showing us the answer. What is political time and how does it get away from us? Take the four days wasted by American legislators before they committed to voting for the Paulson plan; take those four short days that we know really counted doubly, triply, perhaps even more, wreaking irreparable damage. That is the type of procrastination that prevails in situations that qualify as "pre- revolutionary." Is man a predator of man? Does the fear of this predator slumber within us? An anxiety, formerly concealed by a poorly applied varnish of civilization, about a state of nature that is re-emerging? Consider the princes of finance, once so polite, so complicit, so civilized, who have been facing each other at the edge of the abyss, waiting to see who will be the next to fall; consider that dance of wolves, the ferocious ballet of battered predators sniffing at each other, detecting the scent of death on their neighbors, coveting their remains; consider the tango of white-hot hate that has been discreetly called the "drying up of interbank credit." The scent of execution and of collective suicide has been circulating in the middle of the pack. It is as though we have been watching a deadly dance around a fire, where those same people who, through their irresponsibility, devastating egoism and, it must be said, intelligence, turned mad and led the financial world toward implosion, thinking that they could pull themselves out of the furnace by pushing the others in first. And the result has been, for all of us, a suspended apocalypse, in which it is easy to lay out the implacable chain of consequences, but also a situation in which no one knows how to defuse the mechanism. How to respond if account holders attempt to withdraw cash that the banks no longer

54 have? How should we react if electrical and gas utilities default on payment to their employees? What will happen when an angry mob of ruined savers, mainstream borrowers harassed by those who pressured them to go into debt in the first place, and the desperate and unemployed erupt in protest and--according to a scenario that we in France know too well--shout their rage beneath the windows of the speculators, loan sharks and others with golden parachutes? At that time, those who are responsible will have two options. They are all afflicted with an ignorance of the dark, unknown world bristling with new threats that they enter with us. They are all feeling their way, stumbling along. Many leaders have had a terrible time avoiding here a gaffe, a rhetorical stumble there, at quelling the nearly imperceptible bodily ticks that betray one's vertigo. Nonetheless, there are distinctions among them. There are always leaders who--as historian Raymond Aron said of a defeated former French president--ignore the fact that "History is tragic," who believe that everything always works out. After such a long time, they think, isn't History used to playing out its confrontations without hurting anyone? Is it not dedicated to convulsions that are not and will never be more than innocent pirouettes? Conversely, there are those who are sensitive to Tragedy, who know that nothing is more fragile, precarious and quick to disintegrate than a well-established social bond--as French poet Paul Valery wrote, "All that holds it together is magic." You start off with a financial crisis, then the whole cloth begins to unravel little by little: At the beginning you have a terrified crowd, and at the end, a lynch mob. In the second category, we find French President Nicolas Sarkozy. In the second category, we sometimes find leaders who--like him--are concentrated, determined, inhabited by circumstance at the same time they grapple with it: In their gaze, we see a little of the lucid terror that defines all great statesmen. Bernard-Henri Levy's new book, Left in Dark Times: A Stand Against The New Barbarism, was published in September by Random House. This piece was translated from the French by Sara Sugihara.

Look who pays for the bailout

Meet the Henrys (high earners, not rich yet). They make $250,000-plus and get taxed to high heaven. And they're about to get socked again. By Shawn Tully with Joan Caplin LAST UPDATED: OCTOBER 27, 2008: 12:37 PM ET

Bill and Kira Kwon of Peoria, Ill. He is a wealth advisor at Morgan Stanley earning $375,000 a year, and she is a freelance HENRY: High Earner, Not Rich Yet photographer.

The American dream - on hold Meet the HENRYs (high earners, not rich yet). They make $250,000-plus and get taxed to high heaven. And they're about to be socked again. (Fortune magazine) -- Bill Kwon is the embodiment of the American dream. His father - who was arrested by North Korean Communists in the early 1950s for championing democracy - brought the family from Seoul to Illinois when he was a baby. Bill worked himself ragged pursuing every opportunity America's heartland offered, never leaving Peoria. Just out of college, he was earning a six-figure salary at a telecom company and sleeping in his parents' basement. Now he's a wealth advisor earning $375,000 at Morgan Stanley (MS, Fortune 500), with a five-bedroom brick home, a minivan, a son in private school, and three younger kids to follow. "My dad never made more than $25,000 a year," says the burly, outgoing Kwon, 39. "When I was a kid, this was the top neighborhood in Peoria. I never

56 thought I could live here."

For all his blessings, Kwon gets really steamed when politicians and pundits claim that he and other Americans in his income group aren't shouldering their "fair share" in taxes and should pay more. Nor does he appreciate being branded as "rich" when it's far from certain he'll ever build the kind of lavish nest egg the truly wealthy enjoy, especially after the current market meltdown. "I'm not a trust-fund baby," says Kwon. "Raising taxes for people at my income level is like being punished for success, for working hard." Kwon's total tax bill is already more than $100,000, and the bite is taking an ever-rising share of his raises and bonuses, not to mention his wife's income as a photographer. Kwon fears that America risks killing the incentive for people like him by shrinking the rewards for logging extra hours or starting a business, diminishing the dream that brought his father from Korea.

The Kwon family has plenty of company, representing an income group comprising five million households that earn between $250,000 and $500,000 a year and pay a large chunk of it back in taxes. These folks aren't America's hedge fund managers, investment bankers, or CEOs - who boast net worths in the multimillions and qualify as rich right now. Instead, these are the doctors, consultants, and attorneys, the marketing managers and CIOs, the owners of real estate agencies and security firms. They write the contracts, inspire the sales teams, and integrate computer systems. They own many of America's small businesses. A man aspiring to join this cohort, nicknamed Joe the Plumber, has put a face on a big issue in the presidential campaign: Whether it's fair or wise to raise taxes on the powerful job engine of America's corner stores, maintenance firms, and yes, plumbing contractors.

This is the world of the HENRYs, an acronym we'll use to describe people whose financial situation can be summed up by the phrase "high earners, not rich yet." (I coined the term for a Fortune story in 2003 on the alternative minimum tax, or AMT, the bane of the HENRYs.) Put simply, the HENRYs are the bulwark of the professional and entrepreneurial class that drives the economy. Look in the mirror, Fortune reader, and you'll probably see a HENRY. They are relentless strivers. Aspiring HENRYs played by the rules and did everything right: They won the best grades in high school, got accepted at good colleges and grad schools, and worked daunting schedules as medical interns or associates in law firms. They're an upwardly mobile group: Most HENRYs used their talent and grit to advance from the middle class, and those who got a hand from affluent parents are determined to do even better for their kids.

57 "These high earners may come from privileged, upper-middle-class backgrounds or be the children of immigrants," says Phillip Cook, a financial advisor in Torrance, Calif. "What they have in common is that they worked incredibly hard to build their careers and work incredibly hard to move ahead." Now this group of superachievers is being targeted as a cash machine. Barack Obama, the Democratic presidential nominee, has pledged to pay for middle-class tax cuts and credits by raising taxes on the HENRYs. "It's time for folks who make over $250,000 a year to pay their fair share," Obama has declared regularly on the campaign trail. Obama and the congressional Democrats frequently refer to households earning over $250,000 as the "rich" and the "wealthiest Americans." But whether the HENRYs are truly "rich," or ever will be, is debatable. In Fortune's interviews with two dozen HENRYs from Charlotte to Concord, Calif., what emerged was a portrait of families a world away from the private jets, luxury vacation homes, and heated garages with Bentleys and Porsches lined up headlight to headlight that typically represent America's vision of "rich." Kelly Lynch, the owner of a commercial maintenance company in Redondo Beach, Calif., is raising two kids with her partner, Jill Fenske, on a household income of $400,000. She's saving $800 a month for the children's college fund and $4,000 a month for retirement - a number that someday might make her rich. "If I blew my money like other people, I'd feel rich," says Lynch. Her views on taxes are befitting a born entrepreneur: "I think it would be unfair if someone tried to raise my taxes," says Lynch. "I don't think people should be penalized because they earn more." Sure, it's hard to weep for families that earn more than 98% of American households, especially when median family income stands at $50,000 and the middle class is getting pummeled by falling home and stock prices. Unlike millions of Americans, most HENRYs don't need to worry about making the next mortgage or credit card payment. Still, HENRYs are getting a bad rap from those who lump them in with America's conspicuously wealthy. While there's no consensus definition of how much wealth or income makes someone rich in America, here's a reasonable proposal: Many Americans would consider a family wealthy if it enjoyed either a large net worth today, something on the order of $3 million, or an income big enough to pay for a luxurious lifestyle - with enough left over to save for a comfortable retirement. The $3 million figure would generate around $200,000 in income, plenty to retire on tomorrow. If a couple in their 30s, 40s, or 50s has the option to stop working and live on their ample savings - call it "take this job and shove it" money - they can definitely be classified as rich. The HENRYs don't rate as rich by either standard. They're mostly two- income families. And even with two incomes they don't earn enough for luxurious lifestyles, and their savings don't remotely approach the take-this-job level. Hit hard by taxes The reason the HENRYs are strapped for both lifestyle and nest egg is twofold: First, they already face a large and rising burden for federal, state, and property taxes plus the knife of the AMT. "Taxes are by far my biggest expense," says Kwon. Second, the HENRYs invest heavily in a distinct set of high-grade staples that, in effect, defines them. They're all about the kids: saving for private colleges, paying for day care - practically a must, because Mom and Dad are both working - and providing dance, tennis, or gymnastics lessons. These might be seen as luxury items by middle-class workers, but they're absolute necessities to the HENRYs. The big tax bite and what they consider investments in their kids chew up most of the HENRYs' incomes, leaving little for either extravagant living or, in many cases, saving for

58 an affluent retirement. Indeed, the HENRYs consider themselves "well off" and "successful" but nowhere near "rich." "Wealthy people are those who have lots of cash reserves and don't have to go to work," says John Selden, 35, a dentist in Charlotte with a family income of $350,000. Adds David Twa, county administrator of Contra Costa County in California (salary: $250,000): "I feel middle class. To me, rich is people with golf-club memberships." Tony Molino, 50, an attorney in Rancho Palos Verdes, Calif., speaks for legions of HENRYs: "I've worked 50 to 60 hours my entire life, and I don't have a lot left over at the end of the month. I'm comfortable, but when Joe Biden talks about sucking it up, getting patriotic, and paying more taxes, I get livid." The HENRYs interviewed by Fortune indulge in virtually none of the toys that brand families as rich. "I eat fast food and take my kids to soccer," says Kwon. Marie Hoffman, a realtor in Hermosa Beach, Calif., keeps hearing about what affluent Americans are supposed to be buying and swears it's not her. "I see $1,400 dresses advertised in Oprah's magazine, and I can't imagine anyone buying a sheath to wear to work at that price," marvels Hoffman.

59 Business

October 26, 2008 ECONOMIC VIEW But Have We Learned Enough? By N. GREGORY MANKIW LIKE most economists, those at the International Monetary Fund are lowering their growth forecasts. The financial turmoil gripping Wall Street will probably spill over onto every other street in America. Most likely, current job losses are only the tip of an ugly iceberg. But when Olivier Blanchard, the I.M.F.’s chief economist, was asked about the possibility of the world sinking into another Great Depression, he reassuringly replied that the chance was “nearly nil.” He added, “We’ve learned a few things in 80 years.” Yes, we have. But have we learned what caused the Depression of the 1930s? Most important, have we learned enough to avoid doing the same thing again? The Depression began, to a large extent, as a garden-variety downturn. The 1920s were a boom decade, and as it came to a close the Federal Reserve tried to rein in what might have been called the irrational exuberance of the era. In 1928, the Fed maneuvered to drive up interest rates. So interest-sensitive sectors like construction slowed. But things took a bad turn after the crash of October 1929. Lower stock prices made households poorer and discouraged consumer spending, which then made up three-quarters of the economy. (Today it’s about two-thirds.) According to the economic historian Christina D. Romer, a professor at the University of California, Berkeley, the great volatility of stock prices at the time also increased consumers’ feelings of uncertainty, inducing them to put off purchases until the uncertainty was resolved. Spending on consumer durable goods like autos dropped precipitously in 1930. Next came a series of bank panics. From 1930 to 1933, more than 9,000 banks were shuttered, imposing losses on depositors and shareholders of about $2.5 billion. As a share of the economy, that would be the equivalent of $340 billion today. The banking panics put downward pressure on economic activity in two ways. First, they put fear into the hearts of depositors. Many people concluded that cash in their mattresses was wiser than accounts at local banks. As they withdrew their funds, the banking system’s normal lending and money creation went into reverse. The money supply collapsed, resulting in a 24 percent drop in the consumer price index from 1929 to 1933. This deflation pushed up the real burden of households’ debts. Second, the disappearance of so many banks made credit hard to come by. Small businesses often rely on established relationships with local bankers when they need loans, either to tide them over in tough times or for business expansion. With so many of those relationships interrupted at the same time, the economy’s ability to channel financial resources toward their best use was seriously impaired.

60 Together, these forces proved cataclysmic. Unemployment, which had been 3 percent in 1929, rose to 25 percent in 1933. Even during the worst recession since then, in 1982, the United States economy did not experience half that level of unemployment. Policy makers in the 1930s responded vigorously as the situation deteriorated. But like a doctor facing a patient with a new disease and strange symptoms, they often acted in ways that, with the benefit of hindsight, appeared counterproductive. Probably the most important source of recovery after 1933 was monetary expansion, eased by President Franklin D. Roosevelt’s decision to abandon the gold standard and devalue the dollar. From 1933 to 1937, the money supply rose, stopping the deflation. Production in the economy grew about 10 percent a year, three times its normal rate. Less successful were various market interventions. According to a study by the economists Harold L. Cole and Lee E. Ohanian, both of the University of California, Los Angeles, and the Federal Reserve Bank of Minneapolis, President Roosevelt made things worse when he encouraged the formation of cartels through the National Industrial Recovery Act of 1933. Similarly, they argue, the National Labor Relations Act of 1935 strengthened organized labor but weakened the recovery by impeding market forces. LOOKING back at these events, it’s hard to avoid seeing parallels to the current situation. Today, as then, uncertainty has consumers spooked. By some measures, stock market volatility in recent days has reached levels not seen since the 1930s. With volatility spiking, the University of Michigan’s survey reading of consumer sentiment has been plunging. Deflation across the economy is not a problem (yet), but deflation in the housing market is the source of many of our present difficulties. With so many homeowners owing more on their mortgages than their houses are worth, default is an unfortunate but often rational choice. Widespread foreclosures, however, only perpetuate the downward spiral of housing prices, further defaults and additional losses at financial institutions. The Fed and the Treasury Department, intent on avoiding the early policy inaction that let the Depression unfold, have been working hard to keep credit flowing. But the financial situation they face is, arguably, more difficult than that of the 1930s. Then, the problem was largely a crisis of confidence and a shortage of liquidity. Today, the problem may be more a shortage of solvency, which is harder to solve. What’s next? Perhaps the most troubling study of the 1930s economy was written in 1988 by the economists Kathryn Dominguez, Ray Fair and Matthew Shapiro; it was called “Forecasting the Depression: Harvard Versus Yale.” (Mr. Fair is an economics professor at Yale; Ms. Dominguez and Mr. Shapiro are at the University of Michigan.) The three researchers show that the leading economists at the time, at competing forecasting services run by Harvard and Yale, were caught completely by surprise by the severity and length of the Great Depression. What’s worse, despite many advances in the tools of economic analysis, modern economists armed with the data from the time would not have forecast much better. In other words, even if another Depression were around the corner, you shouldn’t expect much advance warning from the economics profession. Let me be clear: Like Mr. Blanchard at the I.M.F., I am not predicting another Great Depression. We have indeed learned a lot over the last 80 years. But you should take that economic forecast, like all others, with more than a single grain of salt. N. Gregory Mankiw is a professor of economics at Harvard. He was an adviser to President Bush and advised Mitt Romney in his campaign for the Republican presidential nomination.

61 New York Magazine The Next New Deal The huge opportunities—and huge risks—of a possible Obama administration. By John Heilemann Published Oct 26, 2008

Evansville, Indiana. (Photo: Charles Ommanney/Reportage/Getty Images)

On a bright, brisk, fat-pumpkin morning in mid-October—the kind of morning you would call glorious were the economy not cratering, the financial system not imploding, the Dow not tumbling at this very moment to its lowest depths in more than five years—Barack Obama is on the courthouse steps in Chillicothe, Ohio, calmly and coolly enlisting the past in the service of claiming the future. “The American story has never been about things coming easy,” Obama declares. “It’s been about rising to the moment when the moment is hard … about rejecting panicked division for purposeful unity; about seeing a mountaintop from the deepest valley. That’s why we remember that some of the most famous words ever spoken by an American came from a president who took office in a time of turmoil: ‘The only thing we have to fear is fear itself.’ ” Obama had been toying with vague FDR allusions for the past three days, but now he’s decided to lay his cards on the table and seize the mantle explicitly. With the specter of a full- blown depression looming, the Age of Roosevelt—the campaign he ran in 1932, the

62 challenges he faced upon assuming office, the “bold, persistent experimentation” he called for and the New Deal edifice he erected in response—is much on the minds of the nominee and his inner circle. “A lot of people around Barack are reading books about FDR’s first hundred days,” says a member of Obama’s kitchen cabinet. “It’s a sign of the shift that’s going on emotionally: from being on this improbable mission to believing, Hey, we’re going to win.” Until recently, talk like that would have brought forth invocations of unhatched chickens from countless Democrats. From the moment it became clear last spring that Obama would be the party’s standard-bearer, the excitement over what he represented has been twinned with a gnawing dread that his astonishing ride would somehow come to a crashing end a few yards short of the White House. That America would prove unready to elect a black president. That the Republicans would once again work their voodoo on the electorate. Or that Obama would choke in the clutch—that, far from being the next FDR or JFK, he would turn out to be the reincarnation of George McGovern or Mike Dukakis or John Kerry. But as the outcome of the race has begun to seem more certain with each passing day—with Obama’s lead in the polls healthy and showing few signs of diminishing, with John McCain’s campaign listing aimlessly and lapsing into rank self-parody, with Sarah Palin devolving into a human punch line—Democrats are slowly, haltingly allowing themselves to believe that victory is truly within their grasp, and hence to contemplate what comes next. Transition. Inauguration. Those first hundred days. Maybe even, perchance, with augmented majorities for the party in both the House and Senate all but in the bag, the dawning of a spanking-new era of Democratic dominion. It requires no prodigious feat of memory, of course, to see how this dream could come a cropper. Back in 1993, Bill Clinton surfed into Washington on a similar wave of enthusiasm and expectation. Democrats then, too, controlled both the upper and lower chambers on Capitol Hill. The party’s agenda was bold, ambitious, far-reaching. And then everything fell to pieces. In something like a heartbeat, Clinton’s reputation as a Third Way centrist was reduced to rubble. The degree of Democratic political malpractice was so severe that it enabled the GOP, in 1994, to snatch the reins of the House and Senate simultaneously for the first time in four decades. This precedent would be unnerving enough for Democrats by itself, but the truth is that the circumstances Obama will confront are infinitely more daunting than those that Clinton faced at the outset of his administration. The recession that facilitated 42’s rise was shallow, and by the time he took office, it was already in the rearview mirror. And although the mounting deficit compelled Clinton to abandon much of the new spending he’d envisioned, the fiscal situation he inherited was nothing like the house of horrors awaiting Obama. Add to that the collapsing real-estate market, the credit crunch, a weak dollar, and rising unemployment, and Obama will find himself staring down the barrel of a downturn so steep and ugly that it could easily consume his whole first term. Oh, and did I forget to mention that the country is at war—in not one but two countries? All too aware that, should he win, these cascading crises will leave Obama with no time to gain his sea legs and terrifyingly little margin for error, he and his people, to a degree few realize, have been planning their transition from campaigning to governing for months with characteristic care and rigor. Like so much about Obama’s historic bid for the presidency, the first few days and weeks and months will be like nothing we have seen before—and all of it grounded in the insight that, mind-boggling as it might sound, winning was the easy part. These are Democrats they’ll be dealing with, after all.

Russia's CDS Spreads Spike: How Much Is Russia's Default Risk Rising? Oct 27, 2008 o The cost of insuring Russian bonds against bankruptcy (spreads of Credit default swaps on Russia's debt ) rocketed to 1,123 bps, higher than Iceland's debt before it sought a rescue from the International Monetary Fund (Telegraph) Although other emerging markets including Baltic countries, Turkey and CEE countries have higher external financing needs and run current account deficits, the reduction in Russia's reserves, the prospect of current account and fiscal deficits, the implicit and explicit state assumption of significant corporate debt as the oil price slumps point to further risk. o Russian sovereign bonds were already trading at higher risk than other EM sovereign bonds even before S&P lowered its credit ratings outlook to negative from stable citing the worsening outlook for public finance in light of lower oil prices and increasing government support of the banking sector o Danske :Russia is scheduled to make external debt repayment of USD 43bn in Q4 08 and USD 59bn is due in H1 09.In the midst of the global credit crisis, these large repayments are raising major concerns not only for foreign investors, but also for the Russian authorities o Russia still looks stronger than some of its peers despite vulnerabilities (Uralsib) - strengthening capital outflows as deposits are converted into dollars which may prompt restrictions on fx transactions. Several other commodity export-led economies could face similar worries, yet they have not been downgraded - the difference may be due to perception of higher political risk and government-led consolidation in Russia o The Russian banks’ dependence on wholesale and external funding is key to the economy’s vulnerability, which can be expected to mount if commodity prices fall further (Citi) Although the CBR's reserves far exceed Russia's short-term debt, they barely cover Russia's total external debt. Intervention to stabilize the rouble is depleting reserves rapidly (though the dollar rally also reduces the value of EUR and GBP holdings). Russia has to roll over $40b in debt this quarter and $150b in the next year. and the Government has promised over $200 billion in short and long-term capital to ease liquidity crunch and substitute for external finance o Russia may run a current account deficit with oil in the $60-70 a barrel range given the ramp up in imports and reduction in oil inflows - capital flows have already reversed and Russian investors flight to USD has pressured the rouble. o Weafer: An oil price in the $60’s/bbl should still mean a defendable ruble, but there will be a lot more pressure at $60/bbl, or lower. A combination of falling oil and a weakening currency will increase downward pressure on equity and bond markets. The correlation between the price of oil and the RTS was not very tight as oil was rising but it has increased on the downward trajectory. o Lacking a developed domestic bond market, the only way for oligarchs to raise money at present is by selling their equity, contributing to massive equity selloffs. Russia's unique

64 fragility is that over $1 trillion of debt needs to financed from a domestic capital pool of $600bn (Bond) o Now that Russia's own sovereign debt is in question, and many other emerging markets are turning to the IMF to stabilize their balance of payments, markets no longer believe Russia is strong enough to guarantee the estimated $530bn of foreign debts accumulated by its companies during the break-neck expansion of the oil boom. (Redeker, via telegraph) Russian Stocks Plunge In Global Rout, Trading Halted Until Oct.28 Oct 24, 2008 o Oct 24: Micex Index plunged 14% to 513.62 when the bourse suspended trading until Oct. 28. The dollar-based RTS Index declined 10% to 570.14 as many global equity markets suspended trading. The falling price of oil and specter of corporate defaults, rouble devaluation and a major decrease in earnings growth led to a renewed selloff. Russian equities have lost almost 70% since their may peak. The Yield on Russia's sovereign 30-yr bond rose above 12%, the highest since 2002 as S&P lowered its credit ratings outlook to negative o Oct 17: The ruble-denominated Micex Index sank 5.8% to 590.26 after earlier advancing as much as 6.4%. The dollar-denominated RTS Index fell 6.3% to 669.04. The RTS is headed for a 19% decline this week, its ninth consecutive week of losses. o Need for cash is contributing to equity liquidations. Fall in the price of oil, avoidance of risk leading to deleveraging from EM, and worries about political risk in Russia have contributed to major selloff in Russian equities which have lost more than 60% since May peak. Redemptions and margin calls from domestic investors contributed to the selloffs Policy Measures o Central bank cut reserve requirement to 0.5% Oct 15. o Capital injections have yet to forestall further declines in Russian equities trading has been suspended frequently o Russian wealth Fund will now invest as much as $6.7b in high rated Russian securities o Russia, has suffered some of the sharpest declines in global equity selloff, unwinding of carry trades etc. Russian market vulnerable to sharp corrections given short-term focus of most investors (Uralsib) Few mutual funds, pension funds. foreign investors held about half the free float (Bond) o Sep 30-Oct 3: the RTS had the biggest weekly drop since 1999 on record declines in commodity prices despite a promised $150b to the domestic banking system including $50b to replace non-renewed external loans o Sep 19: Russian equities gained as much as 25% intraday after they re-opened after being closed most of the week following large selloffs Sep 16-17, following a large bailout package including a promise of a $60b capital injection to the banking system and a cut in the oil export tax o Unicredit Foreign investors may have withdrawn over $63b from Russia from mid August to mid September. IPO stream (at home and abroad) has completely dried up o Merrill: Russian valuations have overshot the most relative to CDS widening this month

65 o Ponin: Three major government missteps contributed to the stock market woes: Putin's and Medvedev's decision not to interfere when government agencies applied pressure to BP in the TNK-BP affair; the harsh words by Putin against Mechel; and the escalation of the Cold War rhetoric following the conflict with Georgia. If the stock market's drop is connected with a global trend, pumping state funds into a sinking market means we would be using taxpayers' money to pay a nice bonus to scared investors who are eager to exit the Russian market and adding to the government's risks o Bond: Although the market is cheap and the domestic story is resilient, it will need a pretty significant catalyst (a rising oil price or a major package of oil tax reforms) to counter global risk aversion and investor shock- but the rising cost of money may create attractive bargains, among the domestic plays Oil Plunges Despite OPEC Production Cut Amid Global Stock Selloff Oct 26, 2008 o Oct 24: WTI crude futures fell as far as $63.05/b, Brent fell to close to $61/b after OPEC cut production by 1.5mbd. Commentators had expected approximately a 1mbd cut after OPEC had warned of a pending supply glut and all OPEC members seemed in favor. OPEC may be unable to stop the downward trend in prices given that global recession is reducing demand. At a minimum no increase in demand for oil as a financial asset is unlikely until financial markets stabilize (Ziemba) More cuts are likely at OPEC's December meeting. o Cut in OPEC's 28mbd overall production ceiling "quota" are split between OPEC members with Saudi Arabia scheduled for a cut of 466,000, Iran for 199,000 cut and Kuwait, UAE both cutting about 130tbd. It rejected Venezuela's suggestion to re-establish a target price range. This ceiling is lower than their current production o OPEC: The slowdown in oil demand, triggered by the financial crisis is serving to exacerbate the situation in a market which has been over-supplied with crude for some time, an observation which OPEC has been making since earlier this year. Forecasts indicate that the fall in demand will deepen, despite the approach of winter in the northern hemisphere. o Saudi Arabia, OPEC's largest (and only swing) producer is thought to be targeting has reversed much of its unilateral supply increase of earlier this year. o US oil and gasoline inventories remain on the build following the gulf coast hurricanes and demand for petroleum products has fallen around 5% on average this year from last year. Demand in other OECD countries is slipping and With China's growth slowing faster than expected, its demand might also slow o OPEC: Slowing economic growth and financial crisis will reduce demand for its oil by almost 900,000b/d next year from current levels (which in turn are lower than before)again cut petroleum demand outlook for next year suggesting that demand for its oil may fall to 31.14mbd in 2009. 2008 average demand for OPEC oil likely to be 32.01mbd, or slightly less than the demand in September, meaning that there is a surplus. o OPEC crude production averaged 32.157mbd in September, a drop of 308,600 barrels a day from August. Saudi output fell by 113,600b/d in September to 9.377mbd (via bloomberg) By October most of the implied cut of September has likely been implemented

66 o Oxford Analytica: the call on OPEC in 2009 will be flat, at best, requiring discipline to prevent lower prices. With Saudi Arabia accounting for most surplus capacity, this should not prove too difficult. Expected additional production from Angola and Ecuador may offset Iran and Venezuela's struggle to maintain exports. o Sep 9: OPEC maintained September 2007 production quotas - implying a removal of the new supply (780mbd) being pumped by Saudi Arabia but avoiding a formal cut OPEC suggested lower demand growth (especially from OECD countries), coupled with higher supply, a strengthening of the US dollar and easing of geopolitical tensions contributed to lower prices which are finally reflecting fundamentals. Given the market's contango structure there is potential for a larger crude build o Allsopp/Fattouh: OPEC's main objective is to defend oil prices from falling below some level deemed unacceptable. OPEC's pricing power varies over time. o Since June, OPEC has been pumping as much as 2mbd than a year ago o CIBC: soaring power demand, and a voracious appetite for motor vehicle fuel, boosted by subsidies for a range of fuel types—including fossil fuel-fired electricity, may reduce net exports of OPEC members by 1mbd/year until 2012 with desalination plants energy accounting for a third of that reduction

Welfare for Detroit Should lower-paid workers help subsidize those averaging $56,650 at GM? Monday, October 27, 2008; A12 AFTER YEARS of decline, U.S. auto companies face the double whammy of a credit crisis and a recession. Car and truck sales fell 26.6 percent in September, the first month since 1993 in which fewer than 1 million vehicles moved off the lots. General Motors, threatened with bankruptcy and burning through $1 billion in cash reserves per month, is groping for a merger with Chrysler. Ford's stock is down more than 70 percent in the past year, and investor Kirk Kerkorian is dumping his shares. The $25 billion federal loan approved by Congress on Sept. 25 may not reach Detroit for six to 18 months because of red tape. So Detroit's allies are pushing for waivers of the usual rules and, perhaps, another $25 billion before the end of the year. And why not? Everyone else seems to be getting a bailout these days. Hundreds of thousands of people depend on Detroit for their jobs, directly or indirectly. Well, we can think of several objections. First, there is the question of whether the U.S. government should be picking winners and losers in a business such as this. It's one thing to bail out the financial sector, whose product -- credit -- is essentially fungible and on which all other businesses depend. Automobiles, however, are not interchangeable, and Congress can't substitute its specific technological and aesthetic preferences for those of the market. What if we lend Detroit $25 billion and still nobody buys its cars? Second, this bailout taxes the less well-off to protect the relatively privileged. The average individual General Motors production worker, whose job would be saved by the bailout, makes $56,650 per year, according to the Center for Automotive Research, and that doesn't count better-paid, white-collar types. Meanwhile, half of all households-- which typically include more than one earner -- make less than $50,000 per year. Where's the justice in that? Congress approved $7,500 tax credits for purchasers of GM's much-touted plug-in hybrid Chevy Volt, built to run 40 miles on a single electric charge. That would knock the net cost of the four-seat Volt, due out in late 2010, down to $32,500 -- not much less than a basic Cadillac CTS costs now. Even then, it could take a decade of Volt driving to recoup the difference in purchase prices between it and the far cheaper Toyota Prius. Assuming a few well-heeled drivers take that deal, why should poorer people be taxed to enable them? The downfall of the American auto industry is indeed a tragedy. But the automakers and the United Auto Workers have only themselves to blame for much of it. For years, they pursued protectionism against foreign competitors rather than tackle them head-on. The automakers say that they need $25 billion from Congress to offset the additional costs of tough new fuel- efficiency standards. Perhaps they wouldn't be in that situation if they had accepted such standards a long time ago and retooled to meet them, rather than persisting in the more familiar, and profitable, business of making gas guzzlers. We would all have been better off if the federal government had enacted a higher gas tax so that the Big Three could have planned production on that basis. A stiffer gas tax, rebatable in some form to consumers, would still be the best way to guarantee a long-term shift to more economical cars. Alas, there's a limit to how much taxpayers can spend ensuring that such cars get built in Detroit.

68 72 F O R B E S NOVEMBER 10, 2008 INTELLIGENT INVESTING

SHELTER FROM THE STORM STEVE H. HANKE There’s plenty of blame to go around, but the main culprit is the Fed. POINT OF VIEW | STEVE H. HANKE IN MY APR. 16, 2007 COLUMN I WARNED THAT THE U.S. was trapped in a dangerous boom-bust cycle that began with cheap credit and would end with collapsing home and stock prices. When the bust came it was worse than I had imagined. Now what? To find a safe harbor in this storm, we must ignore panicked media headlines and understand how we got into such turbulent waters. There is plenty of blame to go around, but the main culprit is the Federal Reserve. In late 2002 Ben S. Bernanke, then a Fed governor and now the chairman, persuaded Alan Greenspan, then chairman, that the U.S. was in the grip of deflation. In consequence, the Fed pushed down on the monetary accelerator. By July 2003 the Fed funds rate had been squeezed down to 1%, where it stayed for a year. This artificially low interest rate set off the mother of all liquidity cycles. The Fed’s laxity stimulated the economy and pushed final sales to an unsustainable growth rate of 7% in nominal terms, i.e., before inflation adjustment. (Final sales, defined as gross domestic product plus net imports minus inventory buildup, is a measure of goods and ervices absorbed in the U.S.) At the same time, the lax monetary policy encouraged investors to take undue risks chasing high yields. To make the most of tiny yields, leverage became the flavor of the day. Carry trades—borrowing in low-yield foreign currencies and investing in higher-yield ones— also became popular. Borrow, borrow, borrow. I watched this top-heavy structure go up with amazement and terror. It had to crumble. It did. The Fed’s policy blunder also weakened the dollar and thus stimulated commodity price inflation. When the dollar goes down relative to other currencies, the price of wheat, corn, rice and oil all go up in dollar terms. The currency began its downward course early in 2002, thanks in large part to Fed policy, and it bottomed out in mid-July 2008, having declined 44% against the euro in that period. At the same time, the price of a barrel of oil climbed sevenfold, from $20 to $146. About half of that climb was a function of rising demand (such as from India and China); the remainder can be laid to the weak dollar. The same happened with other internationally traded commodities such as rice and soybeans. Congress played its part, too. In 2003 and 2004 Fannie Mae and Freddie Mac, the government-sponsored mortgage buyers, were engulfed in accounting scandals. To get Congress off their backs, they became more committed to financing homes for families with low incomes. The ploy worked like a charm. Congressman Barney Frank, who now chairs the House Financial Services Committee, turned a blind eye to their accounting shenanigans and praised their newfound zeal. That was typical. Fannie and Freddie became the largest purchasers of subprime and borderline (Alt-A) mortgages in the 2004–07 period with a total exposure of $1 trillion and thereby contributed mightily to the housing bubble as well as their own later collapse.

69 The Bush Administration wasn’t sitting on its hands while the Fed and Congress were giving birth to a boom-bust disaster. It was making matters worse by spending recklessly and piling up debt. Just weeks before the giant bailout was signed into law, the Congressional Budget Office issued a report stating that the Administration’s spending had put the economy “on an unsustainable path.” It projected that over the next decade the government debt held by the public would grow from $5.4 trillion to $7.9 trillion in today’s dollars, a 46% increase. That didn’t include the unfunded liabilities of the Social Security and Medicare systems, which total $100 trillion in present value. As for the $700 billion bailout, to pass the bill the Administration and Congress had to scare the public to death and promise to lead the alarmed populace to safety—after the government’s multiple failures had created the crisis. At least the scare tactics did their job. But will the bailout work as advertised? Plan A called for the government to buy shaky mortgage paper from banks. Eventually it dawned on Henry Paulson that there was no price he could pay that would both help the banks and protect the Treasury. As this column went to press, the government was veering off in a different direction—using taxpayer capital to buy preferred stock from the banks. Expect more irresponsible political behavior and market panic. We’ll see deleveraging- driven deflation in the near term and more inflation in the long term. Maintain the gold hedges I have recommended before (most recently in my Aug. 11 column) and keep a full stockpile of U.S. Treasury inflationprotected securities. Steve H. Hanke is a professor of applied economics at the Johns Hopkins University and a senior fellow at the Cato Institute in Washington, D.C. Visit his home page at www.forbes.com/hanke.

New world pragmatism Barry Eichengreen There will be no new financial world order on the scale of Bretton Woods in 1944. Here are some modest but important steps that leaders can take

Barry Eichengreen Friday October 24 2008 20.00 BST Before we get carried away with the idea of a new Bretton Woods conference to remake the global economy, it is worth recalling four facts about what made the original gathering a success. First, the original agreement was a response to the shock of the Great Depression and the second world war. The current shock is severe, to be sure, but not that severe. Even the most pessimistic observers expect "only" the deepest recession since the early 1980s, when unemployment in the United States reached 10%, not a depression like that which raised US unemployment to 24% in 1933. And while economic crisis may yet fan geopolitical tensions, no one anticipates repercussions tantamount to the second world war. This crisis has created a willingness to contemplate significant reform, but it is unlikely to support reforms as radical as those reached in 1944. Second, the conference held in Bretton Woods, New Hampshire in that year took place after three years of extensive planning under the intellectual leadership of Harry Dexter White and John Maynard Keynes in the US and British treasuries. This time, in contrast, treasuries on both sides of the Atlantic have been behind the curve. Advance planning, such as it is, has anticipated events by at most a matter of days. Third, the Bretton Woods conference was a meeting of finance and treasury officials, not heads of state. Heads of state are prone to grand statements, not detailed proposals for economic and financial reform – Gordon Brown being a rare exception. For substance, as opposed to posturing, we will have to wait for the follow-up conferences attended by specialists. Finally, the conference took place at a time of unquestioned US hegemony over the western alliance and the global economy. America had the intellectual and financial resources with which to drive the reform process. Now it lacks both. In Europe, France and Germany are squabbling over the form and extent of state intervention in the post-crisis world. In Asia, China and Japan are vying uncooperatively for leadership. Beijing responded favourably to Korea's proposal for a regional bail-out fund, but Tokyo deferred, fearing that this would be dominated by China, given that country's immense dollar reserves. Tokyo then proposed funneling Asian reserves through the IMF, but China deferred, fearing that this initiative would be dominated by Japan, which has long participated in IMF deliberations. Financial diplomacy is evidently more difficult than in 1944.

71 So what to do? Countries participating in the series of summits starting on November 15 should concentrate on stabilising financial markets, which is the immediate problem to be solved. There are other pressing global problems, from climate change to poverty and underdevelopment, but adding them to this agenda will only make deliberations wordier and less productive. It is far from clear, for example, that the US will agree to a new international agreement if it has to compromise not just on financial regulation but on global warming, foreign aid, and sundry other issues. Next, move quickly from the leaders' meeting, which is largely about the photo-ops, to the meeting of finance ministers, where the real business will occur. And once there, focus on pragmatic reforms. Clamp down on regulatory arbitrage. Raise capital requirements. Make the regulatory regime less procyclical. Use taxes and regulation to drive transactions in credit default swaps and other derivative instruments into an organised exchange. This modest approach will not be hailed as a New World Financial Order. But it will be a useful first step toward making the world a safer financial place. And it will minimise the danger that the new Bretton Woods conference will go down in history as a failure.

Doing Business in Spain

Written by Jaime Pozuelo-Monfort

Sunday, 19 October 2008

The World Bank conducts a survey on Doing Business in the world every year. The survey reviews what economic environments are more prone to conducting business and ranks them according to a handful of criteria. The 2009 rankings include 181 nations. Spain worsened its standings from 46 in 2008 to 49 in 2009. The first five spots from best to worse are held by Singapore, New Zealand, the United States, Hong Kong and Denmark. The bottom five spots are held from best to worse by Burundi, Republic of Congo, Guinea-Bissau, Central African Republic and Democratic Repubic of Congo. A summary of the 2009 results was recently released by the World Bank. The conclusions reviewed hereafter correspond to the 2008 results. In a previous article on 5spaniards I related Transparency and Corruption to Foreign Direct Investment. The same conclusion can apply to the business environment. Spain’s standings in the international rankings are mediocre and point out the direction that reform should adopt going forward. In particular the World Bank provides the following data related to the requiremens entrepreneurs face in Spain when opening a business:

STARTING A BUSINESS - Takes 10 procedures - Takes 47 days - Costs 15.1% of the per-capita income - Requires 13.7% of the per-capita income as the minimum capital required

DEALING WITH LICENSES - Takes 11 procedures - Takes 233 days - Costs 64.9% of the per-capita income

EMPLOYING WORKERS - Difficult of hiring index (0-100): 78

73 - Rigidity of hours index (0-100): 60 - Difficulty of firing index (0-100): 30 - Rigidity of employment index (0-100): 33 - Firing cost (weeks of salary): 56

REGISTERING PROPERTY - Porcedures: 4 - Time (days): 18 - Cost (% o property value): 7.1

GETTING CREDIT - Strength of legal rights index (0-10): 6 - Depth of credit information index (0-6): 6 - Public registry coverage (% of adults): 44.9 - Private bureau coverage (% of adults): 8.3

PROTECTING INVESTORS - Extent of disclosure index (0.1-): 5 - Extent of director liability index (0-10): 6 - Ease of shareholder suits index (0-10): 4 - Strength of investor protection index (0.10): 5.0

PAYING TAXES - Payments (numer per year): 8 - Time (hours per year): 298 - Total tax rate (% profit): 62.0

TRADING ACROSS BORDERS - Documents to export: 6 - Time to export (days): 9 - Cost to export (US$ per container): 1,000 - Documents to import (number): 8 - Time to import (days): 8 - Cost to import (US$ per container): 1,336

ENFORCING CONTRACTS - Procedures (number): 39 - Time (days) 525 - Ost (% of claim): 17.2

CLOSING A BUSINESS - Time (years): 1.0 - Cost (% of estate): 15 - Recovery rate (cents on the dollar): 76.9

A second article on Doing Business in Spain will review the conclusions of the 2008 report and provide useful commentary on the conclusions drawn by the World Bank.

Foreclosure prevention efforts and market stability Governor Elizabeth A. Duke Before the Committee on Banking, Housing, and Urban Affairs, U.S. Senate October 23, 2008 Chairman Dodd, Senator Shelby, and other members of the Committee, I appreciate this opportunity to discuss recent actions taken to stabilize financial markets and foreclosure prevention efforts. Financial markets have been strained for more than a year, as house prices declined, economic activity slowed, and investors pulled back from risk-taking. These strains intensified in recent weeks. Lending to banks and other financial institutions beyond a few days virtually shut down. Withdrawals from money market mutual funds and prospects that net asset values would fall further severely disrupted commercial paper and other short-term funding markets. Longer-term credit also became much more costly as credit spreads for bonds jumped and interest rates rose. The problems in credit markets and increasing concerns about the state of the economy caused equity prices to swing sharply and decline notably. Policymakers here and in other countries have taken a series of extraordinary actions in recent weeks to restore market functioning and improve investor confidence, with the aim ultimately to increase the availability of credit and the value of savings. The Federal Reserve has continued to address ongoing problems in interbank funding markets by expanding its existing lending facilities, and recently increased the quantity of term funds it auctions to banks and accommodated greater demand for funds from banks and primary dealers. We also increased our currency swap lines with foreign central banks. To alleviate pressures on money market mutual funds and commercial paper issuers, we implemented several important temporary facilities, including one to provide financing to banks to purchase high-quality asset-backed commercial paper from money funds, and another to provide a backstop to commercial paper markets by purchasing highly rated commercial paper directly from businesses at a term of three months. On Tuesday of this week, we announced another program in which we will provide senior secured financing to conduits that purchase certain highly rated commercial paper and certificates of deposit from money market mutual funds. The financial rescue package recently enacted by Congress, the Emergency Economic Stabilization Act (EESA), provides critically important new tools to address financial market problems. EESA authorized the Troubled Asset Relief Program (TARP), which allows the Treasury to buy troubled assets, to provide guarantees, and to inject capital to strengthen the balance sheets of financial institutions. As provided in the Act, the Federal Reserve Board and its staff are consulting with the Treasury regarding the TARP. In addition, Chairman Bernanke serves as the Chairman of the oversight board for TARP that will, among other things, review the policies that are implemented and make recommendations, as appropriate, regarding the use of authorities under TARP. EESA also temporarily raised the limit on the deposit insurance coverage provided by the Federal Deposit Insurance Corporation (FDIC) from $100,000 to $250,000 per account. Last week, the first use of TARP funds was announced. In particular, the Treasury announced a voluntary capital purchase program, and nine of the nation’s largest financial institutions have agreed to participate. The program is open to financial institutions of all sizes. Under the program, the Treasury would acquire capital of financial institutions on terms that are attractive to the institutions and with features that protect the taxpayer. At the same time, the Federal Reserve Board, the FDIC, and the Secretary of the Treasury in consultation with the President, determined that there were significant risks to the stability of the financial system. With this determination, the FDIC used its authority to expand for a specified period, insurance to non-interest-bearing transactions accounts, such as payroll accounts, and a guarantee for newly issued

75 senior unsecured debt of FDIC-insured depository institutions, including their associated holding companies. A second, complementary, use of TARP funds will be to purchase mortgage assets, including mortgage- backed securities and whole loans. These purchases are designed to remove uncertainty from lenders’ balance sheets and to restore confidence in their viability. Another objective is to improve the modification efforts of servicers on these loans to more effectively prevent avoidable foreclosures. The Federal Reserve System is also working to develop solutions to rising foreclosures. Preventing avoidable foreclosures is good for borrowers, communities, and the economy. A number of efforts are underway. The Federal Reserve has worked with other agencies to put in place the standards and procedures for the new Hope for Homeowners (H4H) program, and I serve on the Oversight Board. These loans can help borrowers who might otherwise face foreclosure because the new loan payments are more affordable and the homeowners get some equity in their homes. Lenders and servicers are analyzing their borrowers for good candidates for the H4H program, and the FHA and its authorized lenders are poised to process applications. We appreciate the additional flexibility provided to the program by Congress in EESA, in particular allowing up-front payments to junior lien holders that agree to release their claims. For some time, we have called upon lenders, investors, and servicers to aggressively pursue sustainable loss mitigation activities. For example, last year the Federal Reserve and the other banking agencies issued supervisory guidance to encourage mortgage lenders and servicers to pursue prudent loan workouts. We continue to support industry-led efforts, especially those of HOPE NOW, in pursuing flexible approaches to stem the rise in foreclosures and to deal with their effects. Earlier this year, we embarked on a joint effort with NeighborWorks America on neighborhood stabilization to help communities develop strategies for addressing increases in foreclosures and vacant properties. The Federal Reserve System is strategically utilizing its presence around the country through its regional Federal Reserve Banks and their branches to address foreclosures. Our history of working closely at the local level with communities enables us to tailor activities to the specific needs of that area. Our efforts have taken a variety of forms. We have employed economic research and analysis to target scarce resources to the communities most in need of assistance. We have provided community leaders with detailed analyses identifying neighborhoods at high risk of foreclosures. This information is helping local groups to better focus their borrower outreach and counseling efforts. In addition, we have sponsored or supported a wide range of activities in local communities. For example, the Federal Reserve System has sponsored a series of “Recovery, Renewal, Rebuilding” forums in cities around the country in which key experts discussed the challenges related to real-estate owned inventories and vacant properties in strong and weak housing markets, and explored effective neighborhood stabilization policies. Four events were held in various parts of the country earlier this year, and the series concluded with a fifth meeting of experts this past Monday in Washington. This series is just an example of many events. All told, the Federal Reserve System has sponsored or co-sponsored more than 80 events related to foreclosures since last summer, reaching more than 6,000 attendees including lenders, counselors, community development specialists, and policymakers. We also have supported events that bring together borrowers with counselors, lenders, and servicers. In August, the Federal Reserve Bank of Boston partnered with the HOPE NOW Alliance, NeighborWorks America, the Kraft family, and the New England Patriots Charitable Foundation, among many others, and held an event at Gillette Stadium. More than 2,100 borrowers seeking help attended. Twenty servicers and twenty non-profit counseling agencies took part, with staff at the event and on dedicated phone lines. Although we do not yet know the outcomes for these homeowners, we are monitoring the results and other Federal Reserve Banks are considering holding similar events. In conclusion, the Federal Reserve has taken a range of actions to stabilize financial markets and to help borrowers and communities. Taken together, these measures should help rebuild confidence in the financial system, increase the liquidity of financial markets, and improve the ability of financial institutions to raise capital from private sources. Efforts to stem avoidable foreclosure and help borrowers through H4H and more aggressive modifications, as well as to develop effective strategies for dealing with foreclosed-upon properties, I believe, will also help homeowners and communities. These steps are important to help stabilize our financial institutions and the housing market, and will facilitate a return to more-normal functioning and extension of credit.

Over-the-counter derivatives Patrick M. Parkinson, Deputy Director, Division of Research and Statistics Before the Subcommittee on Securities, Insurance, and Investment, Committee on Banking, Housing, and Urban Affairs, United States Senate July 9, 2008 Chairman Reed, Ranking Member Allard, and members of the Subcommittee, I am pleased to appear today to discuss the over-the-counter (OTC) credit derivatives market. First, I will provide some information on credit derivatives, the markets in which those instruments are traded, the risks that their use entails, and some key practices for managing those risks. Then I will discuss the oversight of the credit derivatives markets by the prudential supervisors of the firms that act as dealers in credit derivatives, including joint efforts by supervisors and market participants to strengthen the infrastructure of those markets. Finally, I will discuss the potential benefits of central counterparty (CCP) clearing as well as those of exchange trading of credit derivatives. Although the focus of this hearing is on credit derivatives, most of my remarks are applicable to OTC derivatives generally. The OTC Credit Derivatives Market Background Information A credit derivative is a financial contract whose value is derived from the value of debt obligations issued by one or more reference entities. The predominant type of credit derivative is a credit default swap (CDS). In a CDS, a "protection buyer" pays premiums to a "protection seller." In return, in the event of a default or other specified credit event, the protection seller is obligated to pay the protection buyer the notional or par value for the debt, thereby transferring the risk of default from the buyer to the seller. Most reference entities are corporations, including corporations rated investment-grade and those with lower ratings. Over the last few years, CDS referencing mortgage-backed securities and other asset-backed securities (CDS on ABS) also have been traded. A single-name CDS references a single corporation or ABS, while a multiname CDS references a basket of reference entities or, more commonly, an index composed of many single-name CDS. Markets in Which Credit Derivatives Are Traded Although credit derivatives have been listed on exchanges, to date the vast majority of credit derivatives have been executed bilaterally with derivatives dealers in OTC markets. The dealers include 15 to 20 large, globally active commercial and investment banks. The principal centers for trading are London and New York. Trades typically are executed over the telephone or through voice brokers. Use of various electronic trading platforms to facilitate bilateral execution of CDS has been growing, especially in Europe, but remains fairly limited. More than half of trading in CDS is trading between dealers. Other than dealers, the most active participants in CDS markets are asset managers, including both hedge fund managers and managers of regulated investment companies. Estimates of the size of the global market for CDS indicate that the market has been growing very rapidly. Global market estimates published by the Bank for International Settlements show that the notional amount outstanding at year-end 2007 was $58 trillion, about twice the level just a year earlier. The gross replacement cost of those contracts, which measures the current market value of the protection against credit events affecting the $58 trillion of debt, was about $2 trillion at year-end. Growth of index and other multiname CDS has been especially rapid in recent years and those instruments now account for more than 40 percent of both the notional amount and the current market value of all CDS. The very rapid growth of the credit derivatives market reflects their perceived value for managing credit risks. The single-name CDS markets typically are far more liquid than the underlying bond or

77 loan markets, in large measure because the cost of taking short positions is far lower. Fixed-income asset managers use credit derivatives to obtain or adjust their credit exposures. Portfolio managers at banks use single-name CDS to manage concentrations of risk to their largest borrowers. Furthermore, the very liquid markets for CDS indexes allow asset managers to adjust the risk profile of their entire debt portfolios much more quickly and at much lower cost than was possible before these instruments were available. The availability of CDS also facilitates underwriting and making markets in the underlying debt markets, and thereby benefits issuers and investors that do not directly use credit derivatives. Risks of Using Credit Derivatives The use of credit derivatives entails risks as well as benefits. The types of risk are essentially the same as those associated with financial activity generally--market risk, credit risk, operational risk, legal risk, and reputational risk. Of particular importance is counterparty credit risk--that is, the risk that a counterparty to a credit derivatives contract could fail to perform its contractual obligations, resulting in losses to the nondefaulting counterparty. For example, in the case of a CDS, if the protection seller itself becomes insolvent, the protection buyer would lose the value of that protection and would need to replace it by purchasing protection from another seller. If the premiums required by the market for protection against default by the reference entity had risen since the protection had been purchased from the insolvent seller, the protection buyer would be exposed to a loss equal to the present value of the difference between the premiums paid on the new contract and the premiums paid on the original contract. Key Practices for Managing Risks Participants in the credit derivatives market and other OTC derivatives markets manage their counterparty credit risks by carefully selecting and monitoring their counterparties, by documenting their transactions under standard legal agreements that permit them to net gains and losses across contracts with a defaulting counterparty, and by entering into agreements that require counterparty exposures to be collateralized. Market participants effectively preclude firms from acting as dealers if they are not rated A or higher. Dealers evaluate the credit worthiness of their counterparties and assign them internal credit ratings. Those whose internal ratings are equivalent to below investment grade usually are required to enter into collateral agreements that include initial margin requirements as well as variation margin requirements. Transactions with hedge funds typically are supported by collateral agreements, as are transactions between dealers. Laws in the United States and many other jurisdictions have been amended in recent years to clarify that netting and collateral agreements are legally enforceable. Still, the measurement and management of counterparty credit risks on credit derivatives are challenging. Furthermore, as I will focus on today, weaknesses in the infrastructure for the credit derivatives markets and other OTC derivatives markets have created operational risks that could undermine the effectiveness of counterparty risk-management practices. Oversight of the OTC Credit Derivatives Market Although the credit derivatives market often is described as unregulated, by its nature it is subject to significant regulatory oversight. All transactions in the market are intermediated by dealers and all major dealers are commercial or investment banks that are subject to prudential regulation by U.S. or foreign banking regulators or by the Securities and Exchange Commission (SEC). The prudential supervisors devote considerable attention to the dealers' management of the risks associated with activities in the credit derivatives market and other OTC derivatives markets. In particular, they have been issuing guidance on counterparty credit risk management since the mid-1990s and have updated it several times, notably after the near failure in 1998 of Long-Term Capital Management, which was a major participant in the interest rate derivatives market. With the rapid growth of the credit derivatives market and other derivatives markets and the increasing participation of hedge funds in those markets, the management of counterparty exposures to hedge funds has been given careful attention, including a thorough review of relevant risk-management practices by the President's Working Group on Financial Markets (PWG) in 2006. That review fed into the Principles and Guidelines Regarding Private Pools of Capital that the PWG issued in July 2007, which provided updated guidance on the management of such counterparty exposures.

78 The volatility and illiquidity in financial markets over the past year have provided a severe test of major dealers' counterparty risk-management practices. Thus far, the results with respect to hedge fund exposures have been remarkably good. Although quite a few hedge funds have performed very poorly, counterparty credit losses to their dealer counterparties have been negligible. By contrast, the financial difficulties of some monoline financial guarantors have forced some of the firms that act as dealers to write down substantially the value of credit protection that the dealers had purchased from the guarantors on collateralized debt obligations and other structured credit products. Because the guarantors had been considered highly creditworthy and because the exposures against which they sold protection were considered to pose very little credit risk, their CDS counterparties had generally not required the guarantors to enter into collateral agreements. In light of this experience, the Financial Stability Forum's (FSF) April 2008 report to the G-7 Ministers and Central Bank Governors called on prudential supervisors to extend guidance on management of counterparty exposures to hedge funds to other large, highly leveraged counterparties, including other dealers and financial guarantors. Supervisory Efforts to Strengthen the Infrastructure of the OTC Credit Derivatives Market In addition to their efforts to ensure that individual derivatives dealers manage the risks associated with credit derivatives and other OTC derivatives effectively, prudential supervisors, under the leadership of the Federal Reserve Bank of New York (FRBNY), have been working with dealers and other market participants since September 2005 to strengthen arrangements for clearing and settling OTC derivatives transactions. For too many years, post-trade processing of OTC derivatives transactions remained decentralized and paper-based despite enormous growth in transactions volumes. Among other problems, dealers reported large backlogs of unconfirmed trades, a significant portion of which had been outstanding for 30 days or more. The failure to confirm trades promptly can exacerbate counterparty credit risks by allowing errors in counterparties' records of their transactions to go undetected, which could lead them to underestimate exposures or to fail to collect margin when due. Such backlogs also could significantly complicate and delay the close-out and replacement of trades with a defaulting counterparty. By 2005, backlogs of unconfirmed trades were especially large in the credit derivatives market, in part because market participants, including hedge funds, frequently closed out their positions in CDS through a transaction known as a novation. In a novation, one party steps out of the contract and is replaced by another party. The master agreements that govern OTC derivatives trading require the party seeking to step out to obtain the prior written consent of its counterparty, but dealers were frequently accepting novations from market participants without any evidence that they had obtained such prior consent. These sloppy practices not only contributed to backlogs of unconfirmed CDS, but also created confusion about the identities of trade counterparties and thereby undermined the effectiveness of counterparty credit risk management. With encouragement and close monitoring by their prudential supervisors, the dealers worked with market participants to address these weaknesses. By making greater use of available platforms for electronic confirmation of CDS trades, they quickly reduced the backlogs. By September 2006, the dealers reported that, in the aggregate, they had reduced confirmations outstanding more than 30 days by 85 percent. In 2006, the dealers agreed to expand their efforts to tackle backlogs in the equity derivatives market, again by making greater use of electronic confirmation services. Dealers also quickly announced their support for a novation protocol for credit and interest rate derivatives that had been developed by the International Swaps and Derivatives Association. The protocol provides that if the party initiating the novation has not received written confirmation from the original counterparty by the close of business on the date the novation is struck, it is deemed to have two contracts, one with the original counterparty and another with the counterparty that agreed to accept the novation. The protocol thereby provides the party initiating the novation a strong incentive to obtain the original counterparty's consent promptly. Although these achievements were impressive, the financial turmoil during the summer of 2007 convinced prudential supervisors and other policymakers that further improvements in the market infrastructure were needed. Specifically, CDS backlogs grew almost fivefold from June to August

79 2007, reversing much of the previous improvement. Although the backlogs subsequently receded, this episode demonstrated that backlog reductions were not sustainable during volume spikes. Moreover, it underscored that, in many respects, the post-trade processing performance of the OTC derivatives markets still lags significantly the performance of more mature markets and still has the potential to compromise market participants' management of counterparty credit risks and other risks. In their reports on the financial market turmoil, both the PWG and the FSF asked prudential supervisors, under the leadership of the FRBNY, to take further actions to strengthen the OTC derivatives market infrastructure. Specifically, they asked the supervisors to insist that the industry set ambitious standards for trade data submission and resolution of trade-matching errors. More timely and accurate submission of trade data is critical to avoiding the buildup of backlogs following volume spikes. They also asked supervisors to ensure that the industry promptly incorporates into standard CDS documentation a protocol that would permit cash settlement of obligations following a default or other credit event involving a reference entity, based on the results of an auction. Adoption of the cash settlement protocol is intended to address concerns that a physical settlement process for CDS could be disorderly in the event of large-scale or multiple contemporaneous defaults. Finally, the PWG and FSF also recommended that the supervisors ask the industry to develop a longer-term plan for an integrated operational infrastructure for OTC derivatives that covers all major asset classes and product types and addresses the needs of other market participants as well as dealers. The FRBNY convened a meeting of supervisors and market participants on June 9 to discuss how to address the PWG and FSF recommendations. They agreed on an agenda for bringing about further improvements in the OTC derivatives market infrastructure. With respect to credit derivatives, this agenda includes: (1) further increasing standardization and automation, with the ultimate objective of matching trades on the date of execution; (2) incorporating an auction-based cash settlement mechanism into standard documentation; (3) reducing the volume of outstanding CDS contracts via greater use of services that orchestrate multilateral terminations; and (4) developing well-designed central counterparty services to reduce systemic risks. They also agreed to extend the infrastructure improvements in the credit derivatives market over time to encompass the OTC equity, interest rate, foreign exchange, and commodity derivatives markets. Potential Benefits of Greater Centralization of Market Infrastructure

Central Counterparty Clearing of Credit Derivatives A central counterparty is an entity that offers to interpose itself between counterparties to financial contracts, becoming the buyer to the seller and the seller to the buyer. Trades on derivatives exchanges routinely are cleared through a CCP, in part so that market participants can accept the best bids or offers without considering the creditworthiness of the party making the bid or offer. Indeed, in electronic exchanges, the use of a CCP permits anonymous trading. CCP services also have been offered to counterparties in OTC derivatives markets. For example, since September 1999, LCH.Clearnet Limited has operated SwapClear, a London-based CCP for interest rate swaps between dealers. SwapClear clears almost 50 percent of global single-currency swaps between dealers. Several plans are now under development to provide CCP services to the credit derivatives market. A CCP has the potential to reduce counterparty risks to OTC derivatives market participants and risks to the financial system by achieving multilateral netting of trades and by imposing more-robust risk controls on market participants. However, a CCP concentrates risks and responsibility for risk management in the CCP. Consequently, the effectiveness of a CCP's risk controls and the adequacy of its financial resources are critical. If its controls are weak or it lacks adequate financial resources, introduction of its services to the credit derivatives market could actually increase systemic risk. A CCP that seeks to offer its services in the United States would need to obtain regulatory approval. The Commodity Futures Modernization Act of 2000 included provisions that permit CCP clearing of OTC derivatives and require that a CCP be supervised by an appropriate authority, such as a federal banking agency, the Commodity Futures Trading Commission, the SEC, or a foreign financial regulator that one of the U.S. authorities has determined to satisfy appropriate standards. A CCP for

80 credit derivatives with standardized terms that was not regulated by the SEC might need an exemption from securities clearing agency registration requirements. If a CCP for credit derivatives sought to organize as a bank subject to regulation by the Federal Reserve or if we were consulted by any other regulator of a proposed CCP, we would evaluate the proposal against the Recommendations for Central Counterparties, a set of international standards that were agreed to in November 2004 by the Committee on Payment and Settlement Systems (CPSS) of the central banks of the Group of Ten countries and the Technical Committee of the International Organization of Securities Commissions (IOSCO). If one or more CCPs for credit derivatives that meet the CPSS-IOSCO standards are introduced, the Federal Reserve will encourage market participants to use those services to the fullest extent possible. We would also encourage such CCPs to clear trades for a broad range of market participants, either directly or through intermediaries. Market participants should be excluded from participating only if doing so would entail risks to the CCP that it cannot mitigate effectively. Exchange Trading of Credit Derivatives An exchange is a mechanism for executing trades that allows multiple parties to accept bids or offers from other participants. As I have already stated, trades on an exchange usually are intermediated by a CCP. Exchange trading requires a significant degree of standardization of contracts. In many cases, counterparties to OTC derivatives trades seek to customize the terms of trades to meet very specific risk-management needs, so many OTC trades are not amenable to exchange trading. However, many OTC derivatives, including many credit derivatives have become sufficiently standardized that exchange trading is feasible and the scope for exchange trading probably could be expanded by further standardization of contracts while still meeting risk-management needs. Where exchange trading of OTC credit derivatives is feasible, it can produce several benefits. First, trades executed on an exchange usually are intermediated by a CCP and, as I have discussed, a well- designed CCP can reduce risks to counterparties and the financial system. Second, an electronic exchange can be designed so that trades are locked in at execution, essentially achieving trade matching in real time and eliminating confirmation backlogs. Third, exchange trading has the potential to increase market liquidity by allowing participants to directly trade against bids and offers posted by a broader range of parties, including asset managers as well as derivatives dealers. Fourth, exchange trading has the potential to significantly increase transparency with respect to bids and offers and the depth of markets at those bids and offers. For these reasons, policymakers should encourage greater standardization of contracts, which would facilitate more trading on exchanges. However, they should not lose sight of the fact that one of the main reasons the credit derivatives market and other OTC markets have grown so rapidly is that market participants have seen substantial benefit to customizing contract terms to meet their individual risk-management needs. They must continue to be allowed to bilaterally negotiate customized contracts where they see benefits to doing so. Conclusions The credit derivatives market is an important innovation that provides significant benefits to the banks and asset managers that use these instruments and to the financial system generally. However, their use entails risks, including counterparty credit risks, that market participants need to manage effectively. Supervisors need to continue to pay close attention to individual dealers' management of the risks associated with intermediating the credit derivatives market and other derivatives markets. They also need to continue to foster collective actions by dealers and other market participants to move rapidly toward the goal of implementing a clearing and settlement infrastructure for the credit derivatives market and other OTC derivatives markets that is as efficient as the infrastructure for more mature markets. Supervisors and other policymakers should encourage the introduction and use of well-designed CCP clearing services for credit derivatives and should encourage greater standardization of contracts, which would facilitate more trading on exchanges.

81 Financial Times ft.com/alphaville OTC to become UTC, taking ICAP and TP with’em? Posted by Paul Murphy on Oct 28 16:27. Comment. Blindingly obvious, really - which is not to say the market has quite factored in the new reality. From Chris Turner at Goldman Sachs on Tuesday: Counterparty risks to the fore We expect the turmoil among global banks to drive a change in market structure that favours derivative exchanges over the current OTC arrangements. In our view, heightened awareness of counterparty risk is likely to lead to greater clearing of derivative products. The simple argument here is that the interdealer broking community - for so many years the hard-nut core of London’s financial markets - are going to suffer mightily as transparency and centralised counterparty/settlement obligations are imposed on the over-the-counter derivatives markets. Any under-the-counter future, perhaps. More from Turner: Downgrade ICAP and Tullett Prebon to Sell We downgrade both ICAP and Tullett Prebon to Sell and add ICAP to the Conviction List. We reduce EPS forecasts for both stocks by 35%- 50% on average for the next two years, leaving our estimates 35%-50% below consensus. We also reduce our EPS forecasts for the European exchanges by 18%-26% to reflect a more challenging volume environment. The flip side is that established exchanges like the London Stock Exchange Deutsche Borse should benefit. The Goldman analyst reckons the German exchange might capture €300m of European CDS revenues alone, with another €500m by way of interest rate swaps. Unless, of course, Euronext Liffe hoovers up all the business… Is this the real reason for today’s equity market gains? Posted by Neil Hume on Oct 28 13:18. From an email currently circulating the City of London: Gut wrenching declines in US and global equity markets during October coupled with bond market outperformance will undoubtedly require MASSIVE monthly asset rebalancing by US pension funds –- rotating OUT of bonds and INTO stocks. This may have a profound “short-term” impact on performance of risk assets since the required rebalancing appears to eclipse even the large rotation after the 1987 stock market crash. As a very simple example, we asked our quant colleague (xxx) to analyze a balanced portfolio targeting 40% domestic bonds (SBBIG Index) and 60% equities.

82 We assumed that the equity portion is comprised of 75% domestic stocks (MXUS Index) and 25% EAFE international equities (MXEA Index). The attached rebalancing calculations based on closing levels last Friday (Oct 24) suggest that US pension funds would need to reduce bond holdings by a WHOPPING -4.1% while increasing equity allocations by a corresponding +4.1%, all by the close of business at month-end on Halloween Friday (Oct 31). Price action could be bloody scary given terrifying poor liquidity in these markets. For historical perspective, the second largest monthly bond-stock rebalancing rotation was 3.4% in October 1987. Most importantly, US equities did manage to stage a +10.5% during the last four trading days of October 1987 while bonds struggled. As it turns out, that marked the bottom for US equities for the next month and probably helped stocks find some needed footing in 1987. Bottom line: BEWARE the potential bounce in risk assets due to bond-stock rotation this week. FX risk trades may also tend to recover a little lost ground. This entry was posted by Neil Hume on Tuesday, October 28th, 2008 at 13:18 and is filed under Capital markets. Tagged with citigroup. CDS report: European credit derivatives at record wides by David Oakley Tuesday, October 21st, 2008 at 12:40 European credit derivatives were trading at record wides on Tuesday amid growing worries over the implosion of a hedge fund or insurance group because of exposure to failed bank Lehman Brothers. Many hedge funds and insurance groups sold protection to Lehman, the US investment bank which collapsed last month, and are now facing difficulties in repaying contracts, which must be settled in the coming weeks. One trader said: “The next big story could be the implosion of a hedge fund. That is what the market is particularly nervous about at the moment.” The iTraxx Crossover index, which tracks the debt of 50 mainly high-yield companies in Europe, was trading around 780 basis points, or 780,000 euros to protect 10m euros of debt over five years - a record wide - in a sign of the nervousness in the market place. Meanwhile, European sovereign credit default swaps, a form of protection against default on government bonds, were also trading wide amid concerns over the high levels of debt governments will have to issue as they recapitalise the banking system. The so-called peripheral nations of Greece, Italy, Portugal and Ireland are facing the greatest pressures along with Spain, where property prices have crashed. Greek CDS prices over five years are trading around a record 95bp, while Italy and Ireland are trading around 80bp and Portugal and Spain around 70bp. These are twice the level such CDS were at before the collapse of Lehman, which imploded on September 15. In contrast, the CDS prices of European banks have dropped dramatically in the past two weeks. Royal Bank of Scotland, Barclays and HBOS are all trading around 100bp compared with close to 500bp in the case of HBOS at the end of last month. This entry was posted by David Oakley

83 The Anatomy of a Fictional Hedge Fund Collapse by Rob Kirby May 26, 2005

I had a dream last night. I dreamt that this past week on bubble vision 2, CNBC ran an expose entitled Hedge Fund Hallucinations. Well, I must admit, whenever I think of hedge funds and hallucinations – not only do I think of lousy wizards pushing and pulling levers behind curtains, but the notion of ‘waking up some place that doesn’t look like Kansas’ always seems to come to mind – but let’s not get ahead of ourselves, ehhh? In this nightmare, I mean dream, no less than 4 Nobel Laureates worked for a hedgy called STCM [Short Term Capital Management]. As I recall, the fund in my dream had this ‘dream team’ [pun] of 4 Nobel laureates because the management intuitively knew that ‘size was wise’ and the name -- STCM – well, why don’t you just ask anyone if they’ve ever heard of a hedge fund that invests for the long term? [Size does not matter, does it?] So, anyway, this hedgy employs unimaginable leverage [that no one can even imagine who would be so silly to lend them so much] and makes this outrageous bet on foreign debt and gets ‘caught’ the wrong way around – in a ‘perfect storm’ known as a debt default. A liquidity crises ensues before a private corporation masquerading as a Government Central Bank intervenes in the whole affair and sweeps the real machinations of what happened ‘under the carpet’. [lender of last resort or retort? In the dream, STCM had an investor or client known as the Bank of Ipaly. The Bank of Ipaly had a 100 million investment in the hedge fund. What the world never understood was that the Bank of Ipaly also has gold on its books at 35 bucks an ounce. The Ipalians were so smart, they made their investment to the hedge fund “in kind” – using 100 million worth of their gold [which they still valued at 35 bucks an ounce]. So, in essence, the market value of their investment in STCM was really 10-fold [plus some] what was advertised. But using this accounting trick, no one was the wiser! This was so brilliant -- agents acting for other Central Banks caught wind, couldn’t resist - and did the same! So then, the team of Nobel Laureates [being the bright lights that they were] sold all the Bank of Ipaly’s and other’s gold and had many billions in cool hard cash. With those proceeds they leveraged themselves into other paper assets that didn’t do so good [that would be the foreign bonds mentioned above]. The funny thing, STCM never “appeared” to be leveraged to the extent that they really were – to most of their legit bankers [yup, there are some] -- because they had sold all this gold [which no one in the public ever understood they had] to raise cash. This kept all but the most knowledgeable of their financial advisors [like the ones who sold all that gold] in the dark – that is until the foreign bonds defaulted. You see, once the bonds defaulted, then the hedge fund’s lead bankers [Silverman Socks and J.P. Horton] were more than willing to share fault [all the paper losses] with all involved. So does Horton give a who? The gold short was of course secretly absorbed [taken off the books] by the private corporation that masquerades as a Government Central Bank, because they were actively and surreptitiously rigging the world price of gold and wanted to live to fight another day [kinda like Rocky Balboa?]. Don’t worry folks, the dream keeps on getting better!

84 In The After Math or Prelude to the Afternoon of a Financial Farce The implosion of this hedgy was dealt with in an extremely secretive manner since crooked officials needed cover to remove and hide the gold short. The ‘cover’ they chose was that intervention was required to prevent systemic threats to the global financial system. This lame excuse almost worked! The Nobel Laureates and management of the hedgy were sworn to secrecy and would never speak the truth – that’s for sure – being given the choice of their silence or lifetime jail terms or worse. But then anecdotal evidence surfaces that the bailout of STCM was really as a result of ‘a massive gold short’. This gets pointed out on page 29 of a research report titled, Not Free [Cause Nothing Is] prepared by an Investment firm called Stott Securities. The revelation of this gold short would expose a long practiced regime of rigging the world gold price by the private corporation that masquerades as a government central bank – hence something had to be done – but what? Officialdom decides to employ their minions to deride the credibility of all who oppose them. They employ the media to tacitly assist in this regard. The corporate media is more than happy to oblige because, heck, they’re running businesses too! Tune in next week for the Conclusion of: The Anatomy of a Fictional Hedge Fund Collapse http://www.financialsense.com/editorials/reality/2005/0526.html

Fictional Hedge Fund Collapse: The Next Chapter by Rob Kirby August 24, 2005 At the conclusion of our first installment, [when free media had been renamed the Ministry of Truth] the employees of our beloved hedgy STCM [Short Term Capital Management] had just been sworn to secrecy – the secret, of course, being the massive gold short removed from their books by Central bankers, Estate Secretary - Roger the Raccoon, and let’s not forget the noble [or Nobel, take your pick?] assistance from their investment banker cohorts at Silverman Socks and J. P. Horton. Anyway, with the crisis being brought under control, or, shall we say contained in the short term – the Central Gangsters, err Banksters realized they were going to have to concoct something a little bigger – that is to say, more grand – to keep fooling even the dumbest of a dumbed-down Six Pack public, drinking and grazing, believing ‘all was in fact’ – you know, well. The question, how to do it! This was going to require a plan, Stan. So, the Central Banksters decided to hold a convention in Albuquerque, New Mexico, to wheel and deal, deceive and devise and otherwise chart their future intentions where gold was concerned [cause they were clearly running out of the stuff, ehhh?]. Anyhow, as an honest oversight, the Central Banksters forgot to make their room reservations in Albuquerque early enough [or perhaps made a wrong turn somewhere?] and all the hotels were full - and instead they had to make emergency reservations in their 2nd alternative location [after the La Brea Tar Pits were

85 full too] - Washington, D.C. In the words of one of the participants, E. D. Smith – in his best English, “By George, we were in a jam [a sticky situation?] and were looking into the abyss. If the price of gold rose further – we were all hula-hoop-ed. We almost got tarred and feathered! Good thing the tar pits were full [booked]. It’s also a good thing the private corporation that masquerades as a Government Central Bank bailed us out and secretly sold or leased some of their gold stocks and had their stooge J.P. Horton [the who’s who of Derivative Banking] swamp the gold derivatives market with untold billions of synthetic gold short sales as Chasers, ehhhh?” Meanwhile, back at their hastily called convention, where they ended up having the world’s largest Texas Hold ‘Em convention - because everyone knows how much they like to gamble [risk, ehhh?], they formed a pact. Between bluffs, they decided to agree or consent - or otherwise conspire and cajole, concocting a ‘green plan.’ The plan spanned 5 years and involved selling vast hoards of their sovereign gold [shhhh, most of which was already leased and had left the vaults anyway, so keep it quiet and don’t tell anyone, ehhh?] to obfuscate their true intentions - to keep the Freedom dollar strong and the price of gold capped – cause once you’re in this deep, there’s just no way you can afford to fold. Besides, it doesn’t really matter what the size of the fraud you commit, so long as you are doing it in the name of Freedom [or fraud perhaps?]. Now I know that some of you folks are going to think, ‘this story is just too funny’ to be real. Well, I agree, but it gets funnier – believe me. Did I mention to you that the Estate Secretary [Roger the Raccoon] and then Chairman of Silverman Socks [Frank Cortizone] were take- charge type of guys and were instrumental in the diabolical cover up of the Short Term Capital gold swindle? Anyway, with life being as ironical as it is, the Estate Secretary really relished the idea of being a big city banker, while [you’re not going to believe this], being perfectly frank – Cortizone was more interested in politics [where he could rid baseball of steroids] all along. So the two of them switched sides! Meanwhile, the job of Estate Secretary was plugged [or Pugged, take your pick?] during a short but hot summer before getting back to school – Harvard of course! ***Scary Music Interlude*** For any of you who have ever been to the movies before, most of you would be well aware of the Hubris that Hollywood really spews. And the scary music is usually a lead in to the most heinous part of the whole story, right? Well, I’m happy to report, that we’re not going to let you down! All’s Well That Ends Well? Shortly after Frank Cortizone and Roger the Raccoon mustered up the energy to switch sides, so to speak, the greatest calamity – to date – befell Freedomville. Talk about the luck of the Irish - always after me lucky charms, ehh? The real price of loyalty in case you were unaware [or fell asleep under a tree for a number of years, perhaps?] is calamity; the most pervasive fraud [that anyone owned up to] to ever grip Freedomville – namely, the collapse of the corporate energy behemoth – EndRuin. Then it snowed. http://www.financialsense.com/editorials/reality/2005/0824.html

LTCM REVISITED - A FORENSIC ACCOUNT by Rob Kirby KirbyAnalytics.com July 9, 2007

LTCM was a hedge fund based in Greenwich, Connecticut, USA. The fund was formed in 1994 by a group of ex-Salomon Brothers traders led by John Meriwether. The key principals (in addition to Meriwether) included Eric Rosenfield, Lawrence Hilibrand, William Krasker, Victor Haghani, Greg Hawkins and David Modest. LTCM principals included Nobel price winners Robert Merton and Myron Scholes and former regulators including former Federal Reserve Board Vice Chairman David Mullins. It’s been written that, “the presence of Merton, Scholes and Mullins was puzzling. Merton and Scholes were at heart academics engrossed in research. Despite consulting gigs, they were unworldly when it came to the trading wars. Mullins was a career central banker [former vice-chairman of the Federal Reserve]. But they were names.” Or were they simply names? As Adam Hamilton reported back in the year 2000: Persistent rumors exist that LTCM was short 400 tonnes of gold when it went belly up. The US government arranged for someone to supply this gold owed to counterparties very quietly, and forbade any LTCM principals to ever discuss the gold position and disposition in the future. Although the whole LTCM and gold scenario is incredibly intriguing, it is topic for a future essay. It is a fact that one of LTCM’s investors was none other than the Bank of Italy. There have been books penned on this topic and here is a snippet of a review from one of them - "Inventing Money" is published in London by John Wiley & Sons Ltd:

LTCM-BANK OF ITALY PLOT Published in December 1999, Nicholas Dunbar's book on the fall of Long-term Capital Management has more to offer than the usual tale of intellectual arrogance and economic hubris. Amidst the stories of Wall Street "rocket scientists" creating money- making machines of fiendish complexity, there are some nuggets concerning the run up to EMU, and the hedge funds relationship with the Banca d'Italia, in a nutshell: "According to some observers who prefer to remain anonymous, the Bank provided LTCM with market access and privileged information denied to Italian banks - which would yield it a massive profit. In return, LTCM - and a handful of others - would engineer the convergence of Italian debt and get Italy into Emu. The Bank also invested in LTCM, effectively front-running the population of Italy."

87 In October 1994, the Italian Foreign Exchange Office, separate but closely related to the Bank, invested $100 million in the hedge fund. Alberto Giovannini, chairman of the "Council of Experts" advising the Italian Treasury at the time later found a job at none other than LTCM, under Victor Haghani, the man who allegedly engineered the deal. [The initially reported 100 million investment in LTCM by the Bank of Italy was later reported to be a more robust 250 million – RK] In case ANYONE forgets – GOLD is an OFFICIAL RESERVE ASSET. I used to MISTAKENLY BELIEVE that the Bank of Italy had GOLD on its books at 35 bucks an ounce. I was wrong. The Bank of Italy actually revalues [marks to market if you will] its gold bullion every three months and has done so for a VERY long time. While there are European countries [or were at the time] that carried “cheap” gold on their books [like Germany and Switzerland] – Italy was not one of them. So What Was Really Up With The Bank of Italy and a Gold Loan to LTCM? In 1997 – in the drive to meet conditions set down under the Maastricht Treaty to join the European Monetary Union [EMU] – Italy’s national finances were a mess. Folks would do well to understand that – unique among the EU member states - the Italian central bank consists of two institutions – of which the central government has no equity holdings in either: * The Banca d’Italia [BI] * Ufficio Italiano dei Cambi [UIC] The government DOES receive approximately 60% of profits generated by the BI and roughly 25% of profit generated by the UIC. Now, appreciate that way back in 1996 – the UIC purchased 540 tonnes of gold from the BI to “secure” a 2 billion dollar loan from the German Bundesbank. The following year, the UIC resold the gold bullion back to the BI for 10,500 billion lire – generating a capital gain [profit] for the BI of 7,600 billion lire. The government’s “TAKE” on this transaction was some 3,400 billion lire – a number that amounted to .2 % of GDP. This is all explained in greater detail at this link, pages 124 – 127]. This was nothing less than financial chicanery being practiced at the penultimate. The coup-de-grace, so to speak, was the GOLD LOAN to LTCM. Journalist John Brimelow ‘lays out’ the intimate nitty-gritty details of exactly how the personnel of LTCM and the Italian central bank were intertwined: Dunbar directly asserts, and supports with a detailed discussion, what can only be inferred from Lowenstein: that the Italian authorities in effect hired LTCM to groom or manipulate the Italian bond market, in order to accelerate convergence with the other European Monetary System bond markets and to reduce the Italian government interest burden. This permitted the achievement of the Maastrich criteria and allowed Italy to adopt the Euro. "According to some observers...the Bank of Italy provided LTCM with market access and privileged information denied to Italian Banks - which would yield a massive

88 profit. In return, LTCM - and a handful of others - would engineer the convergence of Italian debt..." Dunbar lays out in considerable detail how this was done. One element was the overlooking by the Italians of LTCM's repeated cornering of Italian bond auctions, greatly to the dismay of small local players. Of course, the U.S. Treasury had, since 1989, capped the proportion available to a single buyer at the American auctions at 35%. (It was a subordinate's flouting of this regulation which caused Merriwether's fall at Salomon.) The Italians were not so scrupulous. How many other such cozy arrangements were there? As mentioned in my discussion of Lowenstein’s book, the credit department of UBS took comfort in the fact that about 31% of LTCM was owned by "generally government-owned banks in major markets" who could supply LTCM with market intelligence. Dunbar seems to imply that other ECU markets received LTCM’s ministrations. And he directly says of LTCM that by the end of 1997 “Governments treated it as a valued partner, to be used whenever markets weren't efficient enough to achieve macroeconomic goals." (In this context, of course, "efficient" means "obedient" and "macroeconomic" means "political.”) This leads directly to the question of gold. Dunbar, like Lowenstein makes no reference at all to gold, not even to repudiate the rumors of a large LTCM short position. And indeed such a position must have either been eliminated or else been very well hidden by the time LTCM was invaded by hordes of Goldman and J.P. Morgan investigators in late September '98. By “lending” gold to LTCM – the Italian government was ONCE AGAIN resorting to nothing more than GREED motivated financial trickery. By loaning bullion to LTCM – the Italian government was counting – heck banking - on LTCM’s MAGNIFICENT returns to generate a nice fat profit for the BI – of which they would be entitled to 60 % of the spoils – which would ‘undoubtedly’ help them over the hump in meeting their financial obligations in conforming to EMU guidelines. The Russian Bond Default that unexpectedly “buried” LTCM put an end to those plans – didn’t it? With the Italians no-doubt ‘still licking their wounds’ over this golden fleecing of some 400 tonnes of sovereign gold – it’s no wonder they have NOT BEEN a seller of gold bullion [to date] under ANY of the Washington Agreements – to which they are a signatory. What I’m saying here is that LTCM was bailed out – categorically – because they were “short borrowed Central Bank gold”. A very public failure of LTCM would have undeniably exposed this fact to the whole world. Interestingly, it was the Federal Reserve in cahoots with the U.S. Treasury who would not allow this to happen. Understanding this is very key to understanding what course of action[s] may or may not be forthcoming by monetary officials now that other celebrated Hedge Funds are seemingly ‘blowing up all over the place’. Invest at your own risk. http://www.financialsense.com/fsu/editorials/kirby/2007/0709.html

89 90

NBER WORKING PAPER SERIES

SUDDEN STOPS, FINANCIAL CRISES AND LEVERAGE: A FISHERIAN DEFLATION OF TOBIN'S Q

Enrique G. Mendoza

Working Paper 14444 http://www.nber.org/papers/w14444

NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge, MA 02138 October 2008

I am grateful to Guillermo Calvo, Dave Cook, Mick Devereux, Gita Gopinath, Tim Kehoe, Nobuhiro Kiyotaki, Narayana Kocherlakota, Juan Pablo Nicolini, Marcelo Oviedo, Helene Rey, Vincenzo Quadrini, Alvaro Riascos, Lars Svensson, Linda Tesar and Martin Uribe for helpful comments. I also acknowledge comments by participants at the 2006 Texas Monetary Conference, 2005 Meeting of the Society for Economic Dynamics, the Fall 2004 IFM Program Meeting of the NBER, and seminars at the ECB, BIS, Bank of Portugal, Cornell, Duke, Georgetown, IDB, IMF, Harvard, Hong Kong Institute for Monetary Research, Hong Kong University of Science and Technology, Johns Hopkins, Michigan, Oregon, Princeton, UCLA, USC, Western Ontario and Yale. Many thanks also to Guillermo Calvo, Alejandro Izquierdo, Rudy Loo-Kung and Ernesto Talvi for allowing me to use their classification of Sudden Stop events. The views expressed herein are those of the author(s) and do not necessarily reflect the views of the National Bureau of Economic Research.

© 2008 by Enrique G. Mendoza. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source. 91

Sudden Stops, Financial Crises and Leverage: A Fisherian Deflation of Tobin's Q Enrique G. Mendoza NBER Working Paper No. 14444 October 2008 JEL No. D52,E32,E44,F32,F41

ABSTRACT

This paper shows that the quantitative predictions of a DSGE model with an endogenous collateral constraint are consistent with key features of the emerging markets' Sudden Stops. Business cycle dynamics produce periods of expansion during which the ratio of debt to asset values raises enough to trigger the constraint. This sets in motion a deflation of Tobin’s Q driven by Irving Fisher’s debt-deflation mechanism, which causes a spiraling decline in credit access and in the price and quantity of collateral assets. Output and factor allocations decline because the collateral constraint limits access to working capital financing. This credit constraint induces significant amplification and asymmetry in the responses of macro-aggregates to shocks. Because of precautionary saving, Sudden Stops are low probability events nested within normal cycles in the long run.

Enrique G. Mendoza Department of Economics University of Maryland College Park, MD 20742 and NBER [email protected] 92

“A story is a string of actions occurring over time, and debt happens as a result of actions occurring over time. Therefore, any debt involves a plot line: how you got into debt, what you did, said and thought while you were there, and then—depending on whether the ending is to be happy or sad—how you got out of debt, or else how you go further and further into it until you became overwhelmed by it, and sank from view.” (Margaret Atwood, “Debtor’s Prism,” Wall Street Journal, 09/20/2008, p. W1)

1. Introduction

The Great Depression showed that market economies can experience deep recessions that differ markedly from typical business cycle downturns. The recessions that hit emerging economies in the aftermath of the financial crises of the late 1990s illustrated the same fact. In contrast with the Great Depression, however, the loss of access to world capital markets played a key role in emerging markets crises. That is, these crises featured the phenomenon now commonly referred to as a “Sudden Stop.”

Three striking macroeconomic regularities characterize Sudden Stops: (1) reversals of international capital flows, reflected in sudden increases in net exports and the current account, (2) declines in domestic production and absorption, and (3) corrections in asset prices. Figure 1 illustrates these facts using five-year event windows, centered on Sudden Stop events occurring at date t. The dating and location of Sudden Stops follows Calvo, Izquierdo and Talvi’s (2006) classification.1 The charts show event dynamics for output (GDP), consumption (C), investment (I), the net exports-GDP ratio (NXY) and Tobin’s Q. Data for GDP, C, I, and NXY are from World Development Indicators. Q is estimated for each country as the median across firm-level estimates computed for listed corporations using Worldscope data. Firm-level Q is the ratio of market value of equity plus debt outstanding to book value of equity. The observations in the event windows correspond to cross-country medians of deviations from Hodrick-Prescott trends estimated using 1970-2006 data for each country, except for Q which is not detrended because the data starts in 1994.2

As Figure 1 shows, Sudden Stops are preceded by expansions, with domestic absorption and production above trend, the trade balance below trend, and high asset prices. The median Sudden Stop displays a reversal in the cyclical component of NXY of about 3 percentage points of GDP, from a deficit of about 2 percent of GDP at t-1 to a surplus of 1 percent of GDP at t, and this surplus persists at t+1 and t+2. GDP and C are about 4 percentage points below trend at date t, and I collapses almost 20 percentage points below trend. The economies recover somewhat afterwards, but GDP, C and I remain below trend two years after the Sudden Stops hit. Q reaches a through at date t about 13 percentage points below the pre-Sudden-Stop peak, and it recovers about 2/3rds of its value by t+2.

1 Calvo et al. identified 33 Sudden Stop events with large and mild output collapses in a sample including the 31 countries that JP Morgan defines as emerging markets. Their paper provides further details. Calvo and Reinhart (1999), Calvo, Izquierdo and Loo-Kung (2006), and Milesi-Ferretti and Razin (2000) use other definitions of Sudden Stops, but the actual listings of events are very similar. 2 Note two differences with the event analysis in Calvo et al. First, they study event windows with cross-country averages of country-specific cumulative growth rates. We use medians instead of averages because of substantial cross-country dispersion in cyclical components, and deviations from trend instead of cumulative growth to remove low-frequency dynamics. Second, Calvo et al. focus mainly on Sudden Stops with large output collapses. Here we include all Sudden Stop events.

In addition to the stylized facts illustrated in Figure 1, Sudden Stops are characterized by three important features: First, Sudden Stops are nested within typical business cycles. They are rare events in each country by construction, because a key criterion to identify them is that a country’s international capital flows are significantly below their mean (see Calvo et al. (2006)). Second, they represent business cycle asymmetries (i.e., symmetric episodes of sudden large drops in trade surpluses accompanied by surges in output and absorption are not observed). Third, standard growth accounting shows that a large drop in the Solow residual accounts for a Sudden Stops’ initial output collapse. Part of this is due to factors that bias the Solow residual as a measure of “true” total factor productivity (TFP), such as changes in imported inputs, capacity utilization, and labor hoarding (see Bergoeing et. al. (2002), Mendoza (2006), and Meza and Quintin (2006)). For instance, Mendoza shows that in Mexico’s 1995 Sudden Stop, a large drop in imported inputs accounts for 3.1 percentage points of the 8.5 percent fall in output per worker between the 94:Q3 and 95:Q2.

The characteristics that define Sudden Stops suggest that a dynamic stochastic general equilibrium (DSGE) model aiming to explain this phenomenon should support a stochastic steady state in which infrequent Sudden Stops are nested together with normal business cycles. In addition, the model should display amplification and asymmetry of economic fluctuations during Sudden Stop episodes: Typical realizations of the same underlying exogenous shocks that produce normal business cycles in non-Sudden Stop states should result in reversals of capital flows, economic recessions and declines in asset prices during Sudden Stops. Moreover, Sudden Stops should feature endogenous declines in variables that affect Solow residuals, and these should play a role in the output drop. This paper proposes a model with these properties, and shows that it can deliver quantitative predictions consistent with actual Sudden Stops.

Explaining Sudden Stops is a challenge for a large class of DSGE small open economy (SOE) models, including frictionless real business cycle models and models with nominal rigidities. This is because these models typically assume perfect world credit markets that act as an efficient vehicle for consumption smoothing and investment financing. For example, in response to a large output drop, households formulate optimal plans to smooth the effect on consumption by borrowing from abroad, while in the data the opposite is observed (the external accounts rise sharply precisely when consumption and output collapse). In contrast, the literature on Sudden Stops emphasizes the role of credit frictions. Several studies propose models that predict adjustments in production, absorption and the external accounts as a result of the adverse effects of these frictions (e.g. Auenhaimer and Garcia (2000), Izquierdo (2000), Calvo (1998), Gopinath (2003), Cook and Choi (2003), Cook and Devereux (2006a, 2006b), Martin and Rey (2006), and Gertler, Gilchrist and Natalucci (2007)). The model proposed in this paper follows on a similar path, but it differs in its focus on the amplification and asymmetry of macroeconomic fluctuations that Irving Fisher’s (1933) classic debt-deflation transmission mechanism produces.

The model introduces an endogenous collateral constraint with the debt-deflation mechanism into a DSGE-SOE model driven by three standard exogenous shocks affecting TFP, the foreign interest rate, and the price of imported intermediate goods. The collateral constraint limits total debt, including both standard intertemporal debt and atemporal working capital loans, not to exceed a fraction of the market value of the physical capital that serves as collateral. Thus, the constraint imposes an upper bound on the aggregate

2 94

leverage ratio of the economy. The emphasis is on studying the quantitative significance of this credit friction, along the lines of the growing literature on the macroeconomic implications of credit constraints (as in Kiyotaki and Moore (1997), Bernanke, Gertler and Gilchrist (1998), Aiyagari and Gertler (1999), Kocherlakota (2000), Cooley, Miramon and Quadrini (2004), Jermann and Quadrini (2005), and Gertler, et al. (2007)).

Standard DSGE-SOE models cannot produce Sudden Stops even if working capital and/or imported inputs are added. Agents in these models still have unrestricted access to a perfect international credit market. Negative shocks to TFP and/or the world price of imported inputs induce standard consumption-smoothing and investment-reducing effects. Large shocks could trigger large output collapses driven in part by cuts in imported inputs, but this would still fail to explain the current account reversal and the collapse in consumption (since households would borrow from abroad to smooth consumption). Adding large shocks to the world interest rate or access to external financing can alter these results, but such a theory of Sudden Stops would hinge entirely on unexplained “large and unexpected” shocks. Large, because by definition they need to induce recessions larger than the normal non-Sudden-Stop recessions, and unexpected (i.e. outside the set realizations agents consider possible), because otherwise agents would self-insure to undo their real effects. Paradoxically, large, unexpected shocks often drive reversals of capital flows in the models proposed in the Sudden Stops literature (e.g. Calvo (1988)). In contrast, this paper shows that a debt-deflation-style collateral constraint can provide an explanation for Sudden Stops that does not hinge on large, unexpected shocks.

The debt-deflation collateral constraint adds three important elements to the model’s business cycle transmission mechanism that are crucial for the quantitative results:

(1) The constraint is occasionally binding, because it only binds when the leverage ratio is sufficiently high. When this happens, the economy responds to typical realizations of the exogenous shocks by displaying Sudden Stops. Moreover, if the constraint does not bind, the shocks yield similar macroeconomic responses as in a typical DSGE-SOE model with working capital and perfect credit markets. As a result, the economy displays “normal” business cycle patterns when the collateral constraint does not bind.

(2) The loss of credit market access is endogenous. In particular, the high leverage ratios at which the collateral constraint binds are reached after sequences of realizations of the exogenous shocks lead the endogenous business cycle dynamics of the economy to states with sufficiently high leverage. Since net exports are countercyclical in the model, these high- leverage states are preceded by economic expansions, as observed in emerging economies. However, Sudden Stops have a low long-run probability of occurring, because agents accumulate precautionary savings to reduce the likelihood of large consumption drops. Hence, Sudden Stops are rare events nested within typical business cycles.

(3) Sudden Stops are driven by two “credit channel” effects that induce amplification, asymmetry and persistence in the effects of exogenous shocks. The first is an endogenous financing premium that affects one-period debt, working capital loans, and the return on equity because the effective cost of borrowing rises when the collateral constraint binds. The second is the debt-deflation mechanism: When the collateral constraint binds, agents liquidate capital in order to meet “margin calls.” This fire-sale of assets reduces the price of

3 95

capital and tightens further the constraint, setting off a spiraling collapse of asset prices. Consumption, investment and the trade deficit suffer contemporaneous reversals as a result, and future capital, output, and factor allocations fall in response to the initial investment decline. In addition, the restricted access to working capital induces contemporaneous declines in production and factor demands.

The quantitative analysis of the model is conducted using a baseline calibration based on a detailed analysis of Mexican data, but the focus is on exploring the model’s ability to match the Sudden Stop features observed across countries. The upper bound on leverage is calibrated so that the model’s stochastic stationary equilibrium matches the observed frequency of Sudden Stops in the dataset of Calvo et al. (2006), which is about 3.3 percent. The long-run probability of observing Sudden Stops is reduced by precautionary savings, and hence the model requires an upper bound on the leverage ratio of about 1/5 in order to match the 3.3 percent Sudden Stops probability.

The results show that model economies with and without the collateral constraint exhibit largely the same long-run business cycle co-movements, but the economy with the collateral constraint displays significant amplification and asymmetry in the responses of macroeconomic aggregates to one-standard-deviation shocks. Amplification is reflected in significantly larger average responses conditional on positive-probability states in which the collateral constraint binds. Asymmetry is shown in that the responses to shocks of the same magnitude, but conditional on states in which the collateral constraint does not bind, are about the same in the economies with and without the credit friction.

The ability of the model to replicate observed Sudden Stops is evaluated by conducting stochastic simulations and constructing event analysis windows with the simulated data that are comparable with those shown in Figure 1. The results indicate that the model matches the behavior of output, consumption, investment and net exports, including the collapse when Sudden Stops hit, the periods of economic expansion that precede them, and the pattern of the recovery that follows. The model also replicates the observed dynamics of Tobin’s Q qualitatively, but quantitatively it underestimates the collapse of asset prices. Moreover, in the model’s Sudden Stop events, the Solow residual overestimates the true estate of TFP by about 30 percent.

Sensitivity analysis shows that the loss of access to working capital financing plays a key role in the model’s ability to produce amplification and asymmetry in the responses of GDP and factor allocations, in yielding Sudden Stop dynamics consistent with those observed in the data, and in producing a gap between true TFP and the Solow residual. Increasing the share of imported inputs in production increases the amplitude of the Sudden Stop-induced fluctuations in GDP, factor allocations and working capital, and it also increases the bias between the Solow residual and true TFP (with the former overstating the latter by about 50 percent). The opposite results are obtained if instead of increasing the share of imported inputs in production we lower the Frisch elasticity of labor supply.

The collateral constraint used in this paper is similar to the margin constraint used by Mendoza and Smith (2006) in their extension of the Aiyagari-Gertler (1999) setup to an environment of global asset trading. The model studied here is significantly different, because it is a full-blown equilibrium business cycle model with endogenous capital

4 96

accumulation and dividend payments that vary in response to the collateral constraint, and the constraint limits access to working capital financing. In contrast, Mendoza and Smith study a setup in which production and dividend payments are unaffected by the credit constraint, abstract from modeling capital accumulation, and consider a credit constraint that limits only the access to household debt.

This paper is also closely related to two important strands of the literature that study the quantitative implications of financial constraints for emerging markets business cycles. One is the strand that studies the effects of working capital financing on long-run business cycle co-movements (see Neumeyer and Perri (2005), Uribe and Yue (2006) and Oviedo (2004)). The model of this paper differs in one key respect: Working capital loans require collateral, so that when the collateral constraint binds, the cutoff in working capital loans contributes to the amplification and asymmetry observed in the Sudden Stop responses of output and factor demands to shocks. Moreover, the model is parameterized so that only a small fraction of factor costs needs to be paid in advance. As result, in the absence of the collateral constraint, working capital makes little difference for business cycle dynamics (relative to a frictionless economy).

The second strand of the quantitative business cycles literature related to this paper is the one that introduced the Bernanke-Gertler financial accelerator into DSGE-SOE models with nominal rigidities. Notably, Gertler et al. (2007) calibrated a model of this class to Korean data, and studied its ability to account for the 1997-1998 Korean crash as a response to a large exogenous shock to the real foreign interest rate. In addition, Gertler et al. introduced a mechanism to drive the output collapse together with a decline in productivity as measured by the Solow residual by modeling variable capital utilization. This paper introduces a different financial accelerator mechanism, based on an occasionally binding collateral constraint, and uses imported intermediate goods to produce a decline in the Solow residual.3 The qualitative interpretation of the feedback between asset prices and debt is similar to the one in Gertler et al., but the debt-deflation mechanism yields endogenous Sudden Stops that do not require large, unexpected shocks, co-exist with regular business cycles, and produce asymmetric effects that amplify business cycle downturns. On the other hand, since solving the model requires non-linear global solution methods for DSGE-SOE models with incomplete markets, the model is much less flexible than the framework of Gertler et al. for studying the interaction of the financial accelerator with nominal rigidities and monetary and exchange rate policy.

The paper is organized as follows. Section 2 describes the model and characterizes its competitive equilibrium. Sections 3 and 4 focus on calibrating the model and conducting the quantitative analysis. Section 5 concludes.

2. A Model of Sudden Stops and Business Cycles with Collateral Constraints

The model economy is a variation of the standard DSGE-SOE model with incomplete insurance markets, capital adjustment costs, and working capital financing (e.g. Mendoza (1991), Neumeyer and Perri (2005) and Uribe and Yue (2006)). Two important

3 A previous version of this paper used both imported inputs and variable utilization (see Mendoza (2006)). The latter was harder to calibrate and its contribution was quantitatively smaller.

5 97

modifications are introduced here. First, we introduce an endogenous collateral constraint. Second, the supply-side of the model is modified to introduce imported inputs.

2.1 Optimization problem

The economy is inhabited by an infinitely-lived, self-employed representative firm- household.4 The preferences of this agent are defined over stochastic sequences of

consumption ct and labor supply Lt, for t=0,…,∞ . Preferences are modeled using Epstein’s (1983) Stationary Cardinal Utility (SCU) function, which features an endogenous rate of time preference, so as to obtain a unique, invariant limiting distribution of foreign assets.5

The preference specification is:

⎡⎧∞−t 1 ⎫ ⎤ ⎪⎪ E ⎢⎥exp⎨⎬−−ρ ()()c NL() u c − NL() (1) 0 ⎢⎥∑∑⎪⎪ττtt ⎣⎩t ==00⎪⎪τ ⎭ ⎦ In this expression, u(.) is a standard twice-continuously-differentiable and concave period utility function and ρ(.) is an increasing, concave and twice-continuously-differentiable time preference function. Following Greenwood et al. (1988), utility is defined in terms of the excess of consumption relative to the disutility of labor, with the latter given by the twice- continuously-differentiable, convex function N(.). This assumption eliminates the wealth effect on labor supply by making the marginal rate of substitution between consumption and labor independent of consumption.

There are other approaches in addition to using Epstein’s SCU that yield well-defined stochastic stationary equilibria in DSGE-SOE models (see Arellano and Mendoza (2003) and Schmitt-Grohe and Uribe (2002) for details).6 In models with credit constraints, SCU has the advantage that it can support stationary equilibria in which the constraints can bind permanently. This is because a binding credit constraint drives a wedge between the intertemporal marginal rate of substitution in consumption and the rate of interest. In a stationary state with a binding credit constraint, the rate of time preference adjusts endogenously to accommodate this wedge. In contrast, in models with an exogenous discount factor, credit constraints never bind in the long run (if the rate of time preference is greater

4 Mendoza (2006) presents a different decentralization of a similar setup where firms and households are modeled as separate agents, and face separate collateral constraints. The setup with a self- employed representative firm-household yields very similar predictions and is much simpler to describe and solve (I am grateful to an anonymous referee for suggesting this approach). 5 Since agents face non-insurable income shocks and the interest rate is exogenous, precautionary saving leads foreign assets to diverge to infinity with the standard assumption of a constant rate of time preference equal to the interest rate. 6 Epstein showed that SCU requires weaker preference axioms than those behind the standard utility function with exogenous discounting. Standard preferences require preferences over stochastic future allocations to be risk-independent from past allocations, and past allocations to be risk-independent from future allocations, while SCU only requires the latter. He also proved that a preference order consistent with the weaker axioms can be expressed as a time-recursive utility function if and only if it takes the form of the SCU. Hence, other ad-hoc formulations of endogenous discounting can deliver stationary net foreign asset positions, but they violate the preference axioms.

6 98

or equal to the world interest rate) or always bind at steady state (if the rate of time preference is fixed below the interest rate).

A The economy operates a constant-returns-to-scale technology, exp(εt )F(kt,Lt,vt), that requires capital, kt, labor and imported inputs vt, to produce a tradable good sold at a world- determined price (normalized to unity without loss of generality). TFP is subject to a A random shock εt with exponential support. Net investment, zt = kt+1 - kt, incurs unitary investment costs determined by the function Ψ(zt/kt), which is linearly homogeneous in zt 7 and kt. Working capital loans from foreign lenders are needed to pay for a fraction φ of the cost of imported inputs and labor in advance of sales. The gross interest rate on these loans R R is the world real interest rate Rt=Rexp(εt ), where εt is an interest rate shock around a mean value R. As in Neumeyer and Perri (2005) and Uribe and Yue (2006), working capital loans are provided at the beginning of each period and repaid at the end. Imported inputs P are purchased at an exogenous relative price in terms of the world’s numeraire pt=pexp(εt ), P where p is the mean price and εt is a shock to the world price of imported inputs (i.e., a A R P terms-of-trade shock from the perspective of the SOE). The shocks εt , εt and εt follow a joint first-order Markov process to be specified in more detail later.

The representative agent chooses sequences of consumption, labor, investment, and holdings of real, one-period international bonds, bt+1, so as to maximize SCU subject to the following period budget constraint:

Ab cipvtt++ tt =exp(εφ t ) FkLv ( ttt , , )−( R t − 1)( wLpvqbb ttttttt +) −+1 + (2)

⎡⎛kk− ⎞⎤ where ikkk=+δ ()1 −⎢⎥ +Ψ⎜ tt+1 ⎟ is gross investment. The agent also faces the tttt+1 ⎢⎥⎜ ⎟ ⎣⎦⎝⎠kt following collateral constraint:

b qbtt++11−+≥−φκ R t( wL t t pv tt) qk tt (3)

b In the constraints (2) and (3), wt is the wage rate, qt is the price of bonds and qt is the b price of domestic capital. The price of bonds is exogenous and satisfies qRtt= 1/ , while wt and qt are endogenous prices that are taken as given by the representative agent and satisfy standard market optimality conditions: The price of capital equals the marginal cost of investment, qikkkttttt=∂(,)/()++11 ∂ , and the wage rate equals the marginal disutility of labor,wNLLttt=∂()/ ∂ , where variables with bars are “market averages” taken as given by the representative agent but equal to the representative agent’s choices at equilibrium.

The collateral constraint (3) implies that credit markets are imperfect. In particular, lenders impose a credit constraint in the form of the margin requirement proposed by Aiyagari and Gertler (1999): The economy’s total debt, including both debt in one-period bonds and in within-period working capital loans, cannot exceed a fraction κ of the “marked-to-market” value of capital (i.e. κ imposes an upper bound on the leverage ratio). Both interest and principal on working capital loans enter in the constraint because these

7 Specifying the capital adjustment cost in terms of net investment, instead of gross investment, yields a more tractable recursive formulation of the economy’s optimization problem that preserves Hayashi’s (1982) results regarding the conditions that equate marginal and average Tobin Q.

7 99

are within-period loans, and thus lenders consider that the market value of the assets offered as collateral must cover both components.

The collateral constraint is not derived here from an optimal credit contract. Instead, the constraint is imposed directly as in the models with endogenous credit constraints examined by Kiyotaki and Moore (1997), Aiyagari and Gertler (1999), and Kocherlakota (2000). Still, a credit relationship with a constraint like (3) could result, for example, from an environment in which limited enforcement prevents lenders to collect more than a fraction κ of the value of a defaulting debtor’s assets. As we explain below, in states of nature in which (3) binds, the model produces endogenous premia over the world interest rate at which borrowers would agree to contracts which satisfy (3).

2.2 Competitive Equilibrium & Credit Channels

A competitive equilibrium for the small open economy is defined by stochastic sequences ∞ ∞ of allocations ⎡⎤cLk,, , b ,, vi and prices ⎡qw, ⎤ such that: (a) the representative firm- ⎣⎦ttt++11 t tt0 ⎣ tt⎦0 household maximizes SCU subject to (2) and (3), taking as given wages, the price of capital,

the world interest rate, and the initial conditions (k0,b0), (b) wages and the price of capital satisfyqikkkttttt=∂(,)/()++11 ∂ and wNLLttt=∂()/ ∂ and (c) the representative agent’s choices satisfy kktt= and LLtt= .

In the absence of credit constraints, the competitive equilibrium is the same as in a standard DSGE-SOE model. In fact, removing also imported inputs, the model collapses into a model nearly identical to the models of Numeyer and Perri (2005) or Uribe and Yue (2006). The credit constraints distort the equilibrium by introducing two credit-channel effects. One of these effects is reflected in external financing premia affecting the cost of borrowing in bonds and working capital and the equity premium, and the second is the debt-deflation process. These credit-channel effects can be analyzed using the optimality conditions of the competitive equilibrium.

The optimality conditions of the representative agent’s problem yield the following

Euler equation for bt+1:

01(<−μλ ) =ER⎡ ( λ λ )⎤ ≤ 1 (4) tt t⎣⎢ t++11 tt⎦⎥ where λt is the non-negative Lagrange multiplier on the date-t budget constraint (2), which

equals also the lifetime marginal utility of ct, and μt is the non-negative Lagrange multiplier on the collateral constraint (3). It follows from (4) that, when the collateral constraint binds, the economy faces an endogenous external financing premium on the effective real interest h rate at which it borrows ( Rt +1 ) relative to the world interest rate. This expected external financing premium on debt (EFPD) is given by:

μλε+ cov( ,R ) λ ER⎡⎤hh−= R ttt++11, R ≡t (5) tt⎣⎦⎢⎥++11 t ⎡ ⎤⎡t + 1 ⎤ EEtt⎣λλ++11⎦⎣tt⎦ This premium can be viewed as the premium at which the SOE would choose debt amounts that satisfy the collateral constraint with equality in a credit market in which the constraint is not imposed directly.

8 100

In the canonical DSGE-SOE model, international bonds are a risk-free asset and μt=0 for all t, so there is no premium. In the model examined here, if the collateral constraint binds, there is a direct effect by which the multiplier μt increases EFPD. In addition, there is an indirect effect that pushes in the same direction because a binding credit constraint makes it harder to smooth consumption, and hence the covariance between marginal utility and the world interest rate is likely to increase.

The effects of the EFPD on asset pricing can be derived from the Euler equation for

capital. Solving forward this equation, taking into account that at equilibrium qt equals the marginal cost of investment, yields the following:

⎡ ∞ ⎛⎞⎤j ⎛⎞1 ⎟ ⎢ ⎜ ⎜ ⎟⎟ ⎥ qEtt= ⎢ ∑⎜∏⎜ ⎟⎟()dtj++1 ⎥, (6) ⎜ ⎜ ti+ ⎟⎟ ⎣⎢ j=0 ⎝⎠i=0 ⎝⎠Rti++1 ⎦⎥

⎛⎞⎛⎞2 ()λκμ− zztj++11⎟⎟ tj++ RdFti+ ≡≡ti++ ti ,exp()εδ A (,,)kLv −+⎜⎜⎟⎟Ψ′ ti++11tj++ tj++11 tjtjtj ++1 ++1 ++1⎜⎜⎟⎟ λti++1 ⎝⎠⎝⎠⎜⎜kktj++11⎟⎟ tj++

Thus, the price of capital equals the expected present discounted value of future dividends (d), discounted at a rate that reflects the effect of the collateral constraint.

The above asset pricing formula can be simplified further by combining the Euler equations for bonds and capital to obtain the following expression for the equity premium q (the expected excess return on capital, Rdqqt+1 ≡+( ttt++11)/ , relative to Rt+1):

μκ+ COV(, λ Rq ) ER⎡⎤⎡⎤qh−= R ER −− R tttt++11 (7) tt⎣⎦⎣⎦⎢⎥⎢⎥++11 t tt ++ 11 t Ett[]λ +1 This expression collapses to the standard equity premium if the collateral constraint does not bind and the world interest rate is deterministic. As Mendoza and Smith (2006) explained, when the collateral constraint binds it induces direct and indirect effects on the equity premium similar to those affecting EFPD. The two premia are not the same, however, because in the equity premium the direct effect of the binding collateral constraint ⎡ ⎤ on EFPD is reduced by the term μκtttE ⎣ λ+1 ⎦ , which measures the marginal benefit of being able to borrow more by holding an additional unit of capital. There is also a new element in the indirect effect that is not present in the EFPD, and is implicit in the covariance between q λt+1 and Rt +1 : A binding collateral constraint makes it harder for agents to smooth consumption and self-insure, and hence this covariance term is likely to become more negative when the constraint binds, thereby increasing the equity premium.

Given the result in (7), the asset pricing condition (6) can be re-written as:

⎛⎞∞ ⎡ j ⎛⎞1 ⎤ qE= ⎜⎟⎜ ⎢⎥⎜ ⎟d ⎟ (8) tt⎜ ∑ ⎢⎜∏⎜ q ⎟ ⎥ti++1 ⎟ ⎝⎠j =0 ⎣ i=0⎝⎠ERt []ti++1 ⎦ It follows then from (7) and (8) that, as Aiyagari and Gertler (1999) showed, higher expected returns when the collateral constraint binds at present, or is expected to bind in the future, increase the discount rate of dividends and lower asset prices in the present.

9 101

The external financing premium on working capital financing is easy to identify in the optimality conditions for factor demands:

A ⎡ μt ⎤ exp(εφttttt )FkLv2 ( , , )=++ w⎢ 1 ( r t( ) R t )⎥ (9) ⎣ λt ⎦

A ⎡ μt ⎤ exp(εφttttt )FkLv3 ( , , )=++ p⎢ 1 ( r t( ) R t )⎥ (10) ⎣ λt ⎦ These are standard conditions equating marginal products with marginal costs. In the right-

μt hand-side of (9) and (10), the term ( )Rt reflects the increase in the effective marginal λt financing cost of working capital caused by a binding collateral constraint. This external financing premium on working capital represents the excess over the world interest rate at which domestic agents in a competitive world market of working capital loans would find it optimal to agree to contracts that satisfy constraint (3) voluntarily.

The second credit channel present in the model, the debt-deflation mechanism, is harder to illustrate than the external financing premia because of the lack of closed-form solutions, but it can be described intuitively: When the collateral constraint binds, agents respond to “margin calls” from lenders by fire-selling capital (i.e., by reducing their demand for equity). However, when they do this, they face an upward-sloping supply of equity because of Tobin’s Q. Thus, at equilibrium it is optimal to lower investment given the reduced demand for equity and higher discounting of future dividends, and hence equilibrium equity prices fall. If the credit constraint was set as an exogenous fixed amount, these would be the main adjustments. But with the endogenous collateral constraint, if the constraint was binding at the initial (notional) levels of the price of capital and investment, it must be more binding at lower prices and investment levels, so another round of margin calls takes place and Fisher’s debt-deflation mechanism is set in motion. Moreover, the Fisherian deflation causes a sudden increase in the financing cost of working capital, lowering factor allocations and output.

Interestingly, the effects of the debt-deflation mechanism are non-monotonic, because they are weaker at the extremes in which the SOE can collateralize all of its assets (κ=1) or cannot borrow at all (κ=0) than in the cases in between. When κ=0 there can be no debt- deflation, since the constraint does not respond to asset values (i.e., it becomes an exogenous credit limit). On the other hand, when κ=1 there is no direct effect from the collateral constraint on the equity premium, which leaves only the indirect covariance effects to distort investment and the price of capital relative to the equilibrium with perfect credit markets. In the limiting case without uncertainty, the indirect effects vanish, and κ=1 removes all distortions on investment and the price of capital, and hence there is no debt-deflation mechanism again. Consumption and debt still adjust, but they do so as they would with an exogenous credit constraint. Hence, for the debt-deflation mechanism to operate, credit markets must allow borrowers to leverage their assets but only to some degree.

Mendoza (2006) illustrates the above arguments using a simple numerical example based on a deterministic version of the model. This example is comparable to the one conducted by Kocherlakota (2000). Mendoza found large amplification effects of the collateral constraint on output and asset prices. In contrast, Kocherlakota found small

amplification effects, using a borrowing constraint of the form bqxttt+1 ≥ , where xt can be a fixed factor (e.g., land) or physical capital. The two sets of results are consistent, however,

10 102

because the case with land prevents declines in xt from compounding with the decline in

asset prices in the debt-deflation dynamics, and with capital, the constraint bqxttt+1 ≥ implies κ=1 (which under perfect foresight removes the debt-deflation mechanism).

It is also important to note that a variety of actual contractual arrangements can produce debt-deflation dynamics. The collateral constraint (3) resembles most directly a contract with a margin clause. This clause requires borrowers to surrender the control of collateral assets when the contract is entered, and gives creditors the right to sell them when their market value falls below the contract value. Other widely used arrangements that can trigger debt-deflation dynamics without explicit margin clauses include value-at-risk strategies of portfolio management used by investment banks, and mark-to-market capital requirements imposed by regulators. For example, if an aggregate shock hits capital markets, value-at-risk estimates increase and lead investment banks to reduce their exposure, but since the shock is aggregate, the resulting sale of assets increases price volatility and leads value-at-risk models to require further portfolio adjustments. Mechanisms like these played a central role in the Russian/LTCM crisis of 1998 and the U.S. credit crisis of 2007-2008.

3. Functional Forms and Calibration

3.1 Functional Forms and Numerical Solution

The quantitative analysis uses a benchmark calibration based on Mexican data. The functional forms of preferences and technology are the following:

⎡ Lω ⎤1−σ ⎢⎥c −−t 1 ⎢⎥t ω uc(())−= NL ⎣⎦,,1,σω> (11) tt 1 − σ

ω ⎡ ⎛⎞Lt ⎟⎤ vc(())1tt−= NLγγ⎢⎥ Ln⎜ +−<≤ c t⎟ ,0σ , (12) ⎣⎢ ⎝⎠⎜ ω ⎟⎦⎥

βαη Fk( ttt,, Lv) =≤≤++=> AkLv ttt, 0αβη , , 1, α β η 1,A 0, (13)

⎛⎞zaz ⎛⎞ ⎜⎜tt⎟⎟ Ψ=⎜⎜⎟⎟,0a ≥ (14) ⎝⎠kktt2 ⎝⎠ The utility and time preference functions in (11) and (12) are standard from DSGE-SOE models. The parameter σ is the coefficient of relative risk aversion, ω determines the wage elasticity of labor supply, which is given by 1/(ω -1), and γ is the semi-elasticity of the rate of time preference with respect to composite good c-N(L). The restriction γ ≤ σ is a condition required to ensure that SCU supports a unique, invariant limiting distribution of bonds and capital (see Epstein (1983)). The Cobb-Douglas technology (13) is the production function for gross output. Equation (14) is the net investment adjustment cost function. Following Hayashi (1982), the production and adjustment cost functions are set to be linearly homogeneous in their arguments.

11 103

The model is solved numerically by representing the equilibrium in recursive form and using a non-linear global solution method with the collateral constraint imposed as an occasionally binding inequality constraint (see Mendoza and Smith (2006) and Arellano and Mendoza (2003) for details on algorithms for solving DSEG-SOE models with incomplete markets and collateral constraints). The endogenous state variables are k and b. These are

chosen from discrete grids of NK non-negative values of the capital stock, K={k1

3.2 Calibration

The values assigned to the model’s parameters are listed in Table 1. This calibration is set so that the deterministic stationary equilibrium matches key averages from Mexican data. We adopt three assumptions to make the calibration easier to compare with typical DSGE-SOE calibrations: (1) φ=0 in the deterministic steady state (otherwise working capital payments distort factor shares), (2) the collateral coefficient does not bind at the deterministic steady state, and (3) the CRRA coefficient is set to σ=2.

The measure of gross output (y) in Mexican data that is consistent with the one in the model is the sum of GDP plus imported inputs. The data for these variables are available quarterly (at annual rates) starting in 1993. Using data for the period 1993:Q1-2005:Q2, the annualized average ratio of GDP to gross output (gdp/y) is 0.896 and the ratio of imported inputs to GDP (pv/gdp) is 0.114. The average share of imported inputs in gross output is 0.102, hence η=0.102. This factor share, combined with the 0.66 labor share on GDP from Garcia (2005) implies the following factor shares for the production function (13): ⎛⎞0.66 α = ⎜ ⎟ =0.592 and β=1-α-η=0.306.8 ⎝+⎜1(/pv gdp ) ⎠⎟

We also use Garcia’s (2005) estimates of Mexico’s capital stock, together with our measure of y, to construct an estimate of the capital-gross output ratio (k/y) and to set the value of the depreciation rate. He used annual National Accounts investment data for the period 1950-2000 and the perpetual inventories method to construct a time series of the capital-GDP ratio. The average capital-GDP ratio for the 1980-2000 period is 1.88 with a 1980 point estimate of 1.56. Using these annual benchmarks, we constructed a quarterly capital stock series compatible with the quarterly gross output estimates (starting in 1980 because quarterly investment data, again at annual rates, are available as of 1980:Q1). The annualized quarterly capital stock estimates match Garcia’s annual benchmarks by setting

8 The actual share of labor income in GDP is about 1/3 in National Accounts data but Garcia showed that there are measurement problems in separating capital and labor incomes in the National Accounts. Estimating factor shares using household survey data he estimated the labor share at about 2/3rds, which is in line with the usual estimates for the U.S. and other countries.

12 104

the initial capital-GDP ratio to 1.45 and the depreciation rate to 8.8 percent per year. The 1980:Q1-2005:Q2 average of k/y is 1.758. Combined with the 0.088 depreciation rate, this value of k/y yields an average investment-gross output ratio (i/y) of 15.5 percent.

In the deterministic stationary state, imported input prices and the real interest rate take their mean values p and R. The value of p is set equal to the ratio of the averages of the ratios of imported inputs to gross output at current and constant prices, which is 1.028. The mean value of the annual gross real interest rate is derived by imposing the values of β, (i/y), and δ on the Euler equation for capital evaluated at steady state and solving for R. The resulting expression yields R=1+[δ(β-(i/y))]/(i/y)=1.086. A real interest rate of 8.6 percent is relatively high, but in this calibration it represents the implied real interest rate that, given the values of δ and β, supports Mexico’s average investment-gross output ratio as a feature of the deterministic steady state of a standard SOE model. Note also that with this calibration strategy the deterministic steady state also matches Mexico’s average investment-GDP ratio of 17.2 percent.

The model’s optimality condition for labor supply equates the marginal disutility of labor with the real wage, which at equilibrium is equal to the marginal product of labor. ω A This condition reduces to: LFtt=⋅αεexp( ) ( ). Using the logarithm of this expression, our estimate of gross output, and Mexican data on employment growth, the implied value of the exponent of labor supply in utility is ω = 1.846. This value is similar to those typically used in DSGE-SOE models (e.g. Mendoza (1991), Uribe and Yue (2006)).

Since aggregate demand in the data includes government expenditures, the model needs an adjustment to consider these purchases in order for the deterministic steady state to match the actual average private consumption-GDP ratio of 0.65. This adjustment is done by setting the deterministic steady state to match the observed average ratio of government purchases to GDP (0.11), assuming that these government purchases are unproductive and paid out of a time-invariant, ad-valorem consumption tax. The tax is equal to the ratio of the GDP shares of government and private consumption, 0.11/0.65=0.168, which is very close to the statutory value-added tax rate in Mexico. Since this tax is time invariant, it does not distort the intertemporal decision margins and any distortion on the consumption- leisure margin does not vary over the business cycle.

Given the preference and technology parameters set in the previous paragraphs, the optimality conditions for L and v and the steady-state Euler equation for capital are solved as a nonlinear simultaneous equation system to determine the steady state levels of k, L, and v. Given these, the levels of gross output and GDP are computed using the production function and the definition of GDP, and the level of consumption is determined by multiplying GDP times the average consumption-GDP ratio in the data. The value of γ follows then from the steady-state consumption Euler equation, which yields ln(R ) γ ==0.0166 . As is typical in calibration exercises with SCU preferences ln(1 +−cLω−1 ω ) (see Mendoza (1991)), the value of the time preference coefficient is very low, suggesting that the “impatience effects” introduced by the endogenous rate of time preference have negligible quantitative implications on business cycle dynamics. Finally, the steady-state foreign asset position follows from the budget constraint (eq. (2)) evaluated at steady state. This implies a ratio of net foreign assets to GDP of about -0.86.

13 105

Next we calibrate the stochastic process of the exogenous shocks and compute Mexico’s business cycle moments. Table 2 summarizes key features of Mexico’s business cycles and the Sudden Stop of 1995. The Table provides indicators of business cycle variability, co- movement and persistence of macroeconomic time series using the Hodrick-Prescott filter to detrend the data. The Table also reports moments for estimates of the model’s three exogenous shocks. TFP shocks are measured as the cyclical component of a TFP estimate constructed using the production function (13), together with the capital stock and gross output estimates discussed earlier, the calibrated factor shares, and observed data on L and v (see Mendoza (2006) for details). The price shocks are deviations from trend of the relative price of imported inputs, defined as the deflator of imported inputs divided by the exports deflator (so as to remove effects from changes in the nominal exchange rate or in nontradables prices). Interest rate shocks are the cyclical component of Uribe and Yue’s (2006) measure of Mexico’s real interest rate in world capital markets.

The business cycle moments reported in Table 2 are in line with well-known business cycle facts for emerging economies: Investment is more variable than GDP, private consumption is also more variable than GDP (although nondurables consumption is less variable than GDP), all variables exhibit positive first-order autocorrelations, consumption and investment are positively correlated with GDP and the external accounts are negatively correlated with GDP. In addition, the Table shows that both imported inputs and equity prices are significantly more variable than GDP and procyclical.

The model’s exogenous shocks follow a joint Markov process that approximates their time-series processes in the data. In the data, εA, εR and εP follow stationary AR(1) processes nearly independent of each other, except for a statistically significant, negative correlation between εR and εP. Table 2 lists the standard deviations and first-order autocorrelations of the shocks. The correlation between interest rate and TFP shocks is -0.669. Note that the 1995 Sudden Stop coincided with sizable shocks, but we will show below that Sudden Stops are possible in the model even with one-standard-deviation shocks. Also, typical endogeneity caveats apply to our estimates of εR, because of the link between country risk and business cycles, and εA, because of factors that affect measured TFP in addition to imported inputs, such as capacity utilization and factor hoarding. As a result, the “large” TFP and interest rate shocks reported for the 1995 Sudden Stop probably overestimate the true exogenous shocks that occurred that year.

The joint Markov process is a parsimonious chain with two-point realization vectors for each shock. Each realization is set equal to plus/minus one-standard deviation of the corresponding shock. The Markov transition probability matrix is constructed following the simple persistence rule. This imposes the condition that the first-order autocorrelation of the two correlated shocks (εA and εR) be the same, which is very much in line with the data R A since ρ(ε )=0.572 and ρ(ε )=0.537.

Two parameter values remain to be determined: the adjustment cost coefficient a and the working capital coefficient φ. We set these using the Simulated Method of Moments (SMM) so that the model matches the observed ratio of the standard deviation of Mexico’s gross investment relative to GDP (3.6) and a mean ratio of working capital to GDP of 1/5, in a simulation where the collateral constraint does not bind. This yields the values a=2.75 and φ=0.26. This is a reasonable approach to calibrate a because this parameter does not

14 106

affect the deterministic steady state, but it affects the variability of investment. The working capital-GDP target of 20 percent is an approximation to actual data. Data on working capital financing for Mexico are not available, but the 1994:Q1-2005:Q1 average of total credit to private nonfinancial firms as a share of GDP was 24.4 percent. Note, however, that this measure includes financing at all maturities and for all uses, so it overestimates actual working capital financing. On the other hand, these data include the 1995-2002 period in which Mexican banks were being re-capitalized after the 1994 crisis, and credit declined sharply for “abnormal” reasons that bias the average credit-output ratio downwards.

It is also important to note that φ=0.26 is significantly lower than the working capital coefficients used in the DSGE-SOE models of Neumeyer and Perri (2005) and Uribe and Yue (2006). As Oviedo (2004) showed, with low working capital coefficients, the working capital channel has very weak effects on business cycle moments. Hence, the role of working capital in this model is limited to the amplification and asymmetry that it contributes to when the collateral constraint binds. Its effect on regular business cycle volatility is negligible.

4. Results of the Quantitative Analysis

This section reports the results of a quantitative analysis that evaluates the model’s ability to account for the stylized facts of Sudden Stops, and the magnitude of the amplification and asymmetry in the responses of macroeconomic aggregates to shocks induced by the collateral constraint.

4.1 Long Run Business Cycle Moments

The first result we establish is that long-run business cycle moments are largely unaffected by the collateral constraint. To make this point, we compare in Table 3 the business cycle moments of a frictionless economy without collateral constraints (Panel I) with those from two scenarios with different values of κ in which the constraint binds in some states of nature (Panels II with κ=0.3 and III with κ=0.2). These moments are computed using the model’s limiting distribution of k, b, and e in each scenario. The value of κ=0.2 was chosen to match the observed frequency of Sudden Stops (see 4.2 below), and κ=0.3 is shown for comparison.

The moments listed in Panel I show that the model does well at accounting for Mexico’s key business cycle regularities. The model overestimates the variability of GDP (3.9 percent in the model v. 2.7 percent in the data), but scaling by the variability of output the model does a fair job at matching the variability of the other macro aggregates, and the GDP- correlations and first-order autocorrelations are generally in line with the data. Note in particular that the model does well at accounting for three moments that the RBC-SOE literature emphasizes: consumption is more variable than GDP, the interest rate and GDP are negatively correlated, and net exports are countercyclical. Moreover, in contrast with the findings of Garcia, Pancrazi and Uribe (2006), the model does not yield near-unit-root behavior in the net exports-GDP ratio. In fact, it nearly matches the actual first-order autocorrelation of this variable (0. 769 in the model v. 0.797 in the data).

Panels II and III show that long-run business cycles in economies with collateral constraints are very similar to those observed in the frictionless economy of Panel I. The

15 107

credit friction only has large effects on the moments directly influenced by it: the leverage ratio, the ratio of foreign assets to GDP and the net exports-GDP ratio. The means of the leverage ratio and the foreign assets ratio rise, and the mean of the net exports-GDP ratio falls, the variability of the three declines, and all three become more countercyclical.

A key feature of the model behind the result that long-run business cycle moments in Panels II and III do not differ much from those of the frictionless economy in Panel I is the precautionary savings motive. The high-leverage states at which the credit constraint binds are reached after cyclical dynamics in response to sequences of realizations of the shocks lead the leverage ratio to approach its ceiling. Because of the curvature of the constant-relative- risk-aversion period utility function, agents accumulate precautionary savings to self insure against the risk of large consumption collapses in these scenarios. Note that precautionary savings are present even without the collateral constraint, because even with perfect credit markets this DSGE-SOE model has incomplete markets. Hence the average b/gdp ratio with perfect credit markets at about -33 percent is almost 53 percentage points higher than in the deterministic steady state. With the collateral constraint at κ=0.2, the average b/gdp ratio climbs to -10 percent

4.2 Amplification & Asymmetry with the Collateral Constraint

The second result we demonstrate is that the collateral constraint produces significant amplification and asymmetry in the responses of macro-economic aggregates to shocks. To show this result, Table 4 reports amplification coefficients for model simulations with the collateral constraint. The amplification coefficients correspond to differences in the response of each variable in the economy with the collateral constraint relative to the economy with perfect credit markets, in percent of the latter, for a common (k, b, e) triple. Since there is an amplification coefficient for each triple (k, b, e) in the state space, we report averages computed using the model’s ergodic distribution. The Table shows a set of coefficients for Sudden Stop (SS) states, defined in a manner analogous to those used in the empirical literature (e.g. Calvo et al. (2006)). In particular, SS states are those in which the collateral constraint binds (with positive long-run probability) and the net exports-GDP ratio is at least two percentage points above the mean. Non-SS states include all triples in the state space outside the SS set. The long-run probability of hitting SS states and the average debt ratio at which this happens are shown in the last two rows of the Table.

Panel (1) of the Table reports amplification coefficients for the baseline case with κ=0.2. With this upper bound on leverage, the probability of Sudden Stops is 3.3 percent, which matches the frequency of Sudden Stops in the cross-country panel dataset of Calvo et al. (2006). The SS column shows that when the economy hits a Sudden Stop, the collateral constraint amplifies significantly the response of all macroeconomic aggregates to shocks, relative to what is observed in the same (k, b, e) states in the economy without credit frictions. The increased responsiveness of the aggregates ranges from a decline in GDP below trend that is about 1.1 percent larger to a collapse in investment that is almost 12 percentage points larger. Scaling by the cyclical variability of each aggregate listed in Panel III of Table 3, these excess responses imply business cycles that are larger than typical cycles by factors of about 1/3 for GDP to 1.4 for the net exports-GDP ratio.

16 108

The model’s baseline amplification coefficients in Panel (1) of Table 4 are significantly larger than those computed by Kocherlakota (2000). He found that, varying the share of capital from 0.1 to 0.3, the amplification coefficients were small, ranging from 0.15 to 0.35 for output (v. 1.13 in the model) and 0.004 to 0.008 for asset prices (v. 2.9 in the model). As explained earlier, Kocherlakota’s experiments produce weak amplification because they focus in cases in which either the collateral asset is in fixed supply (which weakens the debt- deflation mechanism), or capital can be pledged as collateral up to 100 percent of its value (which under perfect foresight removes the debt-deflation mechanism completely).

The asymmetry of the amplification effects is illustrated by the stark comparison of the amplification coefficients across SS and non-SS columns. In non-SS states, the responses of macro-aggregates are about the same with the collateral constraint as with perfect credit markets, and scaling by the variability of each aggregate the difference across the two economies is negligible. Since Sudden Stop events are low probability events in the long-run, the business cycle moments shown in Table 3 reflect mainly these non-SS states in which there is no amplification due to the credit constraint, and this is consistent with the previous finding showing that the model with the collateral constraint displays business cycle moments very similar to those of the frictionless economy. A corollary of this result is that relatively rare Sudden Stops coexist with the more frequent, normal business cycles summarized in the moments of Table 3. It is also worth noting that the responses in the SS and non-SS columns are produced by shocks that are at most one-standard-deviation in size (as defined in the vector of realizations of the Markov chain), and that the exogenous shocks hitting the economies with and without the collateral constraint in each of the two columns are identical. Thus, the model displays significant amplification and asymmetry in response to shocks that are relatively small, and it has the feature that symmetric shocks produce asymmetric responses, the extreme case of which is a Sudden Stop.

Panels (2) to (5) of Table 4 show that the result indicating that the collateral constraint induces significant amplification and asymmetry in the macroeconomic effects of exogenous shocks is robust to several parameter changes. Panels (2) and (3) report results for κ=0.3 and κ=0.15 respectively. Panel (4) lowers the net exports-GDP threshold ratio used to define Sudden Stops from an increase of two percentage points above the mean to zero. Panel (5) removes working capital financing by setting φ=0.

Increasing (reducing) κ has small effects on the amplification coefficients, but it reduces (increases) the amplification effect on the leverage ratio and the probability of Sudden Stops. Lowering the net exports-GDP threshold to zero weakens the amplification coefficients somewhat, but again the largest effect is on the probability of Sudden Stops, which rises sharply when the threshold used to define them is lowered significantly. Still, in all these scenarios there is significant amplification and asymmetry. In the scenario without working capital, however, the model cannot generate any amplification in GDP and factor allocations, and the probability of Sudden Stops (keeping κ=0.2) is very low. This is because without working capital, factor allocations and output are not affected contemporaneously by the collateral constraint (capital is predetermined and the external financing premium on optimal factor demands is not present, so labor and intermediate goods are not affected by the collateral constraint). The rest of the macro-aggregates continue to display significant amplification and asymmetry, although the amplification coefficients are smaller than in the scenarios shown in the other panels. Thus, these results highlight the importance of the

17 109

collateral constraint limiting access to working capital for the model’s ability to produce significant amplification and asymmetry.

4.3 Can the Model Explain Observed Sudden Stop Dynamics?

The numerical simulations can also be used to evaluate the model’s ability to account for the actual dynamics of Sudden Stop events in Figure 1 reviewed in the Introduction. To this end, we conduct a 10,000-period stochastic time series simulation of the model, and use the resulting artificial data to construct five-year event windows centered on SS events. Figure 2 shows the SS windows for GDP, C, I, Tobin’s Q, and NXY. To match the methodology used in Figure 1, each window includes the median across the SS events identified in the 10,000 period simulation. We also include for comparison + and – one- standard-deviation bands, the actual event window observations from Figure 1, and the observations from Mexico’s 1995 Sudden Stop. To be consistent with Calvo et al.’s (2006) definition of systemic SS events with mild and large output collapses, a Sudden Stop event is identified as a situation in which the collateral constraint binds, output is at least one standard deviation below trend, and the net exports-GDP ratio is at least one standard deviation above trend.

The event windows in Figure 2 show that the model replicates most of the key features of the dynamics of actual SS events, except for the magnitude of the decline in asset prices. The model predicts that Sudden Stops are preceded by periods of economic expansion, with GDP, C and I above trend and NXY running deficits at t-2 and t-1. In the date of the SS events (date t), the model matches very closely the magnitude of the declines in GDP, C, and I. The reversal in NXY between t-1 and t is also very similar to the one in the data, but the levels in the model overestimate those in the data. The model is also consistent with the data in predicting a slow recovery in dates t+1 and t+2. With regard to Tobin’s Q, the model’s dynamics are qualitatively correct, but quantitatively the decline in asset prices is about 40 percent the size of the actual decline. Relative to the Mexican SS event, the model again matches very well the magnitude of the declines in GDP and C at date t, but it underestimates the pre-Sudden Stop boom and the size of the reversal in NXY.

Figure 3 shows event windows for true TFP (i.e. the productivity shock εA) and for the model’s Solow residual, defined assgdpk≡ /( βηαη/(1−− ) L /(1 )). In the baseline scenario with κ=0.2, the two are very similar except on the date of SS events. When Sudden Stops occur, the Solow residual falls more than true TFP. Thus, the model is also consistent with the data in predicting that part of the decline in GDP observed during SS events cannot be accounted for by changes in measured capital and labor, and that this decline in the Solow residual overestimates actual TFP (albeit the difference is not large). However, it is also important to acknowledge that true TFP still has to fall for the output decline to be realistic, and the reason why TFP would fall like this when a Sudden Stop hits remains an open question beyond the scope of this paper.

In summary, the model with the collateral constraint accounts for several key features of Sudden Stops. Large and infrequent recessions take place in response to shocks of standard magnitude when the economy is highly leveraged, and Sudden Stops are nested within normal business cycles. The economy arrives at these high-leverage states with positive long-run probability, and in these states binding collateral constraints cause large

18 110

amplification, asymmetry and persistence in macroeconomic responses to shocks. One weakness, however, is that the decline in asset prices is smaller than the observed collapse, but it is worth recalling that this, as well as all the other Sudden Stop effects studied in this paper, are in response to one-standard-deviation shocks. Larger shocks would trigger larger responses. Moreover, even at ½ the size of the actual price drop, the model generates significantly more asset price amplification than in previous studies (e.g. Kocherlakota (2000)).

4.4 Sensitivity Analysis

Figure 4 compares Sudden Stop event windows for the baseline economy (with κ=0.2) with those of three alternative specifications: (1) the scenario without working capital (φ=0), (2) a simulation with a higher share of imported inputs in production (η=0.2 v. 0.1 in the baseline), and (3) a scenario with a higher value of ω (3 instead of 1.85), which implies a lower labor supply elasticity (0.5 instead of 1.2). The event dynamics observed in actual data are also included for comparison in Figure 4, and Figure 3 includes event windows comparing Solow residuals with true TFP in each of the three sensitivity analysis scenarios. Note that we consider relatively small changes in parameters because otherwise the economies differ sharply in debt and leverage dynamics, and this requires recalibrating κ in order to study the effects of the occasionally binding collateral constraint. In contrast, with the parameter changes we study here, the value of κ remains at 20 percent in all scenarios.

The simulation without working capital performs much worse than the baseline and the other alternatives in terms of its ability to account for observed Sudden Stop dynamics. Without working capital, the amplitude of the fluctuations observed in SS events is significantly smaller, but more significantly, the model fails to produce periods of economic expansion preceding Sudden Stops, as GDP, C and I are already below trend, and NXY is above trend, before the Sudden Stop hits. This occurs because SS events without working capital are preceded by periods of low and declining productivity (see Figure 3), instead of periods of high and increasing productivity as in the baseline. The expectation of declining productivity leads to a substantial decline in I at almost 15 percentage points below trend and drops in C and I of about 2 percentage points below trend by t-1, and this results in a sharp increase in the trade surplus to about 2.5 percentage points of GDP by the same date. For the same reason, labor and imported inputs fall sharply (instead of risinig) before the Sudden Stop hits, although again because the absence of working capital reduces the amplitude of the economy’s business cycle, the declines in labor and imported inputs at date t are much smaller than in the baseline. The output decline is not smaller at date t because the large decline in I at t-1 reduces the capital stock at t and this enlarges the size of the output drop, which otherwise would be much smaller than in the baseline (in the baseline, I rises at t-1 so the higher capital stock at t contributes to offset the contractionary effect of the declines in labor and imported inputs). Thus, these results reaffirm the previous finding indicating that the collateral constraint limiting access to working capital financing plays a very important role in the model’s performance.

The scenarios with higher imported inputs share and lower labor supply elasticity show that these parameters also play important roles. The shape of the SS dynamics is roughly the same as in the baseline, so the model’s overall performance does not worsen as in the scenario without working capital, but the amplitude of the fluctuations changes. A higher

19 111

share of imported inputs strengthens the production effects of all three shocks present in the model. As a result the declines in GDP, C, working capital, labor, and imported inputs are larger with the higher share of imported inputs, while the dynamics of I and Q are about the same as in the baseline. The fit with the data improves as the drops in output and consumption at date t are nearly a perfect match to those observed in actual SS events. In addition, the simulation with the higher imported inputs share creates a much larger wedge between the Solow residual and true TFP (see Figure 3). An average decline of about 1.2 percent in true TFP when Sudden Stops hit translates into an average decline in the Solow residual that is almost twice as large. Thus, a higher share of imported inputs improves the model’s ability to match observed Sudden Stop dynamics, with larger declines in production, consumption and factor allocations, and with a larger fraction of the output drop accounted for by the Solow residual.

The above results for higher η are important because the baseline calibration value of η=0.1 is probably conservative. Evidence from other countries suggests that imported inputs can have much higher shares. Goldberg and Campa (2006) report ratios of imported inputs to total intermediate goods for 17 industrial countries that vary from 14 to 49 percent, with a median of 23 percent (the ratio for Mexico is about ¼). Moreover, to the extent that domestically produced inputs are substitutes for imported inputs, and purchases of these domestic inputs require working capital financing, the scenario with the higher η is likely to be closer to the one that is empirically relevant, because domestic inputs would respond to a similar amplification mechanism as the one affecting imported inputs.9

The model simulation with lower labor supply elasticity retains the same overall qualitative features of the Sudden Stop events of the baseline simulation: The simulation still produces SS events preceded by periods of expansion and followed by gradual recoveries. With the weakened response of labor supply, however, the amplitude of the fluctuations is smaller, and the gap between true TFP and the Solow residual when the Sudden Stop hits is narrower, so the model does not do as well as the baseline in terms of matching the dynamics observed in actual data. In contrast with what we observed in the exercise that changed the share of imported inputs, lowering the labor supply elasticity does affect the behavior of investment and asset prices, both of which exhibit smaller declines than in the baseline scenario. Thus, these results show that labor supply elasticity of about 1.2, as in the baseline, or higher, is important for the model’s ability to explain observed SS dynamics.

5. Conclusions

This paper shows that the quantitative predictions of an equilibrium business cycle model with an endogenous collateral constraint are consistent with key features of the Sudden Stop phenomenon. The constraint imposes an upper bound on the economy’s leverage ratio by limiting total debt, including working capital loans, not to exceed a fraction of the market value of collateral assets. This constraint only binds in states of

9 Extending the model to include domestic inputs, however, is a challenging task because it requires modeling supply and demand of these inputs with and endogenous price. The model can be modified following Mendoza and Yue (2008) to introduce the two inputs using an Armington aggregator, but the solution algorithm for the setup with occasionally binding, endogenous collateral constraints is harder to solve and runs against the curse of dimensionality.

20 112

nature in which the leverage ratio is sufficiently high, and in turn these high-leverage states are an endogenous outcome of the model’s business cycle dynamics.

The model’s collateral constraint introduces a credit channel with two important distortions: one is in the form of external financing premia affecting the cost of borrowing in one-period debt and within-period working capital loans, and the second is Fisher’s debt- deflation mechanism. This mechanism plays a key role in the ability of the model to explain Sudden Stops. When the leverage ratio is sufficiently high, shocks of standard magnitude that result in RBC-like responses under perfect credit markets trigger the collateral constraint. This causes a fall in investment and equity prices which tightens further the constraint and leads to a spiraling collapse of credit, asset prices and investment, a decline in consumption and a surge in the external accounts. Moreover, the binding credit limit hampers access to working capital, causing a contemporaneous decline in output and factor allocations.

This paper’s quantitative analysis show that the long-run business cycle moments of economies with and without the collateral constraint differ marginally, while the mean responses to one-standard-deviation shocks conditional on Sudden Stop states with positive long-run probability differ sharply across the two economies. Thus, in contrast with findings of previous studies, the collateral constraint produces significant amplification and asymmetry in the responses of macroeconomic aggregates to shocks of standard magnitudes on the same exogenous factors that drive normal business cycles (TFP, interest rates and imported input prices). In addition, because of precautionary saving, Sudden Stops are infrequent events nested within normal business cycles in the stochastic stationary equilibrium. Thus, the model proposed here provides an explanation of Sudden Stops that does not rely on large, unexpected shocks, and integrates a theory of business cycles with a theory of Sudden Stops within the same DSGE framework.

A comparative event analysis of Sudden Stops in the data and in the model shows that the model matches key features of actual Sudden Stops. In particular, Sudden Stops in the model are preceded by periods of economic expansion and external deficits, followed by large recessions and reversals in the external accounts when Sudden Stops hit, and then followed by gradual recovery. Moreover, Solow residuals exaggerate the contribution of true TFP to the Sudden Stops’ output drop. These results are robust to variations in the labor supply elasticity and the share of imported inputs in production. In contrast, the assumption that the collateral constraint limits access to working capital financing plays an important role.

An interesting extension of this framework would be to study a setup with “liability dollarization,” in which foreign debt is denominated in a hard currency (i.e. tradable goods) but largely leveraged on assets and/or incomes in domestic currency and generated by nontradables industries. This is important to consider because Sudden Stops also featured large drops in the relative price of nontradables, and in many cases large nominal devaluations (with exceptions like Hong Kong 1998 and Argentina 1995). A large, exogenous devaluation can be viewed as the cause of a Sudden Stop in this situation, but an alternative is to model a debt-deflation mechanism operating through a fall in the relative price of nontradables. Durdu, Mendoza and Terrones (2008) study a model with this feature in a setup without capital accumulation and where the debt limit is a function of income rather than the value of capital.

21 113

The findings of this paper suggest that the key to reducing the probability of Sudden Stops is in promoting the attainment of levels of financial development that weaken the contractual frictions behind collateral constraints. In contrast, taking as given the underlying uncertainty in the form of aggregate shocks to TFP, world interest rates and relative prices, tighter “marked-to-market” capital requirements or “value-at-risk” targets, designed to manage exposure to idiosyncratic risk, can be counterproductive and raise the probability of observing Sudden Stops. Other policy conclusions derived from this analysis relate to financial contagion and the desirability of holding large stocks of foreign reserves. In the setup of this paper, an economy can have solid domestic policies and competitive, open markets, and still reach a point of high leverage at which a Sudden Stop is caused by a relatively small foreign or domestic shock. If waiting for financial development to eliminate this problem seems naïve, and since tighter credit limits can make things worse, self insurance in the form of a sufficiently large stock of reserves can be a useful way of lowering the probability of Sudden Stops. The analysis by Durdu et al. (2008) provides evidence in favor of this argument.

22 114

References

Aiyagari, S. Rao and Mark Gertler (1999), “Overreaction of Asset Prices in General Equilibrium,” Review of Economic Dynamics. Arellano, Cristina (2005), “Default Risk and Aggregate Fluctuations in Emerging Economies,” mimeo, University of Minnesota. Auenhaimer, Leonardo and Roberto Garcia-Saltos (2000), “International Debt and the Price of Domestic Assets, ” IMF Working Paper 00/177. Bernanke, Ben, Mark Gertler, and Simon Gilchrist (1999), “The Financial Accelerator in a Quantitative Business Cycle Framework,” in J. Taylor and M. Woodford eds. Handbook of Macroeconomics, Volume 1C, ed. by, North-Holland. Calvo, Guillermo A., (1998), “Capital Flows and Capital-Market Crises: The Simple Economics of Sudden Stops,” Journal of Applied Economics, v.1, pp 35-54. , and Carmen M. Reinhart (1999), “When Capital Inflows come to a Sudden Stop: Consequences and Policy Options,” mimeo, University of Maryland. , Alejandro Izquierdo, and Rudy Loo-Kung (2006), “Relative Price Volatility under Sudden Stops: The Relevance of Balance-Sheet Effects,”Journal of International Economics, v. 69, pp. 231-254. , Alejandro Izquierdo, and Ernesto Talvi (2006), “Phoenix Miracles in Emerging Markets: Recovering without Credit from Systemic Financial Crises,” Working Paper No. 570, Research Department, Inter-American Development Bank. Choi, Woon Gyu and David Cook (2003), “Liability Dollarization and the Bank Balance Sheet Channel,” Journal of International Economics. Cooley, Thomas, Ramon Marimon and Vincenzo Quadrini (2004), “Aggregate Consequences of Limited Contract Enforceability,” Journal of Pol. Economy, 111(4). Cook, David and Devereux, Michael B., (2006a), “External Currency Pricing and the East Asian Crisis,” Journal of International Economics, v. 69, pp. 37-63. Cook, David and Devereux, Michael B., (2006b), “Accounting for the East Asian Crisis: A Quantitative Model of Capital Outflows in a Small Open Economies,” Journal of Money, Credit and Banking, forthcoming. Dunbar, Nicholas, (2000) Inventing Money: The Story of Long-Term Capital Management and the Legends Behind it, John Wiley & Sons Ltd., West Sussex: England. Durdu, C. Bora, Enrique G. Mendoza, and Marco Terrones, (2008), “Precautionary Demand for Foreign Assets in Sudden Stop Economies: An Assessment of the New Mercantilism,” Journal of Development Economics, forthcoming. Epstein Larry G. (1983), “Stationary Cardinal Utility and Optimal Growth under Uncertainty,” Journal of Economic Theory, 31, 133-152. Fisher, Irving (1933), “The Debt-Deflation Theory of Great Depressions,” Econometrica 1, 337-57. Gertler, Mark, Simon Gilchrist and Fabio M. Natalucci (2007), “External Constraints on Monetary Policy,” Journal of Money, Credit and Banking, v. 39, No. 2-3, pp.295-330. Garcia Cicco, Javier, Pancrazi, Roberto and Martin Uribe (2006), “Real Business Cycles in Emerging Countries?,” NBER Working Paper No. W12629. Garcia-Verdu, Rodrigo (2005), “Factor Shares from Household Survey Data,” mimeo, Research Department, Banco de Mexico. Goldberg, Linda S. and Jose Manuel Campa (2006), “Distribution Margins, Imported Inputs, and the Insensitivity of the CPI to Exchange Rates,” mimeo, Research Department, Federal Reserve Bank of New York.

23 115

Gopinath, Gita (2003), “Lending Booms, Sharp Reversals and Real Exchange Rate Dynamics,” Journal of International Economics. Greenwood, Jeremy, Zvi Hercowitz and Gregory W. Huffman (1988), “Investment, Capacity Utilization & the Real Business Cycle”, American Economic Review, June. Izquierdo, Alejandro (2000) “Credit Constraints, and the Asymmetric Behavior of Asset Prices and Output under External Shocks,” mimeo, The World Bank. June. Jermann, Urban and Vincenzo Quadrini (2005), “Financial Innovations and Macroeconomic Volatility,” mimeo, Wharton School of Business. Hayashi, Fumio (1982), “Tobin’s Marginal q and Average q: A Neoclassical Interpretation,” Econometrica, v. 50, no. 1, 213-224. International Monetary Fund (2002), World Economic Outlook, September. Kiyotaki, Nobuhiro and John Moore (1997), “Credit Cycles,” Journal of Political Economy, v.105, 211-248. Kocherlakota, Narayana (2000) “Creating Business Cycles Through Credit Constraints,” Federal Reserve Bank of Minneapolis Quarterly Review, v. 24, no. 3, Summer, 2-10. Lane, Philip R. and Gian Maria Milesi-Ferretti, (2001), “The External Wealth of Nations: Measures of Foreign Assets and Liabilities for Industrial and Developing Countries,” Journal of International Economics, 55, 263-294. Martin, Phillippe and Helene Rey, (2006), “Globalization and Emerging Markets: With or Without Crash?,” American Economic Review, forthcoming. Mendoza, Enrique G. (2006), “Endogenous Sudden Stops in a Business Cycle Model with Collateral Constraints: A Fisherian Deflation of Tobin's Q.” NBER WP No. 12564 , (2002), “Credit, Prices, and Crashes: Business Cycles with a Sudden Stop.” In Preventing Currency Crises in Emerging Markets, Frankel, Jeffrey and Sebastian Edwards eds. Chicago: University of Chicago Press. , (1995), “The Terms of Trade, The Real Exchange Rate and Economic Fluctuations,” International Economic Review, v. 36, 101-137, February. , (1991), “Real Business Cycles in a Small Open Economy,” American Economic Review, v. 81, 797-818, September. , and Katherine A. Smith (2006), “Quantitative Implications of a Debt-Deflation Theory of Sudden Stops & Asset Prices,” Journal of Int. Econ. , and Vivian Z. Yue (2008), “A Solution to the Default Risk-Business Cycle Disconnect,” NBER Working Paper no. 13861. Milesi-Ferretti, Gian Maria, and Assaf Razin (2000), “Current Account Reversals and Currency Crises: Empirical Regularities.” In Currency Crises, Paul Krugram ed. Chicago: University of Chicago Press. Neumeyer, P. Andres and Fabrizio Perri (2005), “Business Cycles in Emerging Economies: The Role of Interest Rates,” Journal of Monetary Economics, forthcoming. Oviedo, P. Marcelo (2004), “Intermediation of Capital Inflows: The Macroeconomic Implications of Neoclassical Banks &Working Capital,” mimeo, Iowa State University. Schmitt-Grohe, Stephanie and Martin Uribe (2002), “Closing Small Open Economy Models,” Journal of International Economics. Uribe, Martin and Zhanwei Vivian Yue, (2006), “Country Spreads and Emerging Countries: Who Drives Whom?,” Journal of International Economics, v. 69, 6-36.

24 116

Table 1: Calibrated Parameter Values Paramaters set with ratios from data and deterministic steady state conditions α 0.592 labor share set to yield 0.66 share in GDP as α/(1−η)

β 0.306 capital share set to yield 0.33 share in GDP as β/(1−η) δ 0.088 depreciation rate from perpetual inventories method

R 1.0857 implied by s.s.optimal investment rule

w 1.846 regression estimate using labor supply optimality condition γ 0.0166 implied by s.s. consumption Euler eq.

b/gdp -0.86 implied by s.s. budget constraint

Average ratios from Mexican data (1993-2005)

η= pv/y 0.102 imported inputs/gross output ratio

k/y 1.758 capital/gross output ratio

pv/gdp 0.114 imported inputs/gdp ratio gdp/y 0.896 gdp/gross output ratio

c/gdp 0.65 consumption/gdp ratio

g/gdp 0.110 gov. purchases/gdp ratio

i/gdp 0.172 investment/gdp ratio

g/c 0.168 ratio of public to private consumption Parameters set with SMM a 2.75 targeted to match ratio of s.d. of investment to s.d. of gdp

φ 0.2579 targeted to yield a mean working capital/gdp ratio of 0.2 117

Table 2. Mexico: Business Cycle Statistics and the Sudden Stop of 1995

standard standard dev. correlation first-order Sudden Stop variable deviation relative to GDP with autocorrelation (date in brackets) GDP

GDP 2.723 1.000 1.000 0.749 -8.315 (1995:2) intermediate goods imports 7.850 2.882 0.905 0.759 -27.229 (1995:2) private consumption total 3.397 1.247 0.895 0.701 -8.175 (1995:3) non durables & services 2.490 0.914 0.893 0.676 -5.649 (1995:2) investment 9.767 3.586 0.944 0.816 -30.074 (1995:3) net exports-GDP ratio 2.109 0.775 -0.688 0.797 4.898 (1995:2) current account-GDP ratio 1.560 0.573 -0.754 0.720 3.838 (1995:2) equity prices 14.648 5.379 0.570 0.640 -27.397 (1995:2) intermediate goods prices 3.345 1.228 -0.377 0.737 5.915 (1995:1) world real interest rate 1.958 0.719 -0.590 0.572 6.752 (1995:2) total factor productivity 1.340 0.492 0.519 0.537 -5.082 (1995:2)

Note: The data were expressed in per capita terms, logged and detrended with the Hodrick-Prescott filter. Equity prices are in units of the GDP deflator. Intermediate goods prices are defined as the ratio of the deflator of imported intermediated goods divided by the exports deflator. "Sudden Stop" corresponds to the lowest deviation from trend observed in the corresponding variable (for variables in GDP ratios it is the largest change in percentage points observed in two consecutive quarters). The world real real interest rate is the sum of the return on 3-month U.S. T bills plus the EMBI+ spread for Mexican sovereign debt minus a measure of expected U.S. CPI inflation (see Uribe and Yue (2005) for details). Total factor productivity is measured using a production function for gross output that includes capital, labor and imported intermediate goods. The data are for the period 1993:1-2005:2, except the Uribe-Yue real interest rate, which is for the period 1994:1-2004:1. 118 Table 3. Long-Run Business Cycle Moments Standard Standard correlation first-order variable mean deviation deviation with autocorrelation (in percent) relative to GDP GDP I. Economy without Collateral Constraint gdp 390.135 3.90% 1.000 1.000 0.822 c 263.152 4.21% 1.080 0.861 0.817 i 66.203 13.85% 3.552 0.616 0.493 nx/gdp 0.042 3.00% 0.769 -0.191 0.549 k 752.270 4.39% 1.125 0.756 0.962 b/gdp -0.326 17.57% 4.505 -0.023 0.175 q 1.000 3.33% 0.854 0.379 0.440 leverage ratio -0.266 8.32% 2.133 0.001 0.083 v 42.247 5.85% 1.501 0.830 0.776 working capital 75.993 4.32% 1.107 0.995 0.801 Savings-investment correlation 0.539 GDP-world interest rate correlation -0.665 GDP-int. goods price correlation -0.168

II. Economy with 30% Collateral Coefficient (κ = 0.30)

gdp 389.512 3.96% 1.000 1.000 0.818 c 264.581 4.07% 1.030 0.909 0.792 i 66.093 13.66% 3.453 0.628 0.492 nx/gdp 0.036 2.76% 0.697 -0.213 0.490 k 751.015 4.39% 1.109 0.751 0.963 b/gdp -0.257 12.57% 3.177 -0.071 0.124 q 1.000 3.28% 0.828 0.390 0.438 leverage ratio -0.232 5.86% 1.482 -0.043 0.058 v 42.128 6.02% 1.523 0.836 0.770 working capital 75.777 4.51% 1.139 0.990 0.785 Savings-investment correlation 0.512 GDP-world interest rate correlation -0.657 GDP-int. goods price correlation -0.173

III. Economy with 20% Collateral Coefficient (κ = 0.2)

gdp 388.339 3.85% 1.000 1.000 0.815 c 267.857 3.69% 0.959 0.931 0.766 i 65.802 13.45% 3.496 0.641 0.483 nx/gdp 0.024 2.58% 0.671 -0.184 0.447 k 747.709 4.31% 1.120 0.744 0.963 b/gdp -0.104 8.90% 2.313 -0.298 0.087 q 1.000 3.23% 0.839 0.406 0.428 leverage ratio -0.159 4.07% 1.057 -0.258 0.040 v 41.949 5.84% 1.517 0.823 0.764 working capital 75.455 4.26% 1.107 0.987 0.777 Savings-investment correlation 0.391 GDP-world interest rate correlation -0.645 GDP-int. goods price correlation -0.180 119

Table 4. Amplification and Asymmetry Features of Sudden Stop Events (mean differences relative to frictionless economy in percent of frictionless averages) (1) (2) (3) (4) (5) baseline economy lower collateral coefficient higher collateral coefficient zero net exports κ=0.20 κ=0.30 κ=0.15 threshold no working capital S.S. non S.S. S.S. non S.S. S.S. non S.S. S.S non S.S S.S non S.S states states states states states states states states states states gdp -1.13 -0.11 -1.18 -0.06 -1.21 -0.14 -0.86 -0.06 0.00 0.00

c -3.25 -0.31 -3.17 -0.14 -3.15 -0.42 -2.12 -0.23 -1.54 -0.34

i -11.84 -0.61 -10.73 -0.18 -12.35 -0.91 -7.48 -0.30 -9.71 -1.25

q -2.88 -0.15 -2.64 -0.04 -2.99 -0.22 -1.81 -0.07 -2.53 -0.31

nx/gdp 3.56 0.25 3.32 0.08 3.47 0.34 2.13 0.17 3.11 0.49

b/gdp 3.57 0.25 3.00 0.06 3.60 0.36 2.11 0.18 3.31 0.53

lev. ratio 1.31 0.12 0.89 0.04 1.47 0.18 0.83 0.09 0.90 0.17

L -1.71 -0.16 -1.79 -0.09 -1.83 -0.22 -1.29 -0.10 0.00 0.00

v -3.10 -0.29 -3.21 -0.16 -3.31 -0.40 -2.36 -0.18 0.00 0.00

w. cap. -3.12 -0.29 -3.25 -0.16 -3.34 -0.40 -2.37 -0.18 na na

R 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

p 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

tfp 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

prob. of SS events 3.32%1.07% 3.92% 9.54% 0.07%

b/gdp in SS events -0.21 -0.44 -0.17 -0.20 -0.40

Note: Sudden Stop states are defined as states in which the collateral constraint binds with positive long-run probability and the net exports-GDP ratio is at least 2 percentage points above its mean. 120

Figure 1: Macroeconomic Dynamics around Sudden Stop Events in Emerging Economies (cross‐country medians of deviations from HP trends)

Gross Domestic Product Private Consumption

6.00% 6.00%

4.00% 4.00%

2.00% 2.00%

0.00% 0.00%

‐2.00% ‐2.00%

‐4.00% ‐4.00%

‐6.00% ‐6.00% t‐2t‐1tt+1t+2 t‐2t‐1tt+1t+2

Investment Net exports‐GDP ratio 3.00% 20.00% 15.00% 2.00% 10.00% 1.00% 5.00% 0.00% 0.00%

‐5.00% ‐1.00% ‐10.00% ‐2.00% ‐15.00% ‐20.00% ‐3.00% t‐2t‐1tt+1t+2 t‐2t‐1tt+1t+2

Tobin Q

1.050

1.000

0.950

0.900

0.850

0.800 t‐2t‐1tt+1t+2

Note: The classification of Sudden Stop events in the emerging markets data is taken from Calvo et al. (2006). They define systemic sudden stop events as episodes with mild and large output collapses that coincide with large spikes in the EMBI spread and large reversals in capital flows. Tobin' Q is the ratio of debt outstanding over book value of equity, and it is shown in levels instead of deviation from HP trend. 121

Figure 2 : Sudden Stop Event Windows in Actual Data and Model Simulations (medians of deviations from HP trends)

Gross Domestic Product Private Consumption

0.06 0.08

0.06 0.04 0.04 0.02 0.02 0 0 ‐0.02 ‐0.02

‐0.04 ‐0.04

‐0.06 ‐0.06 t‐2t‐1tt+1t+2 t‐2t‐1tt+1t+2 model +1sd ‐1sd data Mexico model +1sd ‐1sd data Mexico

Investment Net exports‐GDP ratio 0.06 0.3 0.05 0.2 0.04 0.03 0.1 0.02 0.01 0 0 ‐0.1 ‐0.01 ‐0.02 ‐0.2 ‐0.03 ‐0.04 ‐0.3 ‐0.05 t‐2t‐1tt+1t+2 t‐2t‐1tt+1t+2 model +1sd ‐1sd data Mexico model +1sd ‐1sd data Mexico

Tobin Q

1.1

1.05

0.95

0.9

0.85

0.8 t‐2t‐1tt+1t+2 model +1sd ‐1sd data

Note: Events from actual data are as in Figure 1, which uses the definitions from Calvo et al. (2006). Mexican data are for the Sudden Stop of 1995. Sudden Stop events in the model simulations are defined in a manner analogous to Calvo et al, as events in which the collateral constraint binds, output is at least one‐standard‐deviation below trend, and the trade balance‐GDP ratio is at least one‐standard‐deviation above trend. Tobin's Q is shown in levels. 122

Figure 3: Solow Residuals and "True" TFP in Sudden Stop Events (means of deviations from long‐run averages)

Baseline η=0.2

0.01 0.01

0.005 0.005 0 0 ‐0.005 ‐0.005 ‐0.01 ‐0.01 ‐0.015

‐0.015 ‐0.02

‐0.02 ‐0.025 t‐2t‐1tt+1t+2 t‐2t‐1tt+1t+2 Solow residual True TFP Solow residual True TFP

No working capital ω=3

0.01 0.01

0.005 0.005

0 0

‐0.005 ‐0.005

‐0.01 ‐0.01

‐0.015 ‐0.015

‐0.02 ‐0.02 t‐2t‐1tt+1t+2 t‐2t‐1tt+1t+2 Solow residual True TFP Solow residual True TFP 123

Figure 4: Sensitivity Analysis of Sudden Stop Events (medians of deviations from HP trends)

Gross Domestic Product Private Consumption 0.04 0.06 0.03 0.05 0.04 0.02 0.03 0.01 0.02 0 0.01 0 ‐0.01 ‐0.01 ‐0.02 ‐0.02 ‐0.03 ‐0.03 ‐0.04 ‐0.04 ‐0.05 t‐2t‐1tt+1t+2 t‐2t‐1tt+1t+2 baseline no w. cap. eta=0.2 omega=3 data baseline no w. cap. eta=0.2 omega=3 data

Investment Tobin Q 0.2 1.05 0.15 0.1 1 0.05 0.95 0 ‐0.05 0.9 ‐0.1

‐0.15 0.85 ‐0.2 ‐0.25 0.8 t‐2t‐1tt+1t+2 t‐2t‐1tt+1t+2

baseline no w. cap. eta=0.2 omega=0.3 data baseline no w. cap. eta=0.2 omega=3 data

Net exports‐GDP ratio Working capital 0.04 0.03 0.03 0.02 0.01 0.02

0 0.01 ‐0.01 0 ‐0.02 ‐0.03 ‐0.01 ‐0.04 ‐0.02 ‐0.05 t‐2t‐1tt+1t+2 ‐0.03 ‐0.06 t‐2t‐1tt+1t+2

baseline eta=0.2 omega=3 baseline no w. cap. eta=0.2 omega=3 data

Labor Intermediate goods 0.015 0.02 0.01 0.01 0.005 0 0 ‐0.01 t‐2t‐1tt+1t+2 ‐0.005 ‐0.01 ‐0.02 ‐0.015 ‐0.03 ‐0.02 ‐0.04 ‐0.025 ‐0.05 ‐0.03 ‐0.06 t‐2t‐1tt+1t+2 ‐0.07 baseline no w. cap. eta=0.2 omega=3 baseline no w. cap. eta=0.2 omega=3 124

Published on open Democracy News Analysis (http://www.opendemocracy.net)

A crisis-opportunity moment

By Paul Rogers, Created 2008-10-27 14:26

The world's financial nervous breakdown is far from over. The stock-market turmoil continues, the huge debt issues left by the collapse of major institutions remain potentially destabilising, and there could still be major shocks to come (not least in east-central Europe [1]). The reverberations of a crisis that began in the global north are working their way across the global south. These trends indicate that the convulsions of September-October 2008 - from the fall [1] of Lehman Brothers to the emergency bailout-investment model [2] of British prime minister Gordon Brown - are only the prelude to a worldwide economic recession.

The evidence of downturn in what is still quaintly called the "real economy" is accumulating everywhere - with workers in the United States at the sharp end. There, "(employers) are moving Paul Rogers is professor of to aggressively cut jobs and reduce costs in the face of the peace studies at Bradford nation's economic crisis, preparing for what many fear will be a University, northern long and painful recession" (see Neil Irwin & Michael S England. He has been Rosenwald, "Job Losses Accelerate, Signaling Deeper Distress writing a weekly column [2] [3]", Washington Post, 23 October 2008). The scale of current on global security on troubles is indicated by a tale of two months: September 2008 openDemocracy since 26 saw more mass-layoffs in the US than any since September September 2001 2001.

The falling market indices are reinforcing fears in much of the non-west about what is to come. Pakistan, whose reserves of currency are dangerously low, is in discussion over a support-package from the International Monetary Fund which could bring [4] it $5 billion in the short term; the Argentinean stock-market sank [5] by over 16% on 22 October amid the planned takeover of pension-funds by the country's troubled president, Cristina Kirchner [5]; and even China - for so long the most optimistic story in the global economy - is facing difficulties, as economic planners realise how much more must be done to satisfy an expectant population (see Jim Yardley & Keith Bradsher, "Faced with global slump, can China keep up economic growth? [6]", International Herald Tribune, 23 October 2008).

Some on the political left see a renewed opportunity for statist solutions, while others welcome the prospect for effective local responses. But the belief in the possibility of progressive outcomes at all requires a dose of caution, for painful economic circumstances often result in people's arc of concern becoming smaller and less generous (see Andrew Dobson & David Hayes, "A politics for crisis: low-energy cosmopolitanism [6]", 22 October 2008). After all, the world has been here before in the last four decades - twice.

Page 1 of 4 125 Two moments

The first such modern "crisis-opportunity" moment came in the In addition to his weekly early 1970s, when there was a growing awareness of the risk to openDemocracy column, global society from environmental constraints. At the time, the Paul Rogers writes an main concern was not about the then relatively little-understood international security problem of climate change but about pollution, resource- monthly briefing for the depletion and the potential of critical food shortages. The first Oxford Research Group; for United Nations Conference on the Human Environment details, click here [7]. Paul (UNCHE [9]) was held in Stockholm in June 1972; it focused Rogers's most recent book attention on the "limits to growth [10]" debate, and there was is Why We're Losing the much talk of sustainability and appropriate technology. War on Terror [8] (Polity, 2007) - an analysis of the In Britain there was at the time a vigorous if largely middle-class strategic misjudgments of movement towards practical self-sufficiency, reflected in a the post-9/11 era and why a flowering of innovative publications such as the radical science new security paradigm is magazine Undercurrents [11] (1973-84). Alongside this needed energetic current of ideas was awareness of the urgent need to reshape the world economy in favour of the majority world (see "A world in flux: crisis to agency [11]", 16 October 2008).

In the event, plans [12] for a "new international economic order" collapsed as retrenching industrial states focused on their own selfish interests, while stagflation and recession occupied people more than home-insulation and solar-panels. By the start of the 1980s - with two Ronald Reagan and Margaret Thatcher elected to power - there was a new emphasis on diminishing state involvement in the economy (on the grounds that government was "the problem, not the solution") in order to allow the free market to flourish, enrich the deserving....and, eventually, "trickle down" to the poor [13].

The second "crisis-opportunity" moment came in the late 1990s. The anti-globalisation movement, decentralised and amorphous, announced its presence as a new radical force at the World Trade Organisation [14] summit in Seattle in November-December 1999. It struck a chord across the world, far more widely [14] than among its core supporters and sympathisers. The sense of unease among the world's rulers was palpable, starting with the World Economic Forum at Davos that followed; "a crack began to appear in the edifice of Western economic control" (see Losing Control: Global Security in the Twenty-first Century [15], Pluto Press, 2nd edition, 2002).

In the midst of this nervous elite mood, the neo-conservative wave broke in the United States. In its early phase in the first months of 2001, George W Bush implemented a raft of policy changes that revealed that any expectation that his narrow election victory might be the prelude to consensual politics was no more than a pipedream. The new administration refused to ratify the comprehensive nuclear test-ban treaty (CTBT [16]); resisted strengthening the biological and toxin weapons convention (BTWC [17]); allowed the anti-ballistic-missile treaty [18] to wither; opposed [19] commitment to the International Criminal Court; and withdrew [20] from the Kyoto climate-change accords. The "new American century" would be a free-market era with no recognition or acceptance of environmental concerns and little interest in the global socio- economic divide [21].

These changes caused consternation in some European capitals; but there was also a sense that Washington would over-reach itself, and that these early unilateral excesses would not stick. This is where the 9/11 attacks were so significant. The appalling shock [21] of those attacks, just when Washington seemed so confident, massively reinforced the belief that there

Page 2 of 4 126 was only one [22] way forward. Even the Enron and the WorldCom scandals were sidelined as everything became subsumed into the "war on terror" - part of the much wider project of cementing the United States's global leadership (see "The world as a battlefield [22]", 9 February 2006).

The "war on terror" waged in pursuit of this aim can seven years later be seen even more clearly for what it was and is: a deeply counterproductive endeavour that has come to cast doubt on the viability and coherence of the idea of a unipolar world (see "The war on terror: seven years on", parts one [22] and two [22], 2-9 October 2008). On that basis alone, the current financial crisis and deepening recession should be a precious opportunity for a new approach to the world economy that is rooted in a genuine multilateralism (see "Wanted: a new global paradigm [22]", 8 November 2007).

Third chance

A summit of the G20 [23] group of leading industrial and developing countries will be held in Washington on 15 November 2008 - an occasion, therefore, when the existing G8 states will face representatives of emerging economies such as China, India, Brazil and Indonesia around the same table. The agenda and the management of the summit will be vital: it is probable that the event's leading planners [24] will attempt to limit debate to the stabilisation of financial markets, but there is at least a chance that the question of global economic planning and governance will also be aired. Whether the conclusions begin to reflect the realities of the world economy and the needs of the majority of the world's people remains to be seen (see "Bush to host world finance summit [25]", 22 October 2008).

It seems unlikely. At root, the delegates in Washington will be officials from governments representing that 20% or so of the world's population that has since the 1960s consistently forged ahead of the rest (see James Davies, Susanna Sandstrom, Anthony Shorrocks & Edward N Wolff, "The World Distribution of Household Wealth [26]", WIDER Angle, 2/2006 [World Institute for Development Economics Research, Helsinki]). The membership of that elite community is global: it may be 150-million strong in China, 100-million strong in India and over 20-million strong in Brazil, as well as including the majority of the people of north America, western Europe, Japan and Australasia. But this albeit large group is decidedly not the "majority world" and the actions taken by those in officialdom rarely take account of the interests of people in this much larger category.

By coincidence, on the very day that George Bush announced the date of the G20 meeting, a report published by the United Nations Human Settlements Programme (UN-Habitat [27]) on the state of the world's cities [28] concluded that the rising inequalities there were likely to lead to persistent social unrest and the risk of violence (see John Vidal, "Wealth gap creating a social time bomb [29]", Guardian, 23 October 2008). The organisation had earlier confirmed that 2007 was the first year in history that the majority of the world's people were now living in cities; UN- Habitat's new analysis highlights deepening divisions within the world's cities - in wealthy countries such as the United States as well as less-developed ones like India and Brazil (see "A tale of two towns [29]", 21 June 2007).

This is a specific yet fundamental effect of the failure of the global economy to deliver socio- economic justice. It is one that political leaders seldom acknowledge. Even more striking is that these leaders seem to be ignorant of the likely impact of several years of recession on urban populations, which in turn are far more aware of their own marginalisation than was true in the 1980s. The G20 meeting [30] and the other summits to follow carry the risk that policy responses will again be so narrow as to focus on only a minority of the world's people. That will be deeply unethical - and ever more dangerous too.

Page 3 of 4 127 Source URL: http://www.opendemocracy.net/article/a-crisis-opportunity-moment

Links: [1] http://us.ft.com/ftgateway/superpage.ft?news_id=fto102220081432037799&page=1 [2] http://afp.google.com/article/ALeqM5gTmEkk6MQTDztdCUNq5QOHELnPxQ [3] http://mobile.washingtonpost.com/detail.jsp?key=299566&rc=to&p=1&all=1 [4] http://afp.google.com/article/ALeqM5hfJ4iK_jwjPfq81XUmpd3vA7aZRw [5] http://www.abs-cbnnews.com/business/10/22/08/world-stock-markets-reel-recession-fears [6] http://www.iht.com/articles/2008/10/22/business/yuan.php [7] http://www.oxfordresearchgroup.org.uk/paulrogers.htm [8] http://www.polity.co.uk/book.asp?ref=9780745641966 [9] http://www.unep.org/Documents.Multilingual/Default.asp?DocumentID=97 [10] http://www.chelseagreen.com/bookstore/item/limits_to_growth:paperback [11] http://www.intertype.co.uk/undercurrents/index.html [12] http://www.un-documents.net/s6r3201.htm [13] http://sedac.ciesin.columbia.edu/povmap/ [14] http://depts.washington.edu/wtohist/index.htm [15] http://www.plutobooks.com/cgi-local/nplutobrows.pl?chkisbn=9780745319094&main= [16] http://www.ctbto.org/ [17] http://www.opbw.org/ [18] http://www.bits.de/NRANEU/BMD/ABM.htm [19] http://www.amicc.org/usinfo/administration.html [20] http://www.telegraph.co.uk/news/worldnews/northamerica/usa/1328307/US-rejects-global- warming-pact.html [21] http://ucatlas.ucsc.edu/ [22] http://www.brookings.edu/opinions/2007/0221diplomacy_talbott.aspx [23] http://www.g20.org/G20/ [24] http://www.ft.com/cms/s/0/34a9c294-a09b-11dd-80a0-000077b07658.html [25] http://news.bbc.co.uk/1/hi/business/7684740.stm [26] http://www.wider.unu.edu/events/past-events/2006-events/en_GB/05-12-2006/ [27] http://www.unhabitat.org/categories.asp?catid=9 [28] http://www.unhabitat.org/content.asp?cid=5979&catid=5&typeid=6&subMenuId=0 [29] http://www.guardian.co.uk/world/2008/oct/23/population-egalitarian-cities-urban-growth [30] http://www.eurodad.org/whatsnew/articles.aspx?id=3008

This article is published by Paul Rogers, , and openDemocracy.net under a Creative Commons licence. You may republish it free of charge with attribution for non-commercial purposes following these guidelines. If you teach at a university we ask that your department make a donation. Commercial media must contact us for permission and fees. Some articles on this site are published under different terms.

Page 4 of 4 128

NBER WORKING PAPER SERIES

IS POLAND AT RISK OF A BOOM-AND-BUST CYCLE IN THE RUN-UP TO EURO ADOPTION?

Barry Eichengreen Katharina Steiner

Working Paper 14438 http://www.nber.org/papers/w14438

NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge, MA 02138 October 2008

This paper was prepared for the National Report on Poland's Membership in the Euro Area, commissioned by the National Bank of Poland. It was begun while Steiner was visiting UC Berkeley, whose hospitality is acknowledged with thanks. We thank Bernhard Mahlberg for helpful comments. The views expressed herein are those of the author(s) and do not necessarily reflect the views of the National Bureau of Economic Research.

© 2008 by Barry Eichengreen and Katharina Steiner. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source. 129

Is Poland at Risk of a Boom-and-Bust Cycle in the Run-Up to Euro Adoption? Barry Eichengreen and Katharina Steiner NBER Working Paper No. 14438 October 2008 JEL No. F0,F15

ABSTRACT

We ask whether Poland is at risk of the boom-bust problem that has afflicted economies around the time of euro adoption. Our answer, inevitably, is mixed. On the one hand the fact that Poland is an outlier, credit-growth wise, accentuates the danger of a boom if one believes in mean reversion. Our econometrics indicate that the fall in interest rates that will flow from expectations of euro adoption will further feed that boom. On the other hand the fact that interest rates have already converged part way to euro-area levels (and more extensively than in earlier adopters that experienced a sharp fall in rates and a pronounced credit boom), especially in the case of lending to firms, suggests that this shock may be less intense in Poland. And it is certainly conceivable that the same policies and country characteristics (not always visible to the econometrician) that have restrained credit growth in the past may continue to do so in the future. The broader literature also points to two set of factors, the first of which makes the danger of an unsustainable credit boom more immediate, the second of which makes it more remote. In the first category are the continuing limitations of the supervisory framework and the weakness of the finance minister in the budget-making process. In the second are a record of rigorous prudential supervision and the existence of relatively competitive labor markets.

Barry Eichengreen Department of Economics University of California 549 Evans Hall 3880 Berkeley, CA 94720-3880 and NBER [email protected]

Katharina Steiner vienna university of economics and business admini vienna, austria [email protected] 130

Is Poland at Risk of a Boom-and-Bust Cycle in the Run-Up to Euro Adoption?

Barry Eichengreen and Katharina Steiner1

September 2008

1. Introduction

This paper takes as a given that Poland will adopt the euro and asks how it should

manage the transition.2 It considers the boom-bust problem that has afflicted economies around the time of euro adoption. It analyzes why those boom-bust cycles occurred. It explores the consequences. It asks what might have ensured superior outcomes.

Poland is not Portugal, Greece, Ireland, Spain, Estonia, Lithuania or Latvia. It is important, in other words, to avoid mechanical comparisons. Compared to other countries that did or are experiencing credit booms in the run-up to euro adoption, the growth of credit to the private sector has been relatively subdued.3 The increase in housing prices has been relatively

limited. Residential mortgage debt as a percent of GDP remains relatively low.4 The challenge is

that it is not always clear in which direction the differences point. On the one hand, that Poland

has not displayed similar excesses could mean that it has the problem of booming bank lending,

excessive wage growth, and housing-market speculation under control. Firm supervision and

regulation, cautious monetary and fiscal policies, and competitive product and factor markets

may mean that Poland is at less risk of these excesses than its predecessors. On the other hand, it

could simply be that euro adoption has remained sufficiently remote that there has been little

1 University of California, Berkeley and Vienna University of Economics and Business Administration, respectively. This paper was prepared for the National Report on Poland’s Membership in the Euro Area, commissioned by the National Bank of Poland. It was begun while Steiner was visiting UC Berkeley, whose hospitality is acknowledged with thanks. We thank Bernhard Mahlberg for helpful comments. 2 The current plan is for euro adoption in 2011. 3 Some might add the qualifier “until recently.” 4 To avoid misunderstanding, it is important to emphasize that all these are statements about Poland relative to comparator countries.

1 131

reason to expect signs of its effects. But not long from now, as euro adoption draws near, the

same dynamics evident in the early adopters may develop. Indeed, insofar as there is still scope

for movements in prices and quantities, a displacement of Polish markets may occur.

This is not the first paper to consider the credit-boom problem in the run-up to euro

adoption.5 Nor is it the first to consider strategies for managing the transition in Central and

Eastern Europe.6 But it differs from its predecessors in several ways. It does not focus just on member states that experienced pronounced booms in the run-up to euro adoption; it seeks to avoid the problem of selection bias by considering the population of relevant countries. And it looks more closely at Poland’s situation, taking as its focus the structure of its financial markets

and the organization of supervision and regulation.

Section 2 describes the now familiar boom-and-bust scenario. Section 3 then reviews the

experience of other catch-up economies that have adopted the euro. While highlighting the

credit-boom-and-bust cycle experienced in some such countries, it also emphasizes the existence

of heterogeneity within this subset of EU member states and points to the absence of

destabilizing dynamics in some members. This then leads to a discussion of why experience has

varied so widely.

Section 4 through 6 then take a closer look at Poland. Section 4 looks more closely at

credit market and real estate developments. Sections 5 and 6 then ask whether Poland can avoid

the kind of boom-bust cycle that has afflicted other catch-up economies adopting the euro.

Section 5 analyzes Poland’s susceptibility to a credit boom by estimating credit-market dynamics

in a large sample of emerging market economies and using that model to forecast the evolution

of private credit. But forecasting exercises only being as reliable as the assumptions that go into

5 See for example Blanchard (2006), Fagan and Gaspar (2008) and Martin and Schiknecht (2008). Appendix Table A is a summary of contributions to the literature that take the sort of econometric approach that we implement below. 6 On this see Borowski and Brzoz-Brezezina (2004) and Darvas and Szapary (2008).

2 132

them, some uncertainty about the country’s susceptibility nonetheless remains. Section 6

suggests that the outcome will turn on four issues. First, how sharply will interest rates come

down? Second, how effectively will wage discipline be maintained? Third, will the government

be able to resist pressures for increased spending? Fourth, will bank regulators effectively

restrain the impulse for an unsustainable credit boom?

Section 7 summarizes the implications of the preceding analysis.

2. The Boom-Bust Scenario

The standard boom-bust scenario for catch-up economies focuses on real interest rates.

Prior to the adoption of the euro, real and nominal interest rates are high, reflecting capital scarcity and imperfect policy credibility. As accession to the euro area approaches, inflation and nominal interest rates converge toward euro-area levels. Short rates are driven to equality by capital mobility and the ECB’s practice of assigning the short-term sovereign debt instruments of all euro-area member governments to the same liquidity category, implying the same haircut when accepting them as collateral. Convergence of inflation tends to be slower: there is more inertia in product and labor markets, and catch-up economies continue to be characterized by relatively high inflation owing to the Balassa-Samuelson effect.

The consequent decline in real interest rates will stimulate consumption and investment

spending. Households and firms will demand additional credit to finance their spending, and the financial system will respond. The decline in real interest rates will put upward pressure on asset prices, including the price of real estate. Strong demand will make for a buoyant labor market, encouraging workers to escalate their wage demands. With investment up and savings down, the

3 133

current account deficit will widen. Debt service costs having declined, the government will be

tempted to increase spending.

But if the strong growth of real wages persists, international competitiveness will deteriorate. Export growth will slow, and import competition will intensify. As profits are squeezed, firms will cut back on investment, and as growth slows, households will curtail their consumption. As demand falls off, unemployment will rise, and the country will discover that it is saddled with a real overvaluation that can be eliminated only through years of grinding deflation. There will be little scope for using stabilizing policy, since the country lacks monetary independence and the government will have accumulated a substantial debt. The party will be over. The souvenirs will be the memories and the hangover.

A few observations about this story may be helpful. First, the impact effect will depend on the extent of the drop in real interest rates, which will in turn depend on the change in nominal rates and on how far above euro-area levels they were prior to the transition. It will

depend on the continuing inflation differential and on how far behind the euro area the country is

in per-capita-income terms. We should not expect the same real interest rate effect in all cases.

Second, the reaction of households and firms is not entirely irrational. Lower real interest

rates mean positive wealth effects for net foreign debtors. A lower cost of capital will mean

faster growth for capital-scarce economies. It thus makes sense for firms to invest more. It

makes sense for households to increase their consumption in anticipation of higher future

incomes. One would expect to see faster growth in the short run as the economy traverses to a

4 134

higher capital/output ratio.7 One would expect to see transitional current account deficits since the increase in consumption precedes the increase in income.

Third, it is necessary to add disequilibrium dynamics for this temporary acceleration to become an unsustainable boom followed by an extended recession. Real wages have to rise by more than is justified by the increase in the capital stock. Households have to boost their consumption by more than is justified by higher future incomes. The financial system has to increase credit to the public sector by more than is prudent given the fundamentals. The government has to increase its spending by more than is warranted by the decline in debt- servicing costs.8

Putting the point this way not meant to deny these possibilities. To the contrary, it is natural for agents, never having seen this adjustment before, to extrapolate from the present. It is not surprising that they overreact. But there is no reason to expect an equally severe overreaction in all times and places. The extent to which real wages rise and competitiveness deteriorates will depend on the structure of the labor market. The extent of the credit boom will depend on the structure of the financial system and on how it is regulated. Whether fiscal policy is a problem or a solution will depend on the political circumstances of the government. The extent to which everyone extrapolates the present and overreacts will depend on how many other

7 These are presumably the mechanisms through which membership in the EU and the euro area are supposed to give rise to what is known in EU parlance as “convergence” – that is, to closing the gap between per capita incomes in the poorer and richer member stats. 8 Not simply because, like other agents, it gets carried away by the boom but also because it may have been forced to contract public spending or raise taxes by more than is politically sustainable in the period when it was seeking to quality for admission to the euro area, creating a tendency to relax fiscal discipline immediately thereafter – something that the Stability and Growth Pact is designed to address but has not always succeeded in doing in practice.

5 135

catch-up economies have experienced such problems previously and how successfully lessons

are drawn from their experience.9

3. Comparative Experience

Figure 1 shows the behavior of real interest rates in catch-up economies adopting the euro,

other recent entrants to the euro area, and Poland. A sharp drop in real rates (constructed here as

the government bond rate adjusted for concurrent CPI inflation) is evident in Greece, Ireland,

Portugal and Spain. Rates drop from more than five per cent to the neighborhood of zero in all

four cases before moving back up.10 That there has been some reversal is not surprising: zero is

not an equilibrium level for real interest rates. In all four cases the decline in real interest rates

began several years before adoption of the euro. In four cases it persisted for several years

following the change-over.11

It is harder to generalize about the remaining cases. In Slovenia real interest rates had already declined to low levels in 2003, and with the country’s high income there was relatively little Balassa-Samuelson inflation. It is thus not surprising that there is not much evidence of a real interest rate decline as euro adoption drew near. Malta and Cyprus similarly have relatively high incomes and well-developed financial systems. Evidently their relatively small and specialized economies make for volatile real interest rates, complicating inference. In Poland

9 Another way of understanding these points is in the context of exchange-rate-based stabilizations, which are close cousins to euro adoptions. These are episodes where countries bring down inflation by pegging the exchange rate and—hopefully—using that space to implement complementary policies. The decline in inflation is likely to be gradual (though how gradual is a matter of dispute; see Sargent 1986). Interest rates will come down faster. There will be a large capital account surplus and current account deficit as investment surges and flight capital is repatriated. Households will go on a spending spree. Unless fiscal policy is tightened or other measures are taken to damp down demand, there will be an erosion of competitiveness, and eventually boom will turn to bust. This can be thought of as a more extreme version of the same phenomenon considered here. Models of the process include Calvo and Vegh (1999) and Antolia and Buffie (2006). 10 It is not surprising that the timing and, indeed, the shape of this reversal is different in Ireland, given that country’s distinctive industry structure and sensitivity to global high-tech activity. 11 In Portugal the trend was interrupted in 1999-2000.

6 136

real interest rates measured on the same basis as in the other countries have already come down from 6 per cent to less than 4 per cent – that is, half way to a reasonable equilibrium level of 2 per cent. This would seem to imply that the need for a further adjustment and the danger of the associated dislocations, while still there, are less than in the first four cases.

The figures that follow consider the experience of these same countries in the two years prior to euro adoption, in the two years following the changeover, and the two years after that

(see Figures 2 to 4).12 For Poland we use the two most recent years of data at time of writing.

Growth and inflation trajectories are heterogeneous, not surprisingly given that these

variables are affected by myriad other influences in addition to euro adoption.13 Portugal

appears unusual in that its growth rate already slowed in its first two years under the euro, a

pattern that is not evident elsewhere. Either the competitiveness problem was unusually quick to

develop in Portugal, or something else in addition to euro adoption was working to depress

growth.14

One explanation suggested by Blanchard (2006) is that Portugal’s exports have a

relatively low technology content, placing the country squarely in the sights of China.

According to the Monetary Policy Committee of the European System of Central Banks (2005),

some 60 per cent of Portugal’s exports are relatively low tech, compared to 30 per cent for the

euro area. Sustaining output and employment growth thus would have required a decline in unit labor costs relative to other euro area members, where Portugal saw a substantial rise. This explains why Portugal’s experience differed from Ireland’s and Spain’s, but it does not explain why it differed from Greece’s. According to ESCB (2005), Greece had an even higher share of low-tech goods in its exports than Portugal—67 versus 61 per cent—in 2000-1. (In contrast,

12 Data permitting. 13 Notably events elsewhere in the EU and the world. 14 Or both.

7 137

Spain had 41 per cent, Ireland 14 per cent.) Evidently, something else was at work in addition to

the technology content of exports.

Another possibility is that real wages surged ahead even more strongly than in other

catch-up economies, both before and after euro adoption, owing not to market behavior but to

public-sector settlements. Comparing public- and private-sector wages shows that it was the

former that led the increase. In Portugal spending ministries did not face hard budget constraints

owing to the decentralized nature of fiscal policy making. The education ministry could agree to

generous increases in teachers’ pay (teachers being among their constituents) without having to

worry about how to fund them.15 Hallerberg and Wolff (2006) show Portugal as having weaker

budgetary institutions than Ireland and Spain according to their measures of procedural

centralization and agenda-setting power of the finance minister. And with public-sector wages

surging ahead, private-sector salaries followed.

A related hypothesis is that Portugal entered the period with relatively large fiscal deficits

as a result of weak budgetary institutions.16 Even member states with relatively large deficits in

the late 1990s could become founding members of the euro area because they had the ability to

prevent the project from going forward, something that is not true of new EU members seeking

to join the euro area subsequently. And as a relatively small country, Portugal was then subject

to stringent application of the Stability and Growth Pact, its deficit/GDP ratio having risen

further to 4.2 per cent by 2001.17 That member states now seeking to adopt the euro will have to

show greater fiscal discipline in the two preceding years (since they lack leverage to block the

15 In Portugal the relevant minister leads negotiations with the corresponding public-sector union; see European Industrial Relations Observatory (2007). 16 The decentralization of budgetary procedures and the lack of agenda-setting power of the finance minister allowed primary expenditure as a share of GDP to rise from 34 per cent of GDP in 1990-92 to 37 per cent in 1993-5, 38 per cent in 1996-8 and nearly 41 per cent in 1999-2001. Data from Braga de Macedo (2007). 17 Application has become somewhat less stringent as the country’s doldrums have deepened, allowing the budget deficit to widen again.

8 138

process) is reassuring from this point of view. That they may similarly be subject to stringent

application of the Stability and Growth Pact is not unless one is sure that there would otherwise

be a tendency for fiscal policy to run out of control.

4. A Closer Look at Credit and Real Estate Market Developments in Poland

In this section we take a closer look at developments in Poland. Our focus is on credit aggregates and developments in the real estate market.

Since 2003, credit has grown more slowly in Poland than in any other Central and

Eastern European country.18 Figure 5 shows that credit to households and firms followed a broadly similar pattern in Poland and in Central and Eastern Europe as a whole in the 1990s but diverged thereafter. To be sure, even before 2000 there were differences between Poland and the rest of the region: in Poland credit to firms grew unusually fast in 1998; credit to households continued to increase until 2000 whereas it slowed down in the group of comparator countries.

But the most persistent and visible divergences are after the turn of the century. In the CEE-10 as a group, credit to households expanded at rates in excess of 30 per cent per annum and peaking in 2004 at a growth rate of nearly 60 per cent. In Poland, in contrast, the growth of credit to households has remained relatively subdued, hovering between 10 and 20 per cent per annum prior to 2006, when its growth rate reached 30 per cent. Even then, this credit aggregate was growing at a slower path than the CEE-10 average. One explanation for this is that the classification of bank loans as nonperforming was relatively rigorous and stringent in Poland (at least through 2003). Having to classify a relatively large number of loans as nonperforming left

Polish banks cash strapped (Breuss, Fink and Haiss 2004). See also Figure 6. Subsequently,

18Throughout our comparison group is the CEE-10 made up of Hungary, the Czech Republic, Slovakia, Slovenia, Romania, Estonia, Latvia, Lithuania, Croatia and Bulgaria.

9 139

loans previously classified as substandard were reclassified as satisfactory, and credit to

households took off (National Bank of Poland 2004).

Credit to firms has grown more slowly both in Poland and the region as a whole. But the

contrast between Poland and the CEE-10 is equally dramatic. In Poland, credit to firms trended

downward in the first half of the present decade, actually shrinking in 2004-5. This is in contrast

to the CEE-10, where the growth rate was not only positive and significantly higher than in

Poland but also trending upward.19 Investment by Polish firms was therefore relatively subdued.

Firms used a significant portion of their retained earnings to pay off foreign-currency-

denominated liabilities. Possible explanations include the fact that demand growth was not as buoyant as in CEE-10 as a whole, the relatively strict classification of loans as nonperforming, which constrained lending to firms by Polish banks (as described above) and political uncertainty

(including some discussion of the possibility of reversing earlier privatizations). As with credit

to households, there are signs (in the data for 2006-7) that the period of very slow growth of

credit to firms may now be over. Still, experience to date is rather different than in the rest of the

region.

Figure 7 disaggregates not by type of borrower but by currency of denomination. Again,

broadly similar movements (punctuated by temporary divergences like the rapid growth of

foreign-currency denominated credit in Poland in 1998) give way to persistent divergences after

the turn of the century. Between 2002 and 2007, both the domestic-and foreign-currency

components of credit to the private sector grow more slowly in Poland than in the rest of the

19 The Czech and Slovak Republics also experienced negative growth rates of credit to firms at the beginning of the decade, but those growth rates recovered fast.

10 140

region.20 Again, there are hints that the period of relatively subdued credit growth may be

coming to an end. In particular the growth of foreign-currency-denominated credit shoots up in

2006 at the same time that its growth is decelerating in the rest of the region. In 2007, Poland

follows the trend of the comparator group with a decrease in foreign currency credit. The

majority of Polish foreign currency loans are denominated in Swiss francs, but there are also

foreign currency loans in euros and U.S. dollars. The bulk of these foreign currency loans are

housing loans secured by mortgages.21 This suggests where risks may be concentrated going

forward.22

The prices associated with these quantities can be seen in Figures 8-9. Figure 8 does not

point to big differences between Poland and the rest of the region in the cost of domestic- and

foreign-currency-denominated borrowing.23 Figure 9, which again distinguishes households and

firms, is more suggestive. The main change in recent years has been the significant decline in

the cost of loans to Polish firms, which was concentrated in the period 2001-2003. This may have been due to growing competition from foreign banks, whose presence in the Polish market increased significantly in this period. The cost of borrowing has also declined for households, but less dramatically. The result is that borrowing costs for Polish firms are now indistinguishable from those of firms in the euro area. Polish households, in contrast, continue to

face significantly higher costs than their euro-area counterparts. This suggests that if there is

going to be a further drop in interest rates with euro adoption, this will be mainly evident in loans

20 While private credit in foreign currency in CEE-10 increased by 23 percentage points between 2002 and 2005, it declined from 19 percent in 2002 to -3 percent in 2005 in Poland. The only other country besides Poland where the growth of private credit was consistently negative in this period was the Czech Republic. 21 See Polish National Bank (2007). 22 We will have more to say about the Polish housing market below. 23 See Luca and Petrova (2007) for an analysis of drivers of foreign currency credit to firms in CEECs.

11 141

to households, including those for housing-related purposes. Evidently, this is where concerns over the credit-boom-and-bust phenomenon need to focus.

Finally, the real estate market has behaved differently in Poland than the rest of the region

(see Figure 10). In the five years ending in 2007, housing prices rose by less than 2 per cent per annum, slower than anywhere else in Central and Eastern Europe (Egert and Mihaljek 2007).

The behavior of rents is consistent with the relatively subdued behavior of prices: since the turn of the century rents on cooperatively owned properties have essentially been flat in nominal terms, while rents on privately and communally owned properties have risen by about 6 per cent per annum, this in a period when CPI inflation has been averaging 3.5 per cent. But house prices as a share of disposable income are the lowest of any EU country. The share of housing loans in total commercial bank lending to households was lower than in any country but the Czech

Republic.24 The ratio of mortgage debt to GDP as of the end of 2006 was lower than anywhere in the European Union except Romania and Slovenia. The rate of growth of that ratio from 2002 through 2006 was slower than anywhere but Slovenia. Mortgage debt per capita is lower than anywhere but Romania. Loan-to-value ratios on typical new mortgages are as low or lower than anywhere in the EU but Hungary.25 All this said, real estate prices have been frothy in the center of Warsaw, and the fastest growing component of credit to households has been for housing.26

Roughly 40 per cent of those loans are foreign-currency denominated. Again this points to the likely location of risks in the not-too-distant future.

The fact that bank credit to the private sector as a share of GDP remains low—at 30 per cent in 2006 considerably lower than in most of the comparator countries—similarly suggests that there may be scope for a lending surge. In Ireland, Portugal and Spain the ratio of private credit

24 Egert and Mihaljek (2007), Table 2. 25 Data in this and the preceding four sentences are from Miles and Pillonca (2008). 26 IMF (2008b), Figure 1, p.21.

12 142

to GDP was in the range of 45 per cent to 75 per cent prior to the adoption of the euro and rose to more than 100 per cent subsequently. Using earlier data, Cottarelli, Dell’Ariccia and Vladkova-

Hollar (2003) estimated an equilibrium ratio of bank credit to the private sector to GDP for a

country with Poland’s characteristics on the order of 70 per cent. Their estimate should now probably be regarded as a lower bound.27 The size of the disequilibrium thus points to the non-

negligible possibility of a credit boom.

5. An Econometric Analysis of Credit Market Dynamics

One way of addressing the question of whether Poland is likely to experience a credit

boom is to build a model of the determinants of private credit (as a share of GDP) and to

extrapolate on the basis of forecasts of the independent variables. We use a balanced panel of

annual data for 50 middle-income countries including Poland and covering the period 1996-

2006.28 We estimate a model which includes factors driving demand and supply for credit

(building on previous studies such as Cottarelli, Dell’Ariccia and Vladkova-Hollar 2005):

- + - +

PCGDP = ƒ (INTR, RGDPC, INFL, INDX) (1)

where PCGDP is the ratio of private credit to GDP, INTR is the average nominal interest rate,

RGDPC is GDP per capita in constant prices, INFL is the year-on-year CPI inflation rate, and

INDX is a financial openness index measuring the absence of capital controls. (LN in the table

below indicates that a variable is expressed in logs.) Specification tests tell us that the data

27 The estimates in Egert, Backe and Zumer (2006) point to somewhat higher equilibrium levels, although the authors emphasize the uncertainty surrounding their estimates. Those in Kiss, Nagy and Vonnak (2006) suggest somewhat lower equilibrium levels. Details are in Appendix A. 28 The sample is based on the World Trade Organization of middle-income countries. The Czech Republic, Estonia, Hungary, the Slovak Republic and Trinidad and Tobago graduated to the high-income category into 2006 according to the WTO, as did Slovenia in 1997 and Singapore somewhat earlier. The country list and other information related to estimation are in Appendix B.

13 143

should be entered in logs rather than levels; in the one case where this is not clear (that of capital

account openness), we enter the variable both ways. We also include country dummies where

these are needed to pick up shifts in the structural relationship.29

Expected signs of the variables are indicated above where they appear in eqn. 1. We expect the interest rate to enter negatively: as interest rates come down, whether for reasons of euro adoption, because of unrelated capital inflows or otherwise, we expect private credit to boom. Higher levels of per capita income are indicative of economic and financial development and stability conducive to the growth of credit markets and the demand for credit. Low and stable inflation should similarly be conducive to the development of credit markets. Finally we expect a more open capital account (a higher value of INDX) to be associated with a higher private credit ratio insofar as this is indicative of a more liberalized financial environment.

Results are in Table 4. The “a” columns show the estimates when we control for country fixed effects. (The two variants differ by whether capital account openness is entered in levels or logs.) All coefficients are statistically significant and enter with their expected signs; the r- squared is relatively high, and the Hausman test accepts the use of country fixed effects.

Estimates including time-fixed effects (not reported) differ in that the coefficient on INDX is not

significantly different from zero, but this version has a relatively low r-squared and, in any case,

the Hausman test rejects the use of time fixed-effects in favor of period-random effects. We

therefore report instead estimates using period random effects in the “b” columns. The Hausman

test accepts this specification, the distribution of residuals is normal, and the coefficients are all

significant. The one anomaly here, for which we don’t have an explanation, is the negative

29 Most of these are quite intuitive. The list of retained country dummies is Brazil starting in 1998 (currency crisis and inflation stabilization), Colombia starting in 2000 (new reporting system for data), the Dominican Republic starting in 2003 (banking crisis), Indonesia starting in 1999 (currency and financial crisis), Suriname starting in 2002 (new data reporting system), Ukraine starting in 1998 (new and improved data reporting system) and Guyana (reason for the structural break not clear to us).

14 144

coefficient on the capital account openness variable. Finally in the “c” columns we include both

country fixed and time random effects.30 The coefficients are all significant and have the

expected signs. We take this as the preferred specification.31

Using this model we can ask whether Poland is a significant outlier. We construct the fitted value of private credit to GDP for Poland for 2006 and compare it with the actual value.

As expected, the actual value of private credit as a share of GDP (33.3 per cent) is well below the

fitted values shown in Table 5. Our preferred specification in column 2c suggests that the credit/GDP ratio in 2006 should have been 10 percentage points higher than its actual value.

(The country-fixed-effects-based estimates point to slightly smaller discrepancies but are basically compatible. The time random effects estimates, with their anomalous coefficients on the openness index among other things, point to much larger discrepancies.)

Say that the discrepancy disappeared in two years (that the credit/GDP ratio rose by 5

percentage points in each year). With nominal GDP growing at 5 per cent, private credit in

nominal terms would have to be expanding at almost 10 per cent a year.32 This suggests that

there is scope for at least a modest credit boom if Poland converges to predicted levels of credit

in a relatively short period of time. That said, annual rates of growth of nominal credit of 10 per

cent are not as alarming as those seen in some other euro-adopting countries. And a ratio to

30 We also considered a variety of other estimators, such as country random effects and two way random effects, all of which yielded broadly compatible results but none of which were obviously preferred on the basis of the standard specification tests. 31 We also conducted a number of sensitivity analyses of the results. Most of the results were robust to estimating the equation in first differences; only the coefficient on the interest rate showed instability. Estimating the equation for the Central and Eastern European countries again produced similar results, although a smaller sample meant lower levels of precision. In addition, limiting the sample to Central and Easern Europe allowed us to enter a measure of nonperforming loans (as a share of GDP) – data that we do not have for other regions – which turned out to be statistically significant as well. We did some exploration with other explanatory variables – the ratio of public debt to GDP and a measure of corruption/transparency were statistically significant in some specifications. But we stuck with our initial specification for purposes of counterfactual simulation so as to limit the danger of data mining. 32 There are of course second-round effects: the rise in the denominator of the credit/GDP will raise credit growth a bit more according to our estimates.

15 145

GDP of 43 per cent is not as alarming as the ratios in excess of 80 per cent in countries like

Estonia and Latvia in 2006.

A more speculative way of using this model is to assume forecasts for the independent

variables for 2010 and to compute the predicted value of the dependent variable. We extrapolate

the independent variables linearly on the basis of recent growth rates. We then extrapolate the

dependent variable, asking what the credit ratio will look like in 2010 if things continue

unchanged.

If the situation in 2005-06 persists, extrapolation suggests a credit/GDP of 58 per cent, up

from 33 per cent in 2006. Again this points to an annual rate of growth of nominal credit slightly

in excess of 10 per cent.33 If we extrapolate all the independent variables and use our preferred

model (column 2c), we get a predicted ratio for 2010 of 52 per cent. This suggests an annual

average rate of nominal credit growth on the order of 8 ½-9 per cent.

6. Is There a Boom-Bust Cycle in Poland’s Future?

How much at risk is Poland of the kind of boom-bust cycle that has afflicted other catch-

up economies adopting the euro? The answer turns, in our view, on four issues. First, how

dramatically will interest rates come down? Second, will wage discipline be maintained? Third, will the government be able to resist pressure for increased public spending? Fourth and finally, will bank regulators effectively restrain the impulse for an unsustainable credit boom? Our

answer to the first question, “not that dramatically,” is somewhat reassuring. Our answers to the

second, third and fourth questions, alas, are maybe, maybe and maybe.

a) How dramatically will interest rates come down?

33 Again assuming nominal GDP growth of 5 per cent per annum.

16 146

Poland’s real interest rates have already come down half-way from typical catch-up-

economy levels to post-euro-adoption levels, as noted above, both because nominal interest rates

are relatively low and because growth is five per cent, lower than in cases like Ireland in the

1990s. This suggests that the financial impulse will be less than in earlier catch-up economies

adopting the euro. However, as shown in Figure 15, interest rates on loans to households are still

much higher in Poland than in the euro area. The decline in interest rates has been due mainly to

the decline in rates on credit to firms which still leaves scope for a significant impulse from

declining rates on credit to households.

b) Will wage discipline be maintained?

A key question is how labor market institutions will respond. As interest rates come

down with the adoption of the euro and spending increases, will the pressure of demand pass

through into wage settlements, eroding the competitiveness of exports? Evidence on what kind of labor market arrangements encourage appropriate adjustments and which ones are conducive to the kind of problems alluded to in the preceding sentence is notoriously fragile. For what it is worth the hump-shaped hypothesis of Calmfors and Driffill (1988) suggests that both highly decentralized and highly centralized/coordinated systems are likely to work relatively well in this context. In highly centralized and coordinated labor markets, social pacts can be negotiated to restrain the growth of wages in the boom period. In relatively atomistic markets, wages will be free to adjust downward when the boom ends. In intermediate systems where union membership and bargaining coverage are extensive but not encompassing or well coordinated, these happy outcomes are less likely. Different sectoral unions will not internalize the implications of their wage setting for wage setting by other sectoral unions. Leapfrogging will occur. It will be difficult to negotiate a social pact to restrain wage inflation. Similarly, when the boom ends,

17 147

there will be resistance to being the first union to agree to wage reductions. Insiders, who shape union policy, will have different incentives than their unemployed brethren.

Poland appears to be fairly far out in the direction of competitive labor markets.34 Union density is low by EU standards (union coverage somewhat less so, reflecting the ability of the government to extend agreements by employers to non-unionized workers in the same sector when this is a “vital social interest”).35 Employment protection legislation is modest by the

standards of Greece, Spain and certainly Portugal (which had the most restrictive regulations at

the beginning of the present decade).36 Wages in different sectors appear to move differently depending on the particular demand conditions facing them.37 There is a very large effective

number of unions, according to Visser (2004) the largest number of any EU country.

Historically, there has been a relatively low correlation between public sector wages and real

wage growth in Polish manufacturing, which militates against Portugal-style wage inflation driven by public sector settlements.38

All this is reassuring. Indeed, labor market developments in recent years have not been

too bad. While productivity growth and labor force participation may not be all that could be

hoped for, the growth of unit labor costs has been contained, and competitiveness has been

maintained.39 Through 2007 the unit-labor-cost-based real exchange rate was still below 2000

34 Precise rankings differ. That of Visser for 2003 ranks only France and the UK head of Poland in terms of degree of labor market decentralization. See Visser (2004). 35 This is required under the terms of the Labour Code (OECD 2004). 36 See OECD (2004) and Boeri and Garibaldi (2006). 37 See Stockhammer and Onaran (2006). 38 Boeri and Garibaldi (2006) find a lower correlation in Poland and Hungary than in any of the other Central and Eastern European economies (they do not, however, consider the Baltics). 39 This is consistent with the view of Boeri and Garibaldi (2006) that not too much should be made of the slow growth of employment, which reflects the continuing efforts of firms to streamline and reduce labor hoarding.

18 148

levels, an evolution that compares favorably with that in Hungary, the Czech Republic and

(even) Slovakia.40

Less reassuring are two additional observations. First, relatively competitive labor markets are likely to be less good at restraining wage growth during the boom than facilitating

adjustment during the bust. The euro-adoption boom will be real: with lower interest rates,

spending will surge, and with additional demand for domestic goods, labor markets will tighten

and wages will surge. This suggests that Poland will not avoid the boom-bust cycle, although the

bust may be less painful than elsewhere. Second, there is less than full agreement on the hump-

shaped hypothesis. We may know less about how institutional arrangements translate into labor

market outcomes than the preceding discussion suggests.

c) Will the government be able to resist pressure for increased spending?

Among the sources of wage pressure in the boom period are permissive public-sector pay policies. This is a specific instance of the general problem of inadequate fiscal discipline in booms. Theory suggests what kind of fiscal institutions are conducive to the maintenance of fiscal discipline. The budgetary process should be centralized and give the finance minister agenda-setting powers. Parliament should have limited options for disregarding the minister’s deficit target. There should be obstacles to legislative amendments to the budget in mid-year, but the finance minister should have options for restoring balance if the deficit widens.

Hallerberg and von Hagen (2006) consider Poland’s fiscal institutions in this light.41

They give the country poor marks for budget preparation. Although the finance minister has

agenda setting power—he circulates a document specifying the target deficit—spending

40 IMF (2008b), p.6. 41A number of other studies undertake this exercise, including Ylaoutinen (2005), Gleich (2006), and Frabizio and Mody (2006). The advantage of the Hallerberg and von Hagen study is that the authors trace the evolution of Poland’s budgetary institutions over time.

19 149

ministers can respond to his circular with individual budget bids, and their response can lead to

changes in the budget balance figure. Importantly, the final decision on the government’s budget

proposal is made by the full cabinet, not the finance minister, creating common-pool problems of

a sort likely to result in excessive deficits. Poland does better at the legislative stage. Parliament

cannot change the deficit target submitted by the government. The government can call for new

elections if parliament fails to adopt a budget. Finally, the country scores high in terms of

implementation. Changing the budget in mid-course is difficult (a supplementary law is

required), transfers of expenditures across budgetary categories require the approval of the

finance minister, and the finance minister has the power to block expenditures when the deficit

widens unexpectedly. Weighing these considerations, Hallerberg and von Hagen rank Poland

slightly above the Central and Eastern European average, behind Estonia, Latvia and Slovenia

but ahead of the Czech Republic, Slovakia, Lithuania, Bulgaria, Hungary and Romania.42

There is some reassurance in the fact that the three Central and Eastern Europeans coming in ahead of Poland are all strong-currency countries. (One has already adopted the euro, one has a currency board, and one operates within the narrow-band ERM-II.) Overall, Poland comes close to matching their combination of budgetary discipline and flexibility.43 It falls short mainly because of the weakness of the finance minister. Although the finance minister can veto transfers of funds across spending categories and take steps to narrow the deficit when it widens, as noted above, he requires the consent and support of his cabinet colleagues heading up the various spending ministries when formulating the budget, which often results in his being at their

42 The data for these other countries are for 2002, their indicators not having been updated subsequently. Timing appears to explain why the estimates in different studies differ. Thus, there was a new constitution in 1997 and a new Public Finance Act in 1998 that significantly strengthened budgeting institutions in Poland. There also have been changes over time in the power of the finance minister (which was greater in 2001 than 2006), and there are differences in assessment depending on whether indices are constructed on the basis of interviews with policy makers or statutory provisions. See the discussion in Hallerberg and von Hagen (2006), p.36. 43 According to this particular ranking.

20 150

mercy. Polish finance ministers have repeatedly resigned or been dismissed for failing to get

their cabinet colleagues to agree to spending limits. This does not provide reassurance for how

the government will respond to the pressure for public spending created by a euro-accession

boom. The picture, perhaps inevitably, is mixed.

d) Will regulators restrain the impulse for an unsustainable credit boom?

The question here is whether the financial system is well regulated, creating confidence

that boom-and-bust dynamics will continue to be avoided as euro adoption approaches. In

support of a positive answer is the fact that Polish banks are well capitalized.44 There is the fact that the Polish Commission for Banking Supervision (the precursor of the current Financial

Supervisory Authority) promulgated a set of best practices for mortgage-related lending in 2006.

The hope is that this will encourage banks to carefully manage both the rate of growth and composition of their mortgage lending.

It may help to put these issues in context. Financial reform in Poland started with the

Balcerowicz plan in 1990. This plan combined liberalization with macroeconomic stabilization and aimed at creating the legal, economic, financial and administrative conditions needed for transformation to a functioning market economy. Excessive issuance of money was halted, and interest rates were raised to contain inflation.45 The National Bank of Poland (NBP) was prohibited by the parliament from extending long-term credits to the government. Cheap credits and preferential lending by the central bank to state owned firms were curtailed. Supervision of the lending business of commercial banks was reinforced.

44 See e.g. Cottarelli, Dell’Ariccia and Vladkova-Hollar (2003). 45 Among the main challenges was the establishment of a stable, convertible, national currency, the reduction of the high government deficit and external debt burden. Establishment of a demand and supply driven price system, privatization and the establishment of free trade were only some of the additional measures included in the Balcerowicz plan. For more details see Sachs and Lipton (1990).

21 151

Yet the volume of non-performing loans continued to rise during the transition recession

of 1991-2.46 Banks had little incentive to provision against potential loan losses and some had

not accumulated enough capital reserves. In 1992 supervision was intensified to prevent

gambling to survive. Tighter capital requirements were implemented and the NBP acquired the

legal power to enforce capital adequacy and loss provisioning standards. The law also imposed

limits on how much a bank could lend to a single borrower.47 Polish banks were twinned with

Western banks to increase their knowledge of modern banking techniques. Employees were

offered additional training.48 Foreign banks were asked to rehabilitate a private domestic bank in

financial distress when entering the Polish market.

Despite these efforts, the quality of Polish banks’ credit portfolios remained poor (Barisitz

2007). An Enterprise and Bank Restructuring Program (EBRP) was therefore adopted in 1993 to

address the undercapitalization and bad loan problems. Specialized regional banks were obliged

to undergo credit evaluations to qualify for the program. For those who qualified, one-time

recapitalization based on the value of their non-performing loan portfolios at the end of 1991

followed. Banks were also obliged to work out restructuring agreements with bad debtors or

forced bankruptcy reorganization or liquidation of those debtors within a fixed period. This

approach strengthened financial discipline on firms and forced banks to develop and provide

adequate risk assessment capacities.49 Restructuring was extended to the cooperative banking

sector in 1994. Two cooperative banks (PKO BP and PEKAO SA) and the state agricultural

46 See Barisitz (2007). 47 No loan could be for more than 10 percent of capital and total loans to a single borrower could not exceed 15 percent of capital (Mondschean and Opiela 1997). 48 The IMF and the World Bank supported this effort (Mondschean and Opiela 1997). 49 At the same time, the restructuring agreements with bad debtors failed to achieve fundamental changes in the management and operation of non-financial firms. The restructuring agreements which banks signed with debtors dealt primarily with financial conditions and did not address fundamental management or operational changes on the part of the debtor.

22 152

bank (BGZ), were recapitalized in a more centralized approach compared to the above

mentioned program (Barisitz 2007).50

In the mid 1990s, new accounting principles were introduced and a general deposit

insurance scheme was implemented. Bank privatization continued, although the state treasury

retained significant stakes. As a result, unclear property rights hampered bank restructuring. In

1997, banking supervision was reorganized and management processes modernized. A new

Independent Commission for Banking Supervision responsible for identification and decision

making concerning the design of supervisory regulations was established. Executive power remained with the NBP. Foreign owners meanwhile were allowed to control a majority of the equity of banks and the last remaining stakes of the state treasury were sold to private owners.51

Competition in banking increased with foreign owned banks expanding domestic retail business.

Harmonization with EU legislation accelerated with the run-up to EU accession. In 1999 the NBP set two pre-accession priorities within the National Program of Preparations for

Membership in the European Union (NBP, 1999): the adjustment of the NBP for operation within the European System of Central Banks and the harmonization of Polish banking regulations with Community legislation. In 2004, with Poland’s accession to the EU, European legislation on banking, such as the single banking license which aims at facilitating the set-up of branches in different European countries, was introduced in the Polish market. In 2005, banking supervision was again enhanced with the implementation of risk-based consolidated supervision.

In 2006, “Recommendation S” (issued by the Commission for Banking Supervision on

March 15th, 2006) was introduced with the goal of improving the banks’ practices concerning

50 According to Reininger, Schardax and Summer (2001), costs of recapitalization in Poland were lower than in any other CEE country . 51 For more details on privatization and consolidation of the Polish banking sector, again see Mondschean and Opiela (1997).

23 153

credit exposure and strengthening risk management and disclosure to comply with international

practices. The risk weight on housing loans with loan-to-value ratios exceeding 50 percent was

raised to 100 percent. Default risk assessments and regular stress tests for the banks’ mortgage

portfolios were recommended. More information is to be provided to customers to increase their

awareness of risks of foreign-currency borrowing. Banks were advised to offer zloty loans first.

They were instructed not to apply lower customer creditworthiness standards when extending

foreign currency loans. Regulators also warned that further steps to curb foreign currency

lending would be implemented as necessary. All this suggests that vigorous supervision and

regulation limit the danger of an unsustainable credit boom in the run-up to the euro.

At the same time, there are weaknesses in the supervisory framework. Barisitz (2007)

identifies “the often arbitrary and inefficient application of new regulations in Poland which also

reflects lingering deficiencies of the court system. Attaching collateral can be costly.

Information systems on credit histories have room for improvement.” Due to the short lending history in Poland, foreclosure practices remain largely untested. IMF (2007a, p.30) has observed that attempts to clamp down on foreign currency and housing credit growth might simply drive business into the nonbank sector, which would increase supervisory challenges without reducing the associated risks.

7. Conclusions and Recommendations

Financial stability in Poland has rested in recent years on a combination of systematic and idiosyncratic factors that have moderated the rate of credit growth. By systematic factors we mean things like a relatively strict approach to loan classification and other prudential regulations, especially before 2004. By idiosyncratic factors we mean things like the behavior of

24 154

housing prices.52 The result has been to leave credit aggregates below the levels in other Central and Eastern European economies and below the levels one would expect on the basis of the experience of emerging markets generally.

This observation frames the question of whether Poland is at risk of a boom-and-bust-like credit cycle in the run-up to euro adoption. On the one hand the fact that Poland is an outlier, credit-growth wise, accentuates the danger of a boom if one believes in mean reversion. Our econometrics indicate that the fall in interest rates that will flow from expectations of euro

adoption will further feed that boom. On the other hand the fact that interest rates have already

converged part way to euro-area levels (and more extensively than in earlier adopters that

experienced a sharp fall in rates and a pronounced credit boom), especially in the case of lending

to firms, suggests that this shock may be less intense in Poland. And it is certainly conceivable

that the same policies and country characteristics (not always visible to the econometrician) that

have restrained credit growth in the past may continue to do so in the future.

The broader literature also points to two set of factors, the first of which makes the

danger of an unsustainable credit boom more immediate, the second of which makes it more

remote. In the first category are the continuing limitations of the supervisory framework and the

weakness of the finance minister in the budget-making process. In the second are a record of

rigorous prudential supervision and the existence of relatively competitive labor markets.

Thus, while Poland is not doomed to follow other euro adopters that have experienced

disruptive boom-and-bust cycles in the run-up to the euro, neither should this risk be minimized.

Policy makers must remain vigilant as the date of euro adoption approaches, and they must then

52 Housing price developments are of course a consequence as well as a determinant of credit-market developments, although they are affected also by other factors.

25 155

act in response to movements in interest rates, wages, housing prices and credit aggregates signaling the immanence of this danger.

A final somewhat reassuring factor is that Poland’s transition to the euro is unlikely to occur for several more years. By that time a number of additional catch-up economies presumably will have made the move to the single currency. Polish officials are aware of the danger of the credit-boom problem that can accompany entry (Rybinski 2007). The more countries that suffer from it—and the more that demonstrate how it can be successfully averted— the more likely Polish policy makers are to draw appropriate lessons.

26 156

Appendix A: Related Work on Determinants of Credit Growth

Brzoza-Brzezina Kiss et al. Backe et al. Cottarelli et al. Author IMF (2007) (2005) (2006) (2006) (2003) 1998/Jan- 1995/Q4-2003/Q4 1995-2004 1993/Q4-2004/Q4 1973-1996 Time span 2005/Dec Quarterly data Annual data Quarterly data Annual data Monthly data

24 countries from Total of 43 North and Latin CEE-11 Country CEE-8, selected developed and America, PL sample euro countries transition Southeast Asia

economies and Western Europe

Identification of Identification of Identification of Identification of determinants of determinants of determinants of determinants of Identification of private credit private credit private credit Research private credit determinants of development > development > development, the question development and private credit simulation of analysis of the equilibrium the equilibrium development possible loan equilibrium credit/GDP level > credit/GDP levels developments credit/GDP level trend forecasts Dynamic panel, error correction Dynamic panel, VECM for Random effects framework, pooled and fixed Method individual GLS estimation OLS pooled mean effect OLS, countries procedure group estimator, DOLS, MGE IV technique

Aggregated and Change of real Dependent Private Private Real private credit disaggregated disaggregated variable credit/nom. GDP credit/nom. GDP credit/nom. GDP private credit

Dummy variables Nominal short and to account for Change in real Explanatory Real 3-month Real short term long term interest structural breaks average gross variables money market rate interest rate rate of the dependent wages variable GDP/capita (PPS Change in GDP/capita (PPP GDP/capita (PPP Real GDP based); industrial industrial based) based) production production Inflation index to account for Non performing Inflation (CPI) Inflation (CPI) variability and loans threshold effects Change in Bank credit to the Stock of public unemployment public sector debt/nom. GDP rate Liberalization Financial index and index Change in real liberalization on entry policy rate index restrictions Existence of public and private Accounting index registries Housing prices Legal origin

Source: Own compilation.

27 157 Appendix B: Data and Sources

Country sample: 23 Latin American countries: Argentina (AR), Belize (BZ), Bolivia (BO), Brazil (BR), Chile (CL), Colombia (CO), Costa Rica (CR), Dominican Republic (DO), Ecuador (EC), El Salvador (SV), Guatemala (GT), Guyana (GY), Honduras (HN), Jamaica (JM), Mexico (MX), Nicaragua (NI), Panama (PA), Paraguay (PY), Peru (PE), Suriname (SR), Trinidad and Tobago (TT), Uruguay (UY), Venezuela (VE)

6 South-, East-Asian countries: China P.R.: Mainland (CN), Indonesia (ID), Malaysia (MY), Singapore (SG), Thailand (TH)

21 Central, Eastern and Southeastern European countries (CEECs): Albania (AL), Belarus (BY), Bulgaria (BG), Croatia (HR), Czech Republic (CZ), Estonia (EE), Hungary (HU), Latvia (LV), Lithuania (LT), Moldova (MD), Republic of Montenegro, Poland (PL), Romania (RO), Russia (RU), Slovak Republic (SK), Slovenia (Sl), Ukraine (UA), Turkey (TR)

Korea, Albania, Bosnia and Herzegovina (BA), Montenegro, the Republic of Serbia and Turkey were excluded from the sample in order to have a balanced panel.

Time span: 1996 - 2006 Although the short time span suggests using quarterly data, we apply annual data and follow Brzoza- Brzezina (2005, 24) who argues that “since the new Member States have undergone a deep transformation and their time series are not particularly long, models, especially based on quarterly data, are not always of top quality.”

28 158

Appendix C: Variables and Definitions

Private credit: claims on the private sector, stocks Source: IMF (2008b).

Interest rate: lending rate, percent per annum, averages; AR, BO, CL, DO, SV, NI, PE, UY: lending rate (foreign currency, USD), percent per annum, seasonal adjusted Source: IMF (2008b).

GDP Nom.: nominal gross domestic product, flows Source: IMF (2008b); United Nations Statistics Division (2008; SR, 2002-2004); WIIW (2008).

GDP Deflator: GDP deflator at constant prices (2000=100), averages Source: IMF (2008b); Econstats (SR).

Inflation: change in consumer price index, percent per annum, averages, exchange rate index Source: IMF (2008b).

Index of financial liberalization: index measuring the extent of openness in capital account transactions Source: Chinn and Ito (2007).

Dummy variables: Brazil: sturctural break in private credit to GDP (pcgdp), 1998-2006, no reason indicated.

Colombia: structural break in pcgdp, 2000-2006, new reporting system for data.

Dominican Republic: structural break in pcgdp, 2004-2006, no reason indicated.

Guyana: structural break in pcgdp, 1998-2002, no reason indicated.

Indonesia: structural break in pcgdp, 1999-2006, no reason indicated.

Suriname: structural break in pcgdp, 2002-2006, data based on changed and improved classification system of data.

Ukraine: structural break in pcgdp, 1998-2006, changed and improved classification of data. Source: IMF (2008c).

29 159

References

Antolia, Manoj and Edward Buffie (2006), “Exchange-Rate-Based Stabilization, Durables Consumption, and Stylized Facts,” unpublished manuscript, Indiana University (July).

Blanchard, Olivier (2006), “Adjustment within the Euro: The Difficult Case of Portugal,” unpublished manuscript, MIT (February).

Backe, Peter and Tina Zumer (2005), “Developments in Credit to the Private Sector in Central and Eastern European EU Member States: Emerging from Financial Repression – A Comparative View,” Focus on Economic Integration 2, pp.83-109.

Barisitz, Stephan (2007), Banking in Central and Eastern Europe 1980-2006: From Communism to Capitalism, Abingdon: Routledge,

Boeri, Tito and Pietro Garibaldi (2006), “Are Labour Markets in the New Member States Sufficiently Flexible for EMU?” Journal of Banking and Finance 30, pp.1393-1407.

Borowski, Jakub and Michal Brzoz-Brezezina (2004), “Designing Poland’s Macroeconomic Strategy on the Way to the Euro Area,” unpublished manuscript, National Bank of Poland (April).

Braga de Macedo, Jorge (2007), “Competitiveness and Convergence in Portugal,” Working Paper no.4, Ministry of Economies and Innovation (November).

Breuss, Fritz, Fink, Gerhard and Peter Haiss (2004), “How Well Prepared are the New Member States for the European Monetary Union?” Journal of Policy Modeling 26, pp.769-791.

Broza-Brzezina, Michal (2005), “Lending Booms in the New EU Member States: Will Euro Adoption Mater?” ECB Working Paper no. 543 (November).

Calmfors, Lars and John Driffill (1988), “Bargaining Structure, Corporatism and Macroeconomic Performance,” Economic Policy 6, pp.14-61.

Calvo, Guillermo and Carlos Vegh (1999), “Inflation Stabilization and Balance of Payments Crises in Developing Countries,” in John Taylor and Michael Woodford (eds), Handbook of Macroeconomics, Amsterdam: North Holland, vol. 1, pp.1531-1614.

Chinn, Menzie D. and Hiro Ito (2007), “A New Measure of Financial Openness,” unpublished appendix, available at http://web.pdx.edu/~ito/.

Cottarelli, Carlo, Giovanni Dell’Ariccia and Ivanna Vladkova-Hollar (2005), “Early Birds, Late Risers and Sleeping Beauties: Bank Credit Growth to the Private Sector in Central and Eastern Europe and in the Balkans,” Journal of Banking and Finance 29, pp.83-104.

Darvas, Zsolt and Gyorgy Szapary (2008), “Euro Area Enlargement and Euro Adoption Strategies,” Economic Paper no.304, European Commission, Directorate of Economic and Financial Affairs (February).

30 160 Economist Intelligence Unit (2008), Country Data Base, London: EUI.

Econstats (2008), http://www.econstats.com/weo/C152V019.htm.

Egert, Balazs, Peter Backe and Tina Zumer (2006), “Private Sector Credit in Central and Eastern Europe: New (Over) Shooting Stars?” Working Paper no.852, William Davidson Institute (November).

Egert, Balazs and Dubravko Mihaljek (2007), “Determinants of House Prices in Central and Eastern Europe,” Comparative Economic Studies 49, pp.367-388.

Economist Intelligence Unit (2008), EIU country database.

Euromonitor International (2008), GMID database.

European Bank for Reconstruction and Development (2008), Transition Report 2008, London: EBRD.

European Industrial Relations Observatory (2007), “Industrial Relations in the Public Sector,” www.eurofound.europa.eu (nd).

Fabrizio, Stefania and Ashoka Mody (2006), “Can Budget Institutions Counteract Political Indiscipline?” Economic Policy, pp.689-739.

Fagan, Gabriel and Vitor Gaspar (2008), “Adjusting to the Euro,” in Lars Jonung, Christoph Walkner and Max Watson (eds), Building the Financial Foundations of the Euro: Experiences and Challenges, Cheltenham, UK: Edward Elgar, pp.56-83.

Gleich, Holger (2003), “Budget Institutions and Fiscal Performance in Central and Eastern European Economies,” ECB Working Paper no.215 (February).

Hallerberg, Mark and Guntram B. Wolff (2006), “Fiscal Institutions, Fiscal Policy and Sovereign Risk Premia,” Discussion Paper no. 35/2006, Frankfurt: Deutsche Bundesbank.

Hallerberg, Mark and Juergen von Hagen (2006), “Budget Processes in Poland: Promoting Fiscal and Economic Stability,” Warsaw: Program Ernst & Young.

Hsiao, Cheng (2003), Analysis of Panel Data, Cambridge: Cambridge University Press.

International Monetary Fund (2007), “Republic of Poland: Financial Sector Assessment Program - Technical Note – Credit Growth and Financial Stability,” IMF Country Report 01/103 (March).

International Monetary Fund (2008a), “Republic of Poland: Staff Report for the 2007 Article IV Consultation,” www.imf.org (19 March).

International Monetary Fund (2008b), International Financial Statistics, Washington, D.C.: IMF.

International Monetary Fund (2008c), Country reports, various countries, www.imf.org.

31 161 Kiss, Gergely, Marton Nagy and Balazs Vonnak (2006), “Credit Growth in Central and Eastern Europe: Trend, Cycle or Boom?” unpublished manuscript, Magyar Nemzeti Bank (November).

Kraft, Evans. (2005), “Banking Reform in South-East Europe: Accomplishments and Challenges,” in Klaus Liebscher, Josef Christl, Peter Mooslechner and Doris Ritzberger- Grünwald (eds), European Economic Integration and South-East Europe, Cheltenham: Edward Elgar, pp. 351-367.

Luca, Alina and Iva Petrova (2007), “What Drives Credit Dollarisation in Transition Economies?” Journal of Banking and Finance 32/, 858-869.

Martin, Reiner and Ludger Schuknech (2008), “Booms and Busts: Experiences with Internal and External Adjustment,” in in Lars Jonung, Christoph Walkner and Max Watson (eds), Building the Financial Foundations of the Euro: Experiences and Challenges, Cheltenham, UK: Edward Elgar, pp.84-110

Miles, David and Vladimir Pillonca (2008), “Financial Innovation and European Housing and Mortgage Markets,” Oxford Review of Economic Policy 24, pp.145-175.

Mondschean, Thomas and Timoth Opiela (1997), “Banking Reform in a Transition Economy: The Case of Poland,” Economic Perspectives 21/2.

Monetary Policy Committee of the European System of Central Banks (2005), “Competitiveness and Export Performance of the Euro Area,” Occasional Paper no.30, Frankfurt: ECB (June).

National Bank of Poland (1999), Annual Report, www.nbp.pl.

National Bank of Poland (2001), The Polish Banking System in the Nineties, Warsaw: National Bank of Poland.

National Bank of Poland (various years), Financial Stability Report, www.nbp.pl.

National Bank of Poland (2007), “Information Taken From a Survey Conducted for a Project on Foreign Currency Credit in CEECs,” Research Institute for European Affairs, Vienna University of Economics and Bank Austria.

National Bureau of Statistics China (2008), statistical data, http://www.stats.gov.cn/english/statisticaldata/

National Bureau of Statistics Guyana (2008), statistical data, http://www.statisticsguyana.gov.gy.

OECD (2004), “Wage-Setting Institutions and Outcomes,” OECD Employment Outlook, pp.127- 181.

OeNB (2007), “Data on Credit Development in CEECs by the Austrian National Bank,” Vienna: OeNB.

Ötker-Robe, Inci, Zbivniew Polański, Barry Topf and David Vávra (2007), “Coping with Capital Inflows: Experiences of Selected European Countries,” IMF Working Paper no. 190 (July).

32 162 Reininger, Thomas, Franz Schardax and Martin Summern (2001), “The Financial System in the Czech Republic, Hungary and Poland after a Decade of Transition,” Discussion Paper 16/01, Economic Research Centre of the Deutsche Bundesbank.

Rosenberg, Christof (2007), “Euro Adoption: What Have We Learned So Far?” unpublished manuscript, IMF (June), www.imf.org.

Rybinski, Krzysztof (2007), “The Euro Adoption – Assessing Benefits and Costs,” speech at the American Chamber of Commerce, Warsaw (17 January), www.bis.org.

Sachs, Jeffrey and David Lipton (1990), “Poland’s Economic Reform,” Foreign Affairs 69/3, pp. 47-66.

Sargent, Thomas (1986), Rational Expectations and Inflation, New York: Harper Collins.

Schadler, Susan, Zuzana Murgasova and Rachel van Elkan (2004), “Credit Booms, Demand Booms and Euro Adoption,” paper presented to the conference on Challenges to Central Banks in an Enlarged EMU, Vienna, 20-21 February.

Statistics Singapore (2008), http://www.singstat.gov.sg.

Stockhammer, Engelbert and Ozlem Onaran (2006), “National and Sectoral Factors in Wage Formation in Central and Eastern Europe,” Working Paper no. 100, Department of Economics, Vienna University of Economics (December).

Tuma, Zdenek (2004), “Euro Adoption and Financial Stability,” BIS Quarterly Review 8, pp.1-3.

United Nations Statistics Division (2008), Millennium Development Goals Indicators, New York: United Nations.

Visser, Jelle (2004), “Patterns and Varations in European Industrial Relations,” in European Commission, Industrial Relations in Europe 2004, Brussels: European Commission, pp.11-58.

WIIW (2008), Statistical Data Base, Vienna: Wiener Institut fuer Internationale Wirtschaftsvergleiche.

World Bank (2007), “Credit Expansion in Emerging Europe: A Cause for Concern?” World Bank EU8+2 Regular Economic Report (January).

Ylaoutinen, Sami (2005), “Development and Functioning of Fiscal Framework in the Central and Eastern European Economies,” unpublished dissertation, University of Jyvaskyla.

33 163

Table 1. Growth of credit in CEE-11, 2002-2006

Growth of private credit in percent of nom. GDP Growth of credit to in domestic in foreign general government in total to households to firms currency currency percent of nom. GDP Poland 0.70 8.49 -5.83 1.42 -1.18 7.74 Croatia 3.34 12.81 3.33 7.82 1.63 18.11 Czech 6.62 20.57 0.07 7.80 -1.93 -29.17 Republic Hungary 14.34 24.19 10.15 4.97 29.99 21.66 Slovakia 9.03 21.60 3.71 3.85 11.62 14.36 Slovenia 13.38 12.00 13.93 -4.56 41.65 1.41 Romania 20.35 55.84 10.16 26.81 15.15 na Latvia 21.74 41.24 11.42 8.39 29.30 -9.72 Estonia 24.84 31.50 19.77 27.82 23.96 -2.80 Bulgaria 24.09 39.22 17.58 21.07 28.01 40.74 Lithuania 30.13 53.36 21.65 27.81 31.88 -20.85

Note: Trend growth rate of credit ratios to nominal Gross Domestic Product (GDP); Least square regressions of the natural logarithm variable series on a time trend are used to obtain the trend growth rate of the available observations.

34 164 Table 2. Main Initiatives Affecting the Banking Sector, 1989-2007

1989 Banking Act and Act on the National Bank of Poland Introduction of two-tier banking system: nine regional banks, private banks admitted 1990 January: Balcerowicz plan for Poland’s transformation to a market economy 1991 Recapitalization of banks to cover losses from currency devaluation (based on Balcerowicz plan) 1992 Tightening of banking supervision (capital adequacy and loss provisioning standards) Conditional licensing scheme (foreign bank-financed recapitalization of some small credit institutions) 1993 Enterprise and Bank Restructuring Program (EBRP): decentralized recapitalization scheme for regional state owned banks; initiation of hard budget constraints 1994 Act on restructuring of cooperative banks (PKO BP and PEKAO SA) and BGZ (state agricultural bank) Bank Guarantee Fund (implementation of a deposit insurance scheme) 1995 Implementation of new accounting principles (in accordance with EU guidelines) 1997/ Reorganization of banking supervision: Independent Commission for Banking Supervision 1998 in charge of identifying tasks and taking decisions; executive power remains with the NBP 1999 NBP sets priorities within the national program of preparations for membership in the European Union 2001 Economic slowdown coincides with cost cutting and rationalization measures; 2002 Financial situation of PKO BP and BGZ remains fragile, restructuring measures for both; KBC assists Kredytbank with capital injection 2004 Accession to European Union, European banking regulations valid 2005 Adoption of risk-focused consolidated supervision 2006 Recommendation S for improvement of banks’ credit exposure

Source: Barisitz (2007) and NBP (2001).

35 165 Table 3. Rental Rates in Poland, 1996-2006

Rent of dwelling owned Rent of communal Year by a co-operative or company dwelling in zl/1m2 1996 0,67 0,83 1997 0,89 1,05 1998 1,12 1,33 1999 1,35 1,63 2000 1,61 1,98 2001 1,53 1,83 2002 1,38 2,17 2003 1,38 2,28 2004 1,41 2,41 2005 1,44 2,56 2006 1,50 2,75

Source: Personal correspondence with Central Statistical Office of the Government of Poland

36 166

Table 4. Determinants of Private Credit Developments in 44 Emerging and Transition Economies, 1996 to 2006

Explanatory variables Dependent variable: lnpcgdp

Estimation method (1a) country fixed (1b) time (1c) country (2a) country (2b) time random (2c) country effects random effects fixed and time fixed effects effects fixed and time random effects random effects constant 2.425*** 4.380*** 2.650*** 2.793*** 4.280*** 3.167** (6.569) (28.414) (7.949) (7.594) (27.786) 7.989 Lnintr -0.102** -0.294*** -0.141** -0.110*** -0.299*** -0.16** (-2.456) (-6.884) (-2.489) (-2.588) (-7.197) (-2.476) Lnrgdpc 0.160*** 0.078*** 0.139*** 0.139** 0.086*** 0.109** (5.244) (5.268) (3.019) (4.524) (6.572) (2.071) lninfl, -0.075*** -0.302*** -0.053** -0.080*** -0.301*** -0.074*** (-3.811) (-13.031) (-1.997) (-4.184) (-13.617) (-1.982) lnindex_finliberalization, 0.127*** -0.044** 0.140*** (5.126) (-2.520) (5.883) index_finliberalization 0.040*** -0.016*** 0.036*** (5.710) (-3.259) (3.374) dummy_Brazil -0.353*** 0.346*** -0.381*** -0.336*** 0.354*** -0.359*** (-8.162) (5.845) (-10.631) (-7.676) (6.524) (-8.925) dummy_Colombia -0.283*** -0.109*** -0.313*** -0.300*** -0.105*** -0.350*** (-4.312) (-3.389) (-5.992) (-4.320) (-3.683) (-5.266) dummy_Guyana 0.202*** 0.570*** 0.206*** 0.199*** 0.592*** 0.205*** (9.503) (14.706) (9.793) (9.540) (17.644) (9.489) dummy_Indonesia -0.853*** -0.886*** (-10.241) (-11.229) dummy_Suriname 0.591*** -0.160*** 0.563*** 0.550*** -0.144*** 0.499*** (6.682) (-3.276) (5.054) (5.708) (-3.063) (4.179) dummy_Ukraine 2.192*** -0.270* 2.169*** (10.972) (-1.754) (9.541) Adj. r2 0.951 0.347 0.834 0.952 0.347 0.799

F-Value (sign. of r2) 182.105*** 29.559*** 46.885*** 187.650*** 33.056*** 38.055***

*Static panel data model. GLS estimates with country-fixed or random effects using EViews 5.1. No. of observations 484. t-statistics in parantheses, based on heteroskedasticity-robust standard errors (White cross-section s.e. and cov.; d.f. corrected). Asteriks indicate the significance of the coefficients at the 10% (*), 5%(**) and 1%(***) levels. The Hausman test on fixed effects confirmed the reported estimation results at the 5% level. The Jarque Bera test confirmed normal distribution of the residuals. Dummy variables control for structural breaks in the time series of the dependent variable. See the Appendix for definitions and sources of the variables.

Source: See text.

37 167

Table 5. In-sample Actual and Predicted Values of Private Credit to GDP for Poland, 2006 (in percent)

actual pcgdp (a) predicted value of pcgdp (b) absolute deviation (a-b)

1a 33.29 41.22 -7.93 1b 33.29 59.71 -26.42 1c 33.29 41.85 -8.56 2a 33.29 43.27 -9.99 2b 33.29 60.58 -27.29 2c 33.29 43.61 -10.32

Source: See text.

38 168 Table 6. Linear Extrapolation of Credit/GDP Ratio and Forecast Based on Estimated Relationship and Linear Extrapolation of Independent Variables (in percent) extrapolated pcgdp predicted value of pcgdp (b) absolute deviation (a-b) 2010 (a) 58.28 48.47 9.80 58.28 83.37 -25.09 58.28 49.88 8.40 58.28 52.21 6.07 58.28 85.20 -26.92 58.28 52.40 5.87

Source: See text.

39 169 Figure 1: Real Interest Rates around the Time of Euro Adoption

GREECE IRELAND REAL INTEREST RATES (1996-2007) REAL INTEREST RATES (1996-2007)

7.0 6.0

6.0 5.0

5.0 4.0

4.0 3.0

3.0 2.0

2.0 1.0

1.0 0.0

0.0 -1.0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

10-year Government Bond yield adjusted by inflation. 10-year Government Bond yield adjusted by inflation. Source: IFS Source: IFS

PORTUGAL SPAIN REAL INTEREST RATES (1996-2007) REAL INTEREST RATES (1996-2007)

6.0 6.0

5.0 5.0

4.0 4.0

3.0 3.0

2.0 2.0

1.0 1.0

0.0 0.0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

10-year Government Bond yield adjusted by inflation. 10-year Government Bond yield adjusted by inflation. Source: IFS Source: IFS

SLOVENIA CYPRUS REAL INTEREST RATES (2003-2007) REAL INTEREST RATES (1997-2007) 1.6 25.0

1.4 20.0

1.2 15.0 10.0 1.0 5.0 0.8 0.0 0.6 -5.0 0.4 -10.0 0.2 -15.0 0.0 -20.0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

10-year Government Bond yield adjusted by inflation. 10-year Government Bond yield adjusted by inflation. Source: IFS Source: IFS and Global Financial Database

MALTA POLAND REAL INTEREST RATES (1996-2007) REAL INTEREST RATES (1999-2007) 20.0 6.0 15.0 5.0 10.0 4.0 5.0 3.0 0.0

2.0 -5.0

-10.0 1.0

-15.0 0.0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

10-year Government Bond yield adjusted by inflation. 10-year Government Bond yield adjusted by inflation. Source: IFS and Global Financial Database Source: IFS and Global Financial Database

40 170

Figure 2: Selected Macroeconomic Variables

OUTPUT GROWTH RATE

Percent8

0 Cyprus Greece Ireland Malta Poland Portugal Slovenia Spain Pre-Euro Adoption Post-Euro Adoption Period 3

Data is averaged for two years prior to adoption (pre-euro adoption); two years post euro-adoption; and the period after that (period 3). Source: Eurostat

INFLATION

4 3.5 3 2.5 2

Percent 1.5 1 0.5 0 Cyprus Greece Ireland Malta Poland Portugal Slovenia Spain Pre-Euro Adoption Post-Euro Adoption Period 3

Data is averaged for two years prior to adoption (pre-euro adoption); two years post euro-adoption; and the period after that (period 3). Percent change in CPI Index Source: Eurostat

UNEMPLOYMENT

12 Percent 8

0 Cyprus Greece Ireland Malta Poland Portugal Slovenia Spain

Pre-Euro Adoption Post-Euro Adoption Period 3

Data is averaged for two years prior to adoption (pre-euro adoption); two years post euro-adoption; and the period after that (period 3). Source: IFS

41 171 Figure 3: More Macroeconomic Variables

GROWTH RATE OF EXPORTS

30 26 22 18

Percent 10 6 2 -2 Cyprus Greece Ireland Malta Poland Portugal Slovenia Spain

Pre-Euro Adoption Post-Euro Adoption Period 3

Data is averaged for two years prior to adoption (pre-euro adoption); two years post euro-adoption; and the period after that (period 3). Source: Eurostat

REAL EFFECTIVE EXCHANGE RATE

125

100

0 Cyprus Greece Ireland Malta Poland Portugal Slovenia Spain

Pre-Euro Adoption Post-Euro Adoption Period 3

Data is averaged for two years prior to adoption (pre-euro adoption); two years post euro-adoption; and the period after that (period 3). CPI based; base year is 2000 Source: IFS

FISCAL DEFICIT AS A PERCENT OF GDP

-2 Percent of of GDPPercent -4

-6 Cyprus Greece Ireland Malta Poland Portugal Slovenia Spain

Pre-Euro Adoption Post-Euro Adoption Period 3

Data is averaged for two years prior to adoption (pre-euro adoption); two years post euro-adoption; and the period after that (period 3). Data is for net lending (+)/borrowing (-) of General Government Source: Eurostat

42 172

Figure 4: Yet More Macroeconomic Variables

NET CAPITAL INFLOWS AS PERCENT OF GDP

10% 8% 6% 4% 2% 0% -2% -4% -6% -8% Cyprus Greece Ireland Malta Poland Portugal Slovenia Spain

Pre-Euro Adoption Post-Euro Adoption Period 3

Data is averaged for two years prior to adoption (pre-euro adoption); two years post euro-adoption; and the period after that (period 3). Data is for net flows o financial account. Source: Eurostat

HOURLY LABOR COSTS

16 14

12 10

8 6 4

2 0 Cyprus Greece Ireland Malta Poland Portugal Slovenia Spain

Pre-Euro Adoption Post-Euro Adoption Period 3

Data is averaged for two years prior to adoption (pre-euro adoption); two years post euro-adoption; and the period after that (period 3). Euro-based data. No data available for Ireland. Source: Eurostat

GROWTH IN UNIT LABOR COSTS

-1 Percent -2

-3

-4 Cyprus Greece Ireland Malta Poland Portugal Slovenia Spain

Pre-Euro Adoption Post-Euro Adoption Period 3

Data is averaged for two years prior to adoption (pre-euro adoption); two years post euro-adoption; and the period after that (period 3). No data available for Greece before 2001 (its pre-euro adoption period). Data for Portgual are estmated values. Source: Eurostat

43 173 Figure 5: Growth of Credit to the Private Sector by Debtor (1996–2007)

70.0

60.0

50.0

40.0

30.0 Percent

20.0

10.0

0.0

-10.0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Credit to households, PL Credit to firms, PL Credit to households, CEE-10 Credit to firms, CEE-10

Definition: Loans to households include loans to households and non profit institutions serving households (NPISH); loans to firms include loans to non financial firms and non monetary financial firms; loans to general government.

Source: OeNB (2007)

44 174

Figure 6: Share of Nonperforming Loans in Total Loans, Poland and CEE-10, 1996-2006

15 Percent 10

0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Poland CEE-10

Source: EBRD (2008).

45 175 Figure 7: Growth of Credit to the Private Sector by Currency (1996–2007)

80.0

70.0

60.0

50.0

40.0 Percent 30.0

20.0

10.0

0.0

-10.0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Credit in Zloty, PL Credit in foreign currency, PL Credit in domestic currency, CEE-10 Credit in foreign currency, CEE-10

Definition: Private credit in domestic and foreign currency includes credit to households and firms.

Source: OeNB (2007)

46 176

Figure 8: Nominal Interest Rates on Loans by Currency Denomination (1996–2007)

60.00

50.00

40.00

30.00 Percent

20.00

10.00

0.00 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Domestic currency loans, PL Foreign currency loans, PL Domestic currency loans, CEE-10 Foreign currency loans, CEE-6

Definition: CEE-6 include Bulgaria, Croatia, Estonia, Latvia, Lithuania and Slovenia.Average interest rates on loans denominated in (and/or indexed to) foreign and domestic currency. The data do not distinguish between credit to the private and public sector. Longer time series on interest rates of foreign currency loans are not available.

Source: OeNB (2007), National Bank of Poland (2008).

47 177 Figure 9: Nominal Interest Rates on Loans to the Private Sector by Debtor (1996–2007)

30.0

25.0

20.0

15.0 Percent

10.0

5.0

0.0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Loans to households, PL Loans to firms, PL Loans to households, Euro area Loans to firms, Euro area

Definition: Average annual interest rate on loans.

Source: National Bank of Poland (2008) and Economist Intelligence Unit (2008).

48 178

Figure 10: Comparative Housing Prices, Poland vs. CEE-10 (1996 – 2007)

260

240

220

200

180

160

140

120

100 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Average CEE-10, excl. BG and RO Poland

Definition: The index of housing prices is constructed as a weighted average of index of actual rentals for housing prices, index of imputed rentals for housing prices, index of maintenance and repair of dwellings prices, index of water and miscellaneous domestic services prices, and index of electricity, gas and other fuels prices.

Source: Euromonitor International (2008).

49 179

NBER WORKING PAPER SERIES

FOOLING SOME OF THE PEOPLE ALL OF THE TIME: THE INEFFICIENT PERFORMANCE AND PERSISTENCE OF COMMODITY TRADING ADVISORS

Geetesh Bhardwaj Gary B. Gorton K. Geert Rouwenhorst

Working Paper 14424 http://www.nber.org/papers/w14424

NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge, MA 02138 October 2008

This paper has benefited from comments and suggestions from Martijn Cremers, Ned Elton, Bill Fung, Mila Getmansky, Will Goetzmann, Marty Gruber, Raj Gupta, David Hsieh, Jon Ingersoll, Bing Liang, Jonathan Macey, Roberta Romano, Ken Scott, and seminar participants at UMass Amherst. The views expressed are those of the authors and do not necessarily reflect the official position of AIG Financial Products Corp or the National Bureau of Economic Research.

© 2008 by Geetesh Bhardwaj, Gary B. Gorton, and K. Geert Rouwenhorst. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source. 180

Fooling Some of the People All of the Time: The Inefficient Performance and Persistence of Commodity Trading Advisors Geetesh Bhardwaj, Gary B. Gorton, and K. Geert Rouwenhorst NBER Working Paper No. 14424 October 2008 JEL No. G12,G13

ABSTRACT

Investors face significant barriers in evaluating the performance of hedge funds and commodity trading advisors (CTAs). The only available performance data comes from voluntary reporting to private companies. Funds have incentives to strategically report to these companies, causing these data sets to be severely biased. And, because hedge funds use nonlinear, state-dependent, leveraged strategies, it has proven difficult to determine whether they add value relative to benchmarks. We focus on commodity trading advisors, a subset of hedge funds, and show that during the period 1994-2007 CTA excess returns to investors (i.e., net of fees) averaged 85 basis points per annum over US T-bills, which is insignificantly different from zero. We estimate that CTAs on average earned gross excess returns (i.e., before fees) of 5.4%, which implies that funds captured most of their performance through charging fees. Yet, even before fees we find that CTAs display no alpha relative to simple futures strategies that are in the public domain. We argue that CTAs appear to persist as an asset class despite their poor performance, because they face no market discipline based on credible information. Our evidence suggests that investors' experience of poor performance is not common knowledge.

Geetesh Bhardwaj K. Geert Rouwenhorst Vice President School of Management AIG Financial Products Yale University and Box 208200 Rutgers University New Haven, CT 06520-8200 Department of Economics [email protected] 75 Hamilton Street [email protected]

Gary B. Gorton Yale School of Management 135 Prospect Street P.O. Box 208200 New Haven, CT 06520-8200 and NBER [email protected] 181

1. Introduction

Hedge funds command hefty fees because they allege that they can earn above average risk- adjusted returns, based on their skills. This means that the returns generated by their trading strategies must not be easily replicated by lower cost alternatives such as passive indices, mutual funds or ETFs. According to the Government Accountability Office (2008), based on industry estimates, the number of hedge funds has grown from 3,000 to more than 9,000 between 1998 and early 2007, and their assets under management have grown from $200 billion to more than $2 trillion globally. Investors appear to have concluded that these funds are worthwhile investments.

Are hedge funds worthwhile investments? Do they earn above average risk-adjusted returns? What benchmarks should be used for the risk-adjustment? How should investors determine which funds to invest in? It has proven very difficult to answer these questions, because it is difficult to obtain reliable performance data and to determine the relevant benchmarks. Hedge funds are prohibited from direct advertising, but are allowed to indirectly market themselves by reporting their past (possibly paper) returns to private vendors who then sell this performance information through databases to potential investors, news sources, consultants, and researchers. While past returns may be useful for investment choices, the available hedge fund databases are contaminated by a number of biases that affect the ability of investors to make proper inferences. The academic literature has recognized many of these biases (e.g. selection bias, survivor bias, and backfill bias), but the proposed adjustments are often crude or difficult to implement ex-post.

Even when the performance data is available, the issue remains of how to adjust the returns for risk. It is not clear what benchmarks hedge funds should be evaluated against because they are an extremely heterogeneous group and can employ time-varying, state-contingent, and leveraged strategies. In fact, it is not obvious that hedge funds form an “asset class” since their strategies are so diverse. All they have in common is that they have chosen to organize themselves so as to be exempt from various U.S. legal requirements, hence becoming “hedge funds” – a legal definition of the “asset class.”1 Faced with this heterogeneity problem, the literature on hedge funds is not so much performance analysis as it is a descriptive, positive, analysis of these funds’ returns. The focus has been less on whether funds add value for investors than on empirically characterizing fund strategies.

A central point of our work is that biased data and a lack of benchmarks are problems faced by investors and researchers alike. We separate the question of whether fund managers exhibit skill from the question of whether investors receive positive risk-adjusted returns, by looking at both returns net of fees and estimated gross returns. To the extent that fund managers exhibit skill, we ask how the value added is divided between the funds and its investors. We narrow the set of funds to be evaluated to commodity trading advisors (CTAs). 2 There are four reasons for our choice. First, the strategies that CTAs employ are relatively well-known compared to many hedge fund strategies. CTAs report in surveys that they are trend followers and momentum traders. In a survey in 2000, 75 percent of CTAs responded that they are trend followers and 71 percent

1 Hedge funds and CTAs are organized so as to qualify for exemptions from regulations, and disclosure requirements of certain federal securities laws, including the Securities Act of 1933 and the Securities Exchange Act of 1934. For purposes here, to qualify hedge funds must not advertise to the general public and can only solicit participation in the fund from certain large institutions and wealthy individuals. For details see Hall (2008). 2 We use the term CTA (Commodity Trading Advisors) to refer to the legal form of investment vehicles that trades in futures markets and consequently registers with the U.S. Commodity Futures Trading Commission.

Electronic copy available at: http://ssrn.com/abstract=1279594 182

responded that they used momentum as a signal in their trading approach. 3 Second, the commodity and financial futures comprise a smaller strategy space for CTAs compared to equity hedge funds, event-driven funds, or multi-strategy hedge funds, for example. This simplifies the choice of benchmarks for evaluating CTA performance. Third, the available performance history of CTAs is relatively long. While most major hedge fund databases do not provide samples that are free from backfill bias prior to 1994, there exists an early academic literature on public CTAs in the 1980s in which this bias is effectively eliminated (Elton, Gruber, and Renzler (1987, 1989, 1990)). Finally, although CTAs are a subset of the hedge funds universe, they control a significant amount of assets. While there are no official measures of the size of the CTAs’ money-under-management (MUM), BarclayHedge estimates that as of the end of 2007, MUM was $206.6 billion, having grown from $50.9 billion five years earlier – a 306 percent increase.4 We analyze the performance of all CTAs that voluntarily report to the Lipper-TASS database. To eliminate the influence of various biases induced by strategic returns reporting and database construction, more than 80% of the available observations are excluded.5 We show that these corrections greatly influence inference about CTA performance. We estimate that between 1994 and 2007 the average bias-adjusted CTA returns after fees have been statistically indistinguishable from the average return on an investment in US T-bills. The average CTA has therefore not created value for their investors. This conclusion mirrors the finding by Elton, Gruber and Rentzler (1987, 1989, 1990) (EGR) who – almost two decades ago – found that publicly traded commodity funds did not create positive returns for investors. The combined evidence is therefore one of 20 years without performance. The surprising finding therefore is that the considerable attention that the Elton Gruber and Renzler studies received at the time of publication does not seem to have influenced the ability of CTAs to attract assets.

The poor net returns for investors are not necessarily inconsistent with CTA managers possessing skill. For example, it is possible that managers generate excess returns, but capture the rents of outperformance through charging fees. We present some evidence consistent with this view. Using standard procedures to estimate gross returns (i.e., returns before fees), we estimate that the average CTA return has exceeded T-bills by more than 5 percent per annum between 1994 and 2007, but only by 0.85 percent per annum after fees. In order to evaluate whether these gross excess returns are abnormal, we develop a number of simple performance benchmarks. We find that relative to these benchmarks CTAs display no significant skill (alpha). However, the benchmarks can explain relatively little of the variance of CTA returns. It is difficult to explain variation in ex-post gross returns of CTAs, despite the fact that the majority of funds describe their style as trend-following. A regression of individual fund returns on our benchmarks produces an R-squared below 30% for seven out of ten funds. We show that exposure to simple trend following strategies can explain the most of the average outperformance before fees.

The poor performance track record of CTAs raises the question of why the asset class has continued to grow – apparently despite a long history of poor performance. The supply side of the market is easy: CTAs generate fee income of about 4% on assets under management, which also explains the high rates of entry into a market with high attrition rates. Why investors continue to allocate to CTAs is more difficult to answer. Did investors ignore the conclusions of the EGR papers despite the publicity they received at that time?

3 See Waksman (2000). Academic research is in agreement with these CTA self-assessments. Fung and Hsieh (1997) argue that CTAs have one dominant style factor, namely, trend following. 4 See Hhttp://www.barclayhedge.com/research/indices/cta/Money_Under_Management.htmlH . 5 As explained below, we exclude 83,201 of the 102,393 available monthly observations on fund performance post-1993, and all returns prior to 1994.

2 183

We explore several broad explanations. First, average fund performance may not be sufficient as an overall indicator of the attractiveness of an asset class. To the extent that CTAs offer option- like payoffs that exhibit positive skewness, investors may prefer to allocate to CTAs despite poor average returns. However, our data shows that CTAs are equally likely to exhibit positive or negative skewness. It seems unlikely that CTAs are attractive because of the portfolio properties of their performance. While correlations of managed futures programs with traditional asset classes have historically been low, it seems unlikely that investors would allocate $200 billion to an asset class that offers T-bill returns with a standard deviation that is comparable to equities.

An alternative explanation is that investors are unable to overcome the information asymmetry to properly evaluate CTA performance. Although it is difficult to provide direct evidence, several observations are consistent with this view. For example, when academic researchers do not seem to agree on how to properly adjust CTA track records for various biases introduced by strategic reporting, it seems unlikely that investors who often lack access to comprehensive databases can do a substantially better job. Especially since there is no mechanism to create common knowledge about historical CTA performance – it is difficult to learn from the investment experience of others when information is not aggregated, either through market prices, disclosure, or regulatory oversight. In this context it is illustrative that in response to the EGR studies in the 1980s which revealed poor performance of public commodity funds, the industry has reorganized itself into a form that requires less disclosure and regulatory oversight. And while in theory funds can attempt to signal quality through the contract terms they offer investors, we find no systematic relationship between contract terms and fund performance.

Finally, investors may simply be unaware that there is an information asymmetry and the history of poor CTA performance may not be common knowledge. Such an information setting differs from the failure of Akerlof’s (1970) lemons market, in which it is common knowledge that there is an information asymmetry. It appears that CTAs strategically report their performance data to maintain this information environment. We discuss these issues towards the end of the paper. But, note that it puts researchers in a somewhat delicate position. Simply put, we do not have all the data we would like and the available data must be treated with great care, precisely because of the strategic desires of the CTAs. We alert the reader to these difficulties as we proceed.

The literature most directly related to our work is about CTAs. In addition to the Elton, Gruber and Rentzler (1987, 1989, 1990) papers, our work is closely related to Fung and Hsieh (1997, 2001). Fung and Hsieh (1997) argue that the dominant investment style of CTAs is trend following. Fung and Hsieh (2001) construct dynamic factor portfolios to capture this trend following behavior. We show that while the Fung-Hsieh (FH) factors are useful for style analysis, they are less useful for answering the question of whether CTAs create alpha.6 In particular we show that that the FH factors tend to impound an upward bias in fund alphas, because they are inefficient replications of trend-following styles.

The paper proceeds as follows. In Section 2 we introduce the data set used for this study and briefly discuss the various, well-known, biases that exist in CTA and hedge fund data sets. In

6 There is also a literature on hedge funds. For example, Ackerman, McEnally, and Ravenscroft (1999) analyze hedge funds, comparing hedge fund returns, volatility, and Sharpe ratios to the returns and characteristics of the S&P 500 and eight standard market indices. They conclude that hedge funds outperform mutual funds, but not standard market indices. Brown, Goetzmann and Ibbotson (1999) also look at hedge funds and find little evidence of outperformance. Brown and Goetzmann (2003) used a classification algorithm to group hedge funds into similar styles, which then becomes the benchmark for out-of-sample performance evaluation. There are many other papers (e.g., Brown and Goetzmann (1997)).

3 184

addition we construct a performance index for CTAs net of fees, and estimate their (gross) investment returns before fees. In Section 3 we discuss a variety of benchmarks to evaluate the style and performance of CTAs. Given the strategy space of CTAs, we focus is on simple futures based strategies in equity, commodity and currency markets that are in the public domain. We find that CTAs do not add value, in the sense of producing alpha relative to these benchmarks. In Section 4 we review the historical performance of commodity funds in light of the earlier work by Elton Gruber and Rentzler. In section 5 we explore explanations for why CTAs persist despite two decades of poor performance. Section 6 concludes.

2. Fund Performance Data

A CTA is a hedge fund which has registered to trade futures with the Commodity Futures Trading Commission. Like hedge funds, CTAs are essentially prohibited from advertising.7 Faced with this restriction, a primary way to reach potential investors is for the hedge fund or CTA to voluntarily report performance information to private companies, data vendors, which then sell the data. Individual funds can release their own performance data, but not comparative data for advertising purposes.8 For making an investment decision, comparing individual funds to other funds, the vendor data is the only publicly-available source of information for evaluating these funds. The data are purchased by the news media and published in a variety of locations, such as Barron’s or ManagedFutures.com, for investors to observe.9

Because the decision to report performance data by CTAs and hedge funds is entirely voluntary, it introduces a strategic element in the reporting process. The resulting biases lead to an overstatement of performance of hedge funds, which contributes to the inference problem for investors and researchers alike. While many of the biases are well-known, there seems to be less agreement on how to handle these biases when evaluating hedge fund performance. Without reviewing the entire literature, we illustrate some of the major biases in the context of CTAs, and discuss why some attempts to adjust for the biases are suspect.

Consider a naïve investor who is contemplating an investment in CTAs and decides to examine the track record of all currently investable funds. In order to simplify the data collection process, the investor uses the Lipper-TASS database to calculate the average return to CTAs that are currently in existence, going back to 1994. The resulting performance series is given by the top line in Figure 1, which shows the cumulative total returns to an equally-weighted (EW) portfolio of CTAs over this period. The average return (net of fees) on this portfolio was 12.6 % which exceeds the return on T-bills which was about 4.0 % per annum over the 14-year period between 1994 and 2007. Our naïve investor might conclude that CTAs are an attractive investment: they provide an absolute return over T-bills which is significant economically (8.6% per annum) as well as in a statistical sense (t-stat = 2.73). However, this calculation does not correct for various

7 The prohibition on advertising seems problematical with the internet. One need only type “commodity trading advisor” into Google to get a sense of what this means as a practical matter. There are 93,900 hits. 8 Individual CTAs can publicly present performance data. The CFTC under Regulation 4.41(a) adopted “a rule that leaves to the discretion of the [CPO, CTA, or principal] advertising results –whether actual, simulated or hypothetical—the format of that presentation, so long as that format is not false, misleading or deceptive.” See Federal Register Vol. 71, No, 163 (Wednesday, August 23, 2006), p. 49388. 9 See the “Market Lab” section of Barron’s which provides “Commodity Traders Advisors Performance.” Barron’s provides the current monthly return, year-to-date, 12-month return, 3-year return, and 5-year return, the 12-month annualized standard deviation, the 12-month maximum drawdown (%), and the assets under management. Not all of this available for every fund listed. The performance data comes from the CASAM CISM Database (formerly the MAR Database); see Hhttp://www.casamhedge.com/H .

4 185

biases in the database. Figure 1 previews our discussion in the remainder of this section that a correction for survivorship bias and backfill bias would lower the average return to CTAs by about 7.7% to 4.9% per annum, which is only 85 basis points above the average return to T-bills. The correct inference from the data ought to be that the average CTA does not offer absolute returns but merely adds risk.

Figure 1: Measures of CTA Performance The figure shows the cumulative performance of an investment in an equal weighted portfolio of Commodity Trading Advisors that report to the Lipper-TASS Database. The portfolio labeled With Survivorship Bias and Backfill Bias consists of all Funds that were alive at the end of our sample. The portfolio labeled With Backfill bias (no survivorship) includes all monthly return observations in the “live” and “graveyard module” of the database. The portfolio No Backfill or Survivorship Bias includes only fund-returns after the first date of a fund reporting to the database.

Cumulative Return CTAs 600 With Survivorship & With Backfill Bias No Backfill or Backfill Bias (no survivorship bias) Survivorship Bias Annualized Returns 12.6% 9.4% 4.9% volatility 11.8% 9.7% 9.7% 500 Sharpe Ratio 0.73 0.55 0.09

400

300

200

100

0 1995 1997 1999 2001 2003 2005 2007

With Survivorship & Backfill Bias With Backfill Bias (no survivorship bias) No Backfill or Survivorship Bias

2.1 Sources of Bias in Lipper-TASS

There are at least four sources of bias in the Lipper-TASS database:

Selection Bias

The selection bias stems from the strategic reporting decision by a fund. Funds that experience poor performance may decide not to report to the database. Funds that look to attract new investors are more likely to report, while successful funds may stop reporting to the database as their need to advertize may have diminished. This issue has been widely discussed in the literature.

5 186

Look-back Bias

Look-back bias refers to ex-post data withholding by a fund after observing performance. This can take several forms. For example, a fund is unlikely to not report the return(s) prior to liquidation due to poor performance. More generally it is likely that funds delay reporting poor returns. If performance improves subsequently, it may report the delayed returns, or alternatively drop out of the database when fund returns continue to be low. This option to withhold poor performance has been discussed in the literature. What seems to have gone unnoticed in the literature is that funds can ex-post remove their entire performance record from the database. Comparing two versions of the Lipper-TASS database, we find several instances where the entire track record of a fund disappears. Conversations with the vendor confirms that funds can indeed request to have their entire historical track record removed, based on the view that “reporting is entirely voluntary and at the discretion of the funds.” This “look-back bias” affected about 2% of the CTAs between the October 2007 and April 2008 versions of the database. It seems plausible that unsuccessful funds have a larger incentive to remove their performance data ex-post, which would lead to an upward bias in the performance of the funds that remain in the database. Quantification of the magnitude of this bias would require a full record of these deletions, which is unfortunately unavailable.10

Survivorship Bias

The survivorship bias occurs when a fund disappears from the database after it dies. By focusing only on funds that are currently in existence, the naïve investor in our example excluded funds that were dissolved. Because the surviving funds have outperformed their peers, this leads to an upward bias. Malkiel (1995) estimated the size of this bias by comparing the (annualized) returns for the live funds (those funds that still exist at the end of the data sample) to the whole data set of returns (including funds that exited during the sample period).11 Since 1994 Lipper-TASS has maintained a record of non-surviving funds in the “graveyard module” of the database. The top two lines in Figure 1 compare the average return of CTAs that were in existence at the end of 2007 to an equally-weighted performance of all funds in the “live” and “graveyard” modules of the database. Figure 1 illustrates that surviving funds have outperformed the average fund in the database by 3.2% (12.6% minus 9.4%) between 1994 and 2007.12 When we discuss the next source of bias, induced by backfill we will include all CTAs from both the live and graveyard modules.

Backfill Bias

Also known as “instant history,” backfill bias is created when funds are allowed to submit a performance history at the time of first reporting to the database. Because managers are more likely to report funds with a good history, and avoid reporting funds with poor histories, this

10 A by-product of the look-back bias is that it makes it difficult to exactly replicate results of other researchers unless the exact same version of the database is used. Lipper-TASS only distributes the most recent version of the database to current subscribers. 11 Fung and Hsieh (2000), Brown, Goetzmann, and Ibbotson (1999), Ackerman, McNally, and Ravenscraft (1999), and Liang (2000), among others, use this method. The estimates of the bias range from 3.0 percent (from Fung and Hsieh) to 0.2 percent (from Ackerman, McNally, and Ravenscraft). Malkiel and Saha (2005) report that the average difference between live hedge funds and defunct hedge funds is more than 830 basis points over the period 1996-2003. 12 These calculations do not exclude backfilled returns.

6 187

creates an upward bias in the returns prior to the first live reporting date.13 A comparison of the bottom two lines in Figure 1 illustrates the magnitude of this bias that results when “instant histories” of returns before the first reporting date are excluded.

The figure illustrates the wide difference between the average performance of funds when they report to the database in real time (4.9%) and the average performance of all funds including backfill (9.4%). The former return is lower because the average backfilled return of 11.3% considerably exceeds the “live” average return of 4.9%. The backfill bias in CTA returns mirrors the observation by Elton, Gruber and Rentzler, that publicly traded commodity funds in the 1980s generally failed to beat the historical performance reported in their prospectuses.

Early hedge fund studies starting with Park (1995) attempted to correct for this bias by excluding the first portion of the track record of each fund before calculating performance, typically a fixed number of months reflecting the estimated backfill for the “average” fund. This “x-month screen” is a crude measure that leads to overstatement of the measured returns of funds that have a longer backfill period than x months. Recent versions of the Lipper-TASS database contain a field for each fund indicating the date of first reporting to the database. The backfill bias can therefore simply be eliminated by discarding returns prior to the first reporting date. Perhaps surprisingly, many studies continue to apply the x-month screens to account for backfill bias.14 The following table illustrates that x-month screens lead to very different conclusions about the magnitude of the backfill bias for CTAs.

Table 1: Backfill Bias and CTA Performance The table gives the average return expressed as % per annum on an equally-weighted portfolio of CTAs between 1994 and 2007 using different screens for inclusion of funds in the portfolio. Backfill not removed includes all funds and months for which data are available in Lipper-TASS. Backfill removed only uses firm-month observations for funds after their first reporting date to the database. The x-month screen removes the first x months from the performance record of a fund before it enters the portfolio.

EW CTA Index Average Return (% p.a.) Backfill removed (first reporting date) 4.9 Backfill not removed 9.4 12-month screen 8.3 24-month screen 7.8 36-month screen 7.7

Applying a 12-month screen across all funds lowers the average CTA return by only 1.1 % per annum, as compared to 4.5% using the first day of reporting as a screen. Longer screens lower average returns but not to the extent of eliminating returns prior to the first live reporting date. The reason is that there is a great deal of dispersion in the number of backfilled returns across funds. In the Lipper-TASS data set, the average number of backfilled months for all hedge funds is 28 with a standard deviation of 33.86. For CTAs the average number of backfilled months is 43 with a standard deviation of 47.41. This explains why even a conservative 36-month screen is not sufficient to eliminate the backfill bias. Throughout our analysis we will only use funds for which

13 Several papers have quantified this bias, including Posthuma and Van der Sluis (2003), and Malkiel and Saha (2005). 14 Recent examples include Koslowski (2007) and Ter Horst and Verbeek (2007).

7 188

we have an observation on the date of first reporting, and exclude performance data prior to that date.

2.2 Sample

There is a choice of data vendors of Hedge Fund and CTA data. We elect to use Lipper-TASS because it has relatively broad coverage of CTAs and includes flags for the date of first reporting by funds, thus allowing for backfill biases to be taken account of.15

Our sample of CTAs is taken from the April 5, 2008 version of the database. The Lipper-TASS database consists of 10,179 hedge funds. CTAs appear under the primary category “Managed Futures,” which includes 827 funds (327 live and 500 in the graveyard module).16 To avoid the backfill bias we select only those funds for which the date of first reporting is available, which excludes the separate CTA module for which this information is not available and 134 funds in the hedge fund module. Of the remaining funds, 108 were discarded because they did not have any returns after the first date of live entry. Finally we exclude funds (3) that do not report returns net of fees. The resulting sample consists of 582 funds, of which 201 were in existence as of the publication of the database. As of December 2007, our sample covers approximately 20% of all CTAs in terms of money-under-management (MUM).17

2.3 The Cross-section of Performance of CTAs

The poor performance of the average fund, as measured by the average return on the equally- weighted (EW) index, may mask the presence of stellar performers. Figure 2 provides a scatter plot of the average excess net returns and standard deviations of the individual funds in the database. In order to allow for a sufficient number of observations to calculate the average net return by fund, we restrict ourselves to CTAs that report at least 24 monthly observations (excluding backfill) in the database. This limits the number of observations to 312 (down from 582).

For comparison we include the EW CTA net return index. The graph shows large cross-sectional variation among individual CTAs. Annualized average excess net returns range from -42% to +53%, and standard deviations range from 1.9% to 97%. The V-shape of the graph reflects the intuition that funds that take more risk are more likely to exhibit extreme performance. The figure shows that the average standard deviation among individual funds (18.28%) is about double the standard deviation of the EW CTA index (9.70%), which suggests some diversification benefits to holding a portfolio of CTAs. Perhaps surprisingly, the average and median CTA has outperformed the EW index. However, this is caused by an increasing number of funds reporting to Lipper-TASS during the second half of our sample, which is also the period when the average fund performance was higher. In the remainder of the paper we will concentrate on the performance of the EW index rather than individual funds to further analyze the asset class. First, few individual funds have a long time-series to analyze, because the attrition rate of CTAs is high.

15 The CISDM database, while potentially broader in scope lacks such a flag, which prevents us from indentifying backfilled returns. The Barclays database also lacks a backfill flag. The HFR database contains flags for backfilled returns but its coverage of CTAs is not as extensive as Lipper-TASS. 16 Lipper-TASS contains a separate CTA module covering 2,149 funds which overlaps with the hedge fund module. All hedge funds classified as Managed Futures are in the CTA Module, and CTAs that are in the CTA Module but not “Managed Futures” do not appear elsewhere in the Hedge Fund Module. 17 At the end of 2007, our sample contains 205 funds, of which 189 reported a combined MUM of $43.98 billion. BarclayHedge estimated industry-wide MUM to be $206.6 billion.

8 189

Figure 2: Individual CTA Risk and Return The figure shows the annualized average excess return and standard deviation for all CTAs that have at least 24 months of reported returns in the Lipper-TASS database after excluding backfilled returns. Excess returns are calculated as total returns minus the three month T-bill rate.

100 Individual Fund Averages Average Return Standard Deviation mean 3.59 18.28 median 2.98 15.80 EW index 0.85 9.70 80

Standard Deviation Excess Return 20

EW CTA Index 0 -60 -40 -20 0 20 40 60 Average Excess Return

Second, high individual fund volatility further complicates the inference about skill and style. In addition, the portfolio approach naturally takes into account the correlations among individual CTAs which are hard to model. We note that the performance of the EW index is lower than the average fund; this could bias out findings against finding average CTA skill if investors could have forecast this performance. This seems unlikely, but we will present separate results for the recent sub-period.

2.4 Robustness of Performance to Equal-Weighting of Funds

The fact that individual CTA performance has been higher during the second half of the sample, a time when the asset class experienced substantial inflows, suggests that investors may have rationally forecast the performance of successful managers. Also, in light of high entry and attrition rates of funds – discussed in more detail in section 5 – it is possible that the performance of the equally-weighted index is weighted down by a large number of small funds that briefly enter the database. It this is the case, an equally-weighted index would underperform an asset- weighted measure of performance. Unfortunately, this proposition is difficult to test due to incomplete data on asset under management in Lipper-TASS. For those funds that report a history of assets under management, we compared the performance of an equally-weighted index to an asset weighted index of funds, and find that an asset weighted index would have outperformed an equally weighted index by about 3% per annum between 1995 and 2007. This difference is not significant in a statistical sense (t = 1.30).

9 190

We also separately analyzed the performance of large CTAs. Perhaps CTAs signal with their “pedigree.” In other words, new managers that spin-off from established, well-known, funds or who were trained at established, well-known, funds may use the name of the fund where they worked as an advertisement for their trading acumen. Insofar, as these spin-offs have a track record at their prior fund, it is not public, but still there may be a kind of “seal of approval” from the original fund. Many of these might find it easier to attract assets and turn into large funds. To address this issue, we constructed sample of large CTAs in our data set. These tend to include many of the well-known names.18 In any given month a CTA is categorized as large if it had at least $250 million under management over the last 12 months. Then we constructed a Big CTA Index, which consists of the equally-weighted returns of the large CTAs, apart from the above mentioned cut-off we also constructed an equally weighted index of CTAs that had more then $100 million under management. Table 2, below, compares the performance of the equally weighted CTA index with the Big CTA indices discussed above. The comparison is for the period of 1998 to 2007.19

The Big CTA Index contains many of the large, well-known, CTAs – but their non-backfilled performance suggests that pedigree is not a signal, though it may be successful as an advertisement. More importantly, we find very little difference between the performance of large and small CTAs in our sample. For this reason, and because of the relative small number of funds for which data on assets is available, we decided to use the equally-weighted index of CTAs in the remainder of the paper.

Table 2: Big CTA Index Returns 1998-2007 The table gives the annualized average return, standard deviation and Sharpe ratio of the Equally-Weighted portfolio of CTAs in Lipper-TASS, and two portfolios of Big CTAs. A CTA is classified as Big in a year if it reports assets under management at some point during the prior 12 months that exceed USD 250 MM, or 100MM. All return calculations exclude reported returns prior to the date of first reporting to the database.

EW CTA Index CTAs > 250MM CTA s > 100MM Average Return 5.6 6.5 7.7 Standard Deviation 9.9 13.3 13.5 Sharpe Ratio 0.20 0.22 0.30

2.5 CTA Performance Before and After Fees

In addition to net (of fees) returns, we are interested in gross returns for two reasons. First, gross returns measure the payoffs to the fund’s portfolio investments and speak to the question of whether a manager has the ability to generate positive investment returns. A comparison of gross and net returns indicates how the returns to skill are shared between the fund and its investors. The discrepancy is potentially large, because CTA fees resemble those of hedge funds: in our sample fixed fees on money-under-management range from 0.167% to 8.0% per annum while variable performance fees range from 0% to 50%. The average fixed fee is 2.15% and the variable fee averages 19.5% across funds. The second reason to study returns before fees is that gross returns are potentially better suited for performance analysis because the fee structure may induce additional nonlinearities in the post fee returns.

18 Some of the well known names included in the sample are: Campbell & Company Inc., Graham Capital Management, Man Investments Ltd, Winton Capital Management Ltd, and Aspect Capital Ltd. 19 There are very few big CTAs prior to 1998.

10 191

Table 3: CTA Excess Returns and Fees

The table gives the annualized average excess return and standard deviation of the equally-weighted portfolio of all CTAs in the Lipper-TASS database before and after fees, between 1994 and 2007. Before fee returns are estimated using net of fee data and fee information using the methodology outlined in French (2008).

Average Standard Deviation t-statistic After Fees 0.85 9.70 0.33 Before Fees 5.37 9.79 2.05

Brown et al. (2004) and French (2008) estimate gross returns for hedge funds from net returns and fee information. We follow French (2008) in the construction of gross returns for Managed Futures funds in Lipper-TASS, using the reported net returns. We make two assumptions implementing French’s model, namely that fees accrue on a monthly basis, and that high watermarks, when applicable, increase at the rate of return on T-bills. Table 3 summarizes the effect of fees on performance.

The table shows that:

1. As a consequence of fees, the estimated average return on a fund’s investments of 5.37% exceeds the return earned by investors (85 bps) by 4.52% per annum. Although not included in the table, of this difference 2.19% can be attributed to the fixed component of the fee structure, and 2.33% to variable performance fees.

2. We can reject the hypothesis that the average CTA has no ability to outperform T-bills. The gross excess return is marginally significantly different from zero (t = 2.05). However, most of this outperformance accrues to the fund management through levying fees, leaving on average 85 bp per annum for fund investors, an amount that is indistinguishable from zero in a statistical sense.

2.6 Performance Summary

The conclusion from this section is that the properly bias-adjusted average return to investors from CTAs has been poor between 1994 and 2007. Relative to T-bills, the average value added after fees – which is what investors care about – has been 85 basis points per annum. And in order to earn these returns, investors had to accept volatility at the fund level that has been comparable to investing in equity indices. Our observations closely resemble the central conclusions of the EGR studies (1987, 1990) which document poor performance of public commodity pools between 1979 and 1988. Why is it that CTAs not only have survived since the EGR publications, but have thrived as measured by the growth of money managed by the industry? We will return to a discussion of these issues in Sections 5 and 6.

The poor returns to CTAs do not imply an absence of skill of CTA fund managers. Our results are consistent with a world in which CTAs produce “alpha” before fees but successfully capture most of the rents they generate through charging (high) fees. In the next section we will attempt to identify particular investment strategies of CTAs. This is of interest because CTAs describe their style as predominantly trend-following, and academic research has documented that certain trend following (or momentum) strategies are profitable. Do CTAs extract fees from following simple strategies that are in the public domain? Or does a substantial component of their fees come from other sources that generate alpha?

11 192

In the next section, we will examine the correlation of CTA returns with various versions of simple dynamic strategies, which we will use as benchmarks for performance analysis. This will provide the answer to two questions. First, is there a predominant style for CTAs and how pervasive is this style? Second, how does CTA performance compare relative to these benchmarks?

3. Normative Asset Based Benchmarks

A central characteristic of hedge fund strategies is that they invest in active strategies, take both long and short positions and generally use leverage. For these reasons it has been difficult to specify appropriate benchmark returns that are comprised of passive strategies that capture the potential non-linear nature of hedge fund returns (see, for example, Hasanhodzic and Lo (2007) for a discussion). In the first subsection we illustrate the difficulties of developing benchmarks by looking at risk factors developed by Fung and Hsieh. Then, in subsection 3.2, we set out our own “Normative” benchmarks, factors that we think CTAs ought to reasonably outperform. In subsection 3.3 we analyze CTA gross return performance against the Normative benchmarks. Subsection 3.4 looks at subperiods. Subsection 3.5 summarizes our analysis of individual fund performance, as opposed to the EW index.

3.1 Fung and Hsieh Factors

Fung and Hsieh (FH) (2001) demonstrate that CTAs actively engage in trend-following strategies which generate option-like characteristics in their payoff structures. This motivates FH to conduct a style analysis in which they compare CTA returns to a dynamically traded portfolio of look- back (options) straddles. Fung and Hsieh (2004) label their approach “Asset Based Style Analysis”.

We follow a similar approach in this paper, and construct a set of active strategy returns for each of three asset classes for which there exist liquid futures markets: commodities, foreign exchange, and equities. Our focus is slightly different from FH in that we are not merely interested in creating “positive” benchmarks that successfully describe the style of hedge funds. In addition we want our benchmarks to be “normative” and useful in evaluating the performance of hedge funds against these benchmarks. In particular, when the benchmarks are dynamic trading strategies themselves, there can be a tradeoff between the objective of capturing style and measuring performance. To illustrate this issue consider the following regression of the EW Index of before- fee CTA (gross) excess returns on the FH factors (using their notation):20

20 The five factors (PTFSBD, PTFSFX, PTFSCOM, PTFSIR, and PTSSTK) are factors that have been constructed by Fung and Hsieh (2001) to represent nonlinear trading strategies designed to capture “trend following” by CTAs. Each acronym starts with the prefix “Primitive Trend-Following Strategy” and then includes Bonds (BD), Foreign Exchange (FX), Commodity Markets (COM), Interest Rates (IR), and Stocks (STK). Construction of these factors involves rolling a pair of lookback straddles for various asset classes. Applying the analysis to CTAs, Fung and Hsieh interpret their results as supporting the view that CTAs follow nonlinear, option-like, strategies. Fung and Hsieh (2001) conclude that the use of their nonlinear factors “supports our contention that trend followers have nonlinear option-like strategies” (p. 337).

12 193

R 77.0 02.0 ×+= PTFSBD + 04.0 × PTFSFX + 05.0 × PTFCOM − 02.0 × PTFSIR 04.0 ×+ PTFSSTK EW )81.3( )40.2( )32.4( )06.3( − )91.2( )90.2( R 2 = 25.0 where the dependent variable represents the excess gross returns of the equal weighted portfolio of CTAs and the independent variables are the excess returns of the FH style factors corresponding to bonds (PTFSBD), currencies (PTFSFX), commodities (PTFCOM), interest rates (PTFSIR), and equities (PTFSSTK). As explained above, we analyze returns gross of fees because that return series captures the talent of the average manager. The regression shows that the various style factors explain about 25 percent of the variance of CTA excess gross returns. And controlling for exposure to the various styles, the average CTA earns an excess return of 0.77 percent per month (t = 3.81), which is about 9.2 percent annualized. The regression seems to indicate that the style factors are somewhat successful in capturing various aspects of CTA return variance, and provides evidence of positive excess gross returns after controlling for style (“alpha”).

The interpretation of the regression alpha is complicated by the fact that the style factor returns correspond themselves to dynamic trading strategies, which may be inefficient replications of that particular style. Although the payoffs to trend-following rules can mimic those of look-back options strategies described by FH, it is likely that CTAs will achieve these payoffs by directly trading in futures markets rather than options markets. The return on trading look-back straddles would understate the achievable returns to the trend-following style. In what follows, we will show that trend-following characteristics are as easily captured by simple momentum strategies, which outperform the FH style factors and change the inference about the presence of “alpha.”21 Table 4 gives the excess returns on the FH factors between 1994 and 2007.

Table 4: Performance of the Fung and Hsieh Style Factors The table gives the annualized average excess returns and standard deviations of the style factors reported to capture “trend following” in FH (2001). These factors are constructed as the returns on look-back option straddles in bonds (BD), Currencies (FX), Commodities (Com), Interest Rates (IR) and Stocks (STK). Sample consists of monthly data between 1994 and 2007. Data Source: http://faculty.fuqua.duke.edu/~dah7/HFRFData.htm

Mean Standard Arithmetic Geometric Deviation PTFSBD -19.5% -27.2% 51.1% PTFSFX -4.0% -20.7% 64.9% PTFSCOM -9.3% -17.5% 46.1% PTFSIR 5.2% -17.0% 85.8% PTFSSTK -64.7% -53.6% 44.0%

The table shows that the (geometric) average excess returns of the FH style factors has been negative (and highly volatile) over the 14-year period between 1994 and 2007. The issue is more dramatically illustrated in Figure 3, which plots the cumulative return for the five FH factors: a dollar invested in each of these factors at the end of 1993 would have lost more than 90 cents of its value by 2007. Unlike passive benchmarks, dynamic rebalanced portfolios may require frequent trading. And because the style returns are measured before transactions costs, accounting for trading costs in options markets would further lower the reported averages in the table.

21 See also Cremers, Petajisto, and Zitzewitz (2008) of a discussion of the effect of nonzero alphas of benchmark indices on performance attribution.

13 194

Measurement error in the style returns induces measurement error in the alpha of CTAs relative to these style benchmarks. Because the average CTA positively loads on the FH style portfolios, the resulting alpha will exceed the raw excess gross return to CTAs. To the extent that the style returns reflect inefficient replication of the trading strategies followed by CTAs, this will lead to an upward bias in the alpha. It seems unlikely that CTAs would choose to follow styles that have earned negative returns over a 14-year period. Instead, it seems more plausible that the negative style returns and the apparent positive alpha are merely a reflection of inefficient benchmarks.

In fairness, the FH style factors were not, of course, intended for the purpose of performance evaluation, yet the example illustrates the tradeoff between capturing style and performance evaluation when the style portfolios are not passive benchmarks. To the extent that options are expensive, a strategy that buys straddles to mimic trend-following behavior will exhibit negative excess returns.

In the spirit of the FH analysis we propose to evaluate the style of CTAs by correlating their returns to those of dynamic trading strategies in equities, currencies and commodities. Our strategies differ in two respects from FH. First they are relatively cheap to trade, and therefore are more useful for performance evaluation. For example we evaluate the performance of CTAs against a set of simple momentum strategies which are likely to capture the basic characteristics

Figure 3: Cumulative Performance Fung-Hsieh Trend Following Factors The figure shows the cumulative total return, standard deviation, and Sharpe Ratio of the trend following factors (PTFS) reported by Fung and Hsieh (2001). These factors are constructed as the returns on look-back straddles in bonds (BD), currencies (FX), Commodities (COM), interest rates (IR) and Stocks (STK). Sample consists of monthly data from 1994 to 2007 Data Source: http://faculty.fuqua.duke.edu/~dah7/HFRFData.htm

350 PTFSBD PTFSFX PTFSCOM PTFSIR PTFSSTK Annualized Returns -15.5% -0.01% -5.3% 9.3% -60.7% Volatility 51.1% 64.9% 46.1% 85.9% 44.1% 300 Sharpe Ratio -0.38 -0.06 -0.20 0.06 -1.47

250

200

150

100

0 Dec-93 Dec-95 Dec-97 Dec-99 Dec-01 Dec-03 Dec-05 Dec-07

PTFSBD PTFSFX PTFSCOM PTFSIR PTFSSTK

14 195

of trend-following but are likely to be cheaper than option straddles. Second, our choice of benchmarks is not just based on what CTAs self-purportedly do (trend-following), but also on what they ought to be doing, in the sense that the strategies are dynamic strategies in the public domain. In addition to momentum, we select for each asset class a second style factor that is in the public domain and has been documented to be correlated with average returns. These factors are value (price-to-book) for equities, interest rate differentials (the carry trade) for currencies, and the basis (backwardation) for commodity futures. We call these benchmarks “Normative” benchmarks.

3.2 Normative Benchmarks Performance

We construct the Normative Benchmarks by constructing rules-based active strategies using primitive assets that include currency futures, commodity futures and country equity indices. The active strategies are intended to capture known sources of return as well as self-declared styles of CTAs. In our selection of benchmark portfolios, we are guided by the academic literature, to ensure a reasonable expectation that these benchmarks are indeed in the public domain and therefore available to CTAs. In line with the previous evidence of CTA trend-following, we construct a momentum factor for each of the three major asset classes: currencies, commodities, and equities. In addition, we construct actively traded portfolios based on the forward bias in currencies (“carry trade”), a factor to capture inventory effects (“backwardation”) in commodities markets, and a factor related to cross-country value in equity markets (the price-to-book ratio (PB)). A detailed discussion of the construction of these factors is contained in the Appendix.

Table 5: Annualized Average Excess Returns and Standard Deviation of Normative Benchmarks 1993/12 – 2007/12 The table gives the average excess return, standard deviation, and t-statistic for a test of non-zero average excess return for the Equally-Weighted portfolio of CTAs and portfolios of dynamically traded futures of Commodities, Equities, and Currencies. Dynamic portfolios are constructed by monthly sorting commodity, equity and currency futures on past performance (Momentum), end of prior month futures Basis (Commodities, and Currencies) or Price-to-Book (Equities). Long-Only indices take long positions in the top half of the relevant assets in this ranking, while Long-Short takes a long position in the top half and a short position in the bottom half of the futures in the ranking. Panel A: Long-only Average Volatility t-stat (Average) EW CTA Index After Fees 0.9% 9.7% 0.33 EW CTA Index Before Fees 5.4% 9.8% 2.05 Hi Momentum 15.1% 11.6% 4.88 Commodities Hi Basis 13.0% 11.7% 4.17 Hi Momentum 9.4% 14.5% 2.43 Equities Low PB 8.7% 15.0% 2.15 Hi Momentum 2.6% 7.2% 1.34 Currencies Hi Basis 4.1% 6.8% 2.27 Panel B: Long-Short Average Volatility t-stat (Average) LS Momentum 15.9% 14.3% 4.17 Commodities LS Basis 11.9% 12.6% 3.52 LS Momentum 3.9% 8.6% 1.71 Equities Low minus Hi PB 3.0% 6.5% 1.73 LS Momentum 0.7% 5.2% 0.51 Currencies LS Basis 4.0% 5.8% 2.60

15 196

Table 5 summarizes the excess returns to the Normative Benchmarks between 1994 and 2007. For each asset class and strategy, we report both the excess returns of the High (“HI”) characteristic portfolio (Long-Only), as well as the excess return difference of High minus Low (Long-Short or “LS”).

The table illustrates the following points:

1. Contrary to the FH factors, all of our Normative Benchmarks have earned positive risk premiums. Long-Only excess returns range from 2.6% per annum (FX Momentum) to 15.1% (Commodities Momentum). With the exception of the FX Momentum premium, all Normative factor premiums are significantly different from zero.

2. The Normative factor excess returns exceed the average return of the equal-weighted return CTA index after fees and, with the exception of currencies, exceed the average return of the equal-weighted return CTA index before fees.

3. The Long-Short excess returns in Panel B are slightly lower than the Long-Only excess return, but with the exception of LS FX exceed the excess returns on the EW CTA after fees index. Only the commodities strategies exceed the EW CTA before fees returns.

The table is compelling in the sense that average CTA performance (before or after fees) is poor relative to most of our Normative Benchmarks. The full sample Sharpe Ratio of CTAs after fees is only 0.09, and the Sharpe ratio before fees is 0.55, as compared to the Sharpe ratios of the active strategies which exceed 0.94 in the case of commodities, and 0.14 in the case of currencies.

The finding of relatively low gross returns already suggests that CTAs follow strategies that are different from those embedded in the benchmarks. A formal performance evaluation is the subject of the next subsection of the paper. Regression of estimated gross excess returns on the Normative Benchmarks addresses the question whether CTAs earn alpha relative to a set of strategies that are in the public domain.

3.3 The Performance CTAs and Asset Based Style Benchmarks

Table 6 contains the regression results of the gross excess return of the EW CTA index on the excess returns of the various factors using data from 1994 to 2007, as well as two sub-sample periods. The slope coefficients and R-squared of these regressions are informative about average CTA style, while the constant term provides us with the estimate of alpha conditional on the style factors. All specifications include the S&P 500, the Lehmann Aggregate Bond Index, and the Gorton and Rouwenhorst (2006) Equally-Weighted Commodity Index (GRCI). The motivation for including these three indices is twofold: first, they benchmark the CTA returns relative to the basic passive asset class exposures. Second, we find that including passive benchmarks seems to alleviate the problems outlined in section 3.2, where the regressions alphas are biased by inefficient replication of style factors.

In addition to the passive benchmarks, we contrast three dynamic style benchmarks. The first is the Mount Lucas Index (MLM), a commercially-produced index that equally-weights 25 different

16 197 Table 6: The Abnormal Performance of CTAs

The table gives the results of a regression of the excess return of the Equally-Weighted portfolio of CTAs, gross of fees and corrected for survivorship bias and backfill bias on three groups of style factors. MLM refers to the excess return of the Mount Lucas Index, F&H are excess returns of the five trend following style factors of Fung and Hsieh (2001), and the LS are the excess returns of asset based style factors based on long-short positions in Commodities (based on momentum and the basis), Stocks (based on price-to-book and momentum), and Currencies (momentum and the basis). In addition to the sets of style factors each regression includes the excess return of three passive benchmarks: SP500, Lehman Aggregate Bond Index, and the Equally-Weighted Commodity index described in Gorton and Rouwenhorst (2006). In parentheses below the coefficients are t-statistics, corrected for heteroskedasticity.

Sample: 1994-2007 Sample: 1994-2000 Sample: 2001-2007 MLM F & H MLM F & H MLM F & H Index Factors LS Ind ex Factors LS Ind ex Factors LS 0.265 0.147 0.582 0.028 0.000 -0.277 0.117 -0.313 0.484 0.651 1.069 0.562 Constant (1.22) (0.74) (2.85) (0.13) (0) (-1.01) (0.44) (-1.15) (1.52) (2.41) (3.23) (1.74) 0.367 0.403 0.369 Commodities LS Momentum (4 .23 ) (2 .4 6) (3.5 3) -0.110 -0.196 -0.078 Commodities LS Basis (-1.06) (-1.03 ) (-0.62 ) 0.106 -0.319 0.504 Equities Hi minus Low Momentum (0.7) (-1.57) (1.92) 0.039 0.028 -0.256 Equities Low minus Hi PB (0 .19 ) (0.1) (-0.88 ) 1.331 1.272 1.615 FX LS Momentum (5.5) (3.95) (3.72) 0.048 0.203 -0.220 FX LS Basis (0 .23 ) (0 .8 2) (-0.52 ) 0.571 0.485 0.767 MLM (5.4) (3.23) (3.9) 0.025 0.034 0.033 PTFSBD (2.45) (2.27) (2.27) 0.040 0.028 0.053 PTFSFX (4.22) (2.86) (3.26) 0.046 0.048 0.044 PTFSCOM (2.63) (1.73) (2.24) -0.012 -0.017 -0.012 PTFSIR (-1.78) (-1.46) (-1.26) 0.038 0.016 0.061 PTFSSTK (2.68) (0.86) (2.97) Rbar-squared 0.097 0.201 0.296 0.233 0.036 0.106 0.204 0.155 0.119 0.282 0.343 0.329 Durbin-Watson 1.908 1.846 1.882 1.923 2.090 2.008 2.028 2.182 1.714 1.795 1.878 1.788 Num of Obs 168 168 168 168 84 84 84 84 84 84 84 84

17 198

futures contracts that cover foreign exchange, energy, financials, metal and agricultural futures.22 It is rebalanced monthly, as follows. If the 200 day moving average is greater then the closing price of the future then it takes a short position otherwise it takes a long position. The MLM index is a widely used benchmark for CTAs, which is why we include it here. We compare including MLM to the FH style regression and our Normative Benchmarks. White heteroskedasticity-adjusted t-values are in parentheses. The Table shows that:

1. The estimated alpha of CTAs after controlling for passive exposure to stocks, bonds and commodities CTAs is 0.265% per month (3.2% annualized) between 1994 and 2007, which is insignificantly different from zero. Although we estimated earlier that CTAs earn positive excess returns before fees, we cannot reject absence of skill after controlling for passive asset class exposures.

2. The only instances in which we can reject absence of alpha are for the FH and MLM benchmarks during the 2001-2007 sub-period, and FH factors for 1994-2007. A comparison with the constant term of a regression on only the passive benchmarks shows that the MLM and FH alphas are driven by negative realized excess returns to the MLM and FH factors during the second half of the sample. CTAs do not add value relative to the Normative Benchmarks.

3. Style benchmarks can explain up to about 30% of the full sample variance of the EW CTA gross of fees index (as measured by adjusted R-squared). However, the explained variation remains low, which suggests that there may be potentially important, omitted style factors.

4. Among the two styles, CTAs tend to have higher average exposure to Momentum factors than Basis or Value. This is consistent with previous studies that identified trend-following as the major CTA style. Among the three asset classes, Momentum exposure is highest in Currencies and Commodities.

The overall conclusion of this section is that the average CTA – as measured by the EW gross of fees performance index – has failed to deliver alpha to investors. The predominant style has been one of trend-following, most pronounced in currencies, but the combined factors have a maximum explanatory power of 33% (over the period 2001-2007). This evidence does not support the hypothesis that CTAs on average adhere to the trading strategies as embodied in the Normative Benchmarks and capture the apparent profits of these strategies through charging fees. Instead, it suggests that CTA performance has a large idiosyncratic component, and that the combination of poor performance and high fees has on average resulted in absence of alpha for investors.

3.4 Individual Fund Analysis

Looking at the alphas on individual CTAs, with at least 24 months of returns, relative to the Normative Benchmarks, we find that 21 percent of the CTAs have an alpha which is significant at the five percent level. These alphas are about evenly divided among positive and negative, and the average of these significant alphas is −0.14 percent per month. So, CTAs perform poorly on average, but even those with individually significant alpha are not particularly good performers.

22 For the Mount Lucas Index (MLM), see Hhttps://www.mtlucas.com/about.aspxH . On MLM, see Mulvey, Kaul, and Simsek (2003).

18 199

Given the relatively small strategy space of CTAs it is perhaps surprising that style analysis can explain only about a third of the variance of the returns on the EW CTA gross of fees index. As shown in Figure 4, however, a style analysis conducted at the individual fund level reveals a similar picture. The histogram shows the distribution of the R-squared of a regression of individual fund level returns on the Normative Benchmarks. The figure shows that 74% of the funds have an R-squared that is below the index-level regression (0.33). And about 24% of regression R-squareds is below 10%. This further complicates the inference problems that investors in CTAs face. Even if investors are able to obtain clean performance data, an analysis of the self-proclaimed style of CTAs can explain less than 30% of the return variance for most of the funds. For 12% of the CTAs the adjusted R-squared is not positive. The overall conclusion of our style analysis is therefore that proclaimed style explains very little of the variance of individual CTA returns as well as of the returns to the broader asset class.23

Figure 4: Distribution of R-squared of Individual Fund Returns on Normative Style Factors For each CTA with at least a 24-month return history after controlling for backfill and survivorship bias, we regress the excess fund gross return on the excess return of the 6 style benchmarks that capture Momentum (Commodities, Currencies and Equities), Basis (Commodities, Currencies) and Value (Price-to-Book, Stocks). The figure provides a histogram of the distribution of the R-squareds of the fund-level regressions.

6 Relative% Frequency

0 -0.275 -0.175 -0.075 0.025 0.125 0.225 0.325 0.425 0.525 0.625 Adj-R-squared (bin midpoint)

4. Why do CTAs Persist? An Historical Perspective

Our conclusion about the lack of return and alpha of CTAs is surprising because the asset class has experienced substantial inflows over time. For example, BarclayHedge estimated money- under-management of CTAs to be $206.6 billion as of December 2007, which is an increase of

23 Results are qualitatively similar for net-of-fee fund level returns.

19 200

306 percent from 2002 when assets were estimated to be $50.9 billion. Apparently investors increased allocations following years of poor performance. This finding is perhaps even more striking considering the broader historical context of commodity funds provided by the studies of Elton, Gruber and Renzler (EGR 1987, 1989, 1990). They studied the performance of publicly traded commodity funds between 1979 and 1985, and document a similarly poor performance in their earlier sample. EGR attributed the existence and persistence of poorly-performing CTAs to misinformation. They find that no fund is able to outperform its prospectus track record. EGR concluded that the historical returns series provided in the prospectuses of public commodity funds were misleading. The past performance was unreasonably upward biased and investors had no other information to rely on.

The EGR studies were widely reported on in the press, sometimes in scathing terms. See, for example, Newswire (September 18, 1986), the Wall Street Journal (September 29, 1986), the Toronto Star (October 5, 1986), Chicago-Sun Times (November 17, 1986), the Washington Post (March 1, 1987, September 28, 1987), the St. Petersburg Times (September 26, 1987), the San Francisco Chronicle (October 5, 1987), the New York Times (August 20, 1988), Business Week (November 28, 1988), The Economist (December 1, 1990), Forbes (September 2, 1991). As an example, here is one part of an article on their studies by Jane Bryant Quinn in the San Francisco Chronicle (February 21, 1989): The larger - and more intractable - scandal lies in the entirely legal deceptions that surround the selling of commodities funds in the first place. Brokerage firms mislead you as a matter of course, with the full approval of the market's so-called regulators.

The problems lie in the sales brochures and prospectuses for new commodities funds. They "disclose" the portfolio manager's past performance, which is never anything less than spectacular. Gains may be claimed of 50 percent, 60 percent, even 70 percent a year.

But those astonishing track records can be a clever form of fiction. They're not wrong, exactly. But they're biased and misleading. They greatly exaggerate the manager's chance of success.

For proof, I give you a study by three New York professors - Edwin Elton and Martin Gruber of New York University's Graduate School of Business, and Joel Rentzler of the Baruch College of the City University of New York. They took 77 new commodity funds, and compared the managers' past performance with how well the funds actually did in practice. The verdict: disaster.

Given the widespread publicity, it is hard to believe that investors would continue investing. But, things did change, in two important ways. First there was a regulatory reporting change. Second, the form of the investment vehicle changed.

On the regulatory front, subsequent to the EGR papers, the Securities and Exchange Commission put out a STATEMENT OF THE COMMISSION REGARDING DISCLOSURE BY ISSUERS OF INTERESTS IN PUBLICLY OFFERED COMMODITY POOLS SECURITIES AND EXCHANGE COMMISSION, Release Nos. 33-6815; 34-26508 [S7-1-89]; 17 CFR Parts 231 AND 241, February 1, 1989, which said in part::

Certain recently published studies suggest that the actual performance of publicly held commodity pools was significantly lower than the performance disclosed in

20 201

the prior performance tables included in commodity pool disclosure documents.* While the findings and issues raised in these studies are currently being reviewed by the staff of the CFTC, the Commission believes that it should provide guidance to issuers of publicly offered commodity pools at this time. Although the positions expressed in this release and the CFTC's interpretive statement currently reflect the respective agencies' views regarding appropriate disclosure in commodity pool disclosure documents, the Commission is interested in receiving views on the interpretive positions expressed in those statements. Commentators may wish to make the same submission to both agencies. The Commission expects to consult with the CFTC concerning the comments received in response to their respective statements with a view towards determining whether further action is necessary or appropriate.

* See Elton, Gruber & Rentzler, New Public Offerings, Information and Investor Rationality: The Case of Publicly Offered Funds, 62 J. Bus. 1-15 (January, 1989). The authors hypothesized that the findings of the study were at least in part due to the following factors: 1) public commodity pools have larger transaction costs and management fees than private commodity accounts; 2) only trading advisers with recent successful track records are likely to go public; and 3) trading advisers can select the period of time for disclosing their prior performance, resulting in an upward bias in performance results. See also Edwards & Ma, Commodity Pool Performance: Is the Information Contained in Pool Prospectuses Useful? Working Paper Series No. 16, Center for the Study of Futures Markets, Columbia Business School (January, 1988). [Footnote in original.]

Subsequently, filings of public commodity funds included the SEC statement in their filings.24 Also, the CFTC did change the reporting requirements.

The second change appears to have been a response from the commodity fund industry to the EGR publicity. After the EGR studies and ensuing publicity, and after the CFTC reporting requirement changes, commodity trading advisors appear to have stopped the frequent use of publicly-offered funds, which required a prospectus following the new rules. Rather, commodity fund managers began to structure themselves like hedge funds, which require less disclosure. One possibility is that this change in organizational form was enough to entice investors to continue to invest.

5. Explaining the of Persistence CTAs

Data sets that have been strategically manipulated not only make it hard for econometricians to draw inferences, investors have the same problem. It is difficult to evaluate performance and, as we have discussed, even to determine CTAs’ style. In this section we delve into two related issues. First, we ask whether CTA return distributions have desirable characteristics that are not captured by means and variances. In particular we examine whether individual CTA returns exhibit skewness and coskewness with other asset classes, which might explain why the asset class can persist despite offering poor returns on average.

The second set of explanations focuses on the information asymmetry between investors and funds. We first investigate whether there is evidence that talented CTAs try to overcome the

24For example, the JWH Global Trust S-1 on Nov. 26, 1996 (see Hhttp://www.secinfo.com/dRqWm.9rzv.2.htmH , section Ex-99.01). Other examples include Kenmar Global Trust, July 25, 1996 (Hhttp://sec.edgar-online.com/1996/07/25/00/0001005477-96- 000208/Section21.aspH), and also MAN-Ahl 130/LLC S-1/A Nov. 11, 2005, Ex-99.01: Hhttp://www.secinfo.com/dsvRm.zcZk.8.htm#1stPageH .

21 202

information problem by signaling. Secondly, is there evidence that investors are aware of the information issues, concerning, for example, biased performance data?

5.1 Zero Alpha, but Positive Skewness: Are CTAs Lotteries?

Perhaps CTAs generate desirable skewed returns. It is well-known that if investors have utility functions that display decreasing absolute risk aversion, then they will have a preference for positively skewed returns (see, e.g., Markowitz (1952) and Arrow (1971)). Also, the literature in behavioral finance suggests that investors may have a preference for skewed payoffs (see, e.g., Barberis and Huang (2007)). There is some evidence that investors indeed do have a preference for skewness. For example, Levy and Sarnat (1984) find a strong preference for positive skewness in a study of mutual funds. Also, see Polkovnichenko (2005). It is possible to reconcile the poor performance of CTAs with the growth in assets-under-management if CTA returns exhibit positive skewness. We calculate the skewness of individual CTAs using the full-sample of available returns for all the funds that have at least 24 months of returns, 312 funds. Figure 5 below provides a histogram of the sample skewness of individual funds.

Figure 5: The Distribution of Individual CTA Return Skewness For each CTA with at least a 24-month return history after controlling for backfill and survivorship bias, we calculate the skewness of monthly returns. The figure provides a histogram of the distribution of the calculated fund-level skewness.

4 Relative% Frequency

0 -3.5 -3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3 3.5 Skewness (bin midpoint)

The histogram in Figure 5 shows that the number of CTAs with positive skewness is about the same as the number of funds with negative skewness. The median estimated skewness is 0.056 and the average is 0.0213. The figure also shows that some CTAs do exhibit large skewness, but these are equally divided over the positive and negative tails of the empirical distribution. Absent

22 203

skewness, it is unlikely that CTAs attract assets because they offer lottery-like payoffs to investors.

A related issue concerns coskewness with other asset classes. Fung and Hsieh (2001) argue that CTAs funds are attractive because they do well when other asset classes are not doing well, in particular when there are “tail events” in which the other asset classes are doing particularly poorly. Vice versa, perhaps the other asset classes are doing well when CTAs are doing particularly poorly. We look at this issue in a very simple way. We look at the performance of the S&P500, the Lehman Aggregate Bond Index (LABI), and the equally-weighted CTA Index in the months where each had the 5 percent worst months of performance and the 5 percent best months of performance in our sample period. For those months we ask how the other asset classes performed. Table 7 shows the results.

Table 7: Tail Correlation during Extreme Events For Each asset class, including the S&P500, the Lehman Aggregate Bond Index, and the Equally-weighted CTA Index, we compute the average annualized return for the other indices in the months where the specified index had its 5 percent worst performing months and the 5 percent of the best performing months.

Panel A: Best and Worst 5% of S&P500 Months

Worst 5% S&P500 Months CTA S&P500 LABI Monthly Average ER 4.1% -9.3% 0.7% Best 5% S&P500 Months CTA S&P500 LABI Monthly Average ER -1.0% 7.9% 0.2%

Panel B: Best and Worst 5% of Lehman Aggregate Bond Index Months

Worst 5% LABI Months CTA S&P500 LABI Monthly Average ER -1.0% -1.1% -4.7% Best 5% LABI Months CTA S&P500 LABI Monthly Average ER 2.3 2.9% 4.5%

Panel C: Best and Worst 5% of the EQ CTA Index Months

Worst 5% CTA Months CTA S&P500 LABI Monthly Average ER -4.9% 2.1% -1.8% Best 5% CTA Months CTA S&P500 LABI Monthly Average ER 6.6 -2.4% 1.6%

Looking at Panel A of the table, CTAs do well in the months when the S&P500 is doing very poorly and conversely do poorly when the S&P500 is doing well. Fung and Hsieh (2001) also make this point. This pattern is also true when we select the worst months and best months for CTAs, Panel C of the table. While there is nothing that stands out in this regard for bonds; see Panel B. What is less clear is whether this tail behavior is sufficient to justify an investment in

23 204

CTAs despite the poor performance. This would not seem to be justification since this type of diversification can be achieved at much lower cost using passive indices of commodity futures; see Gorton and Rouwenhorst (2006). The correlations of CTA returns with traditional asset classes in the tails seem unlikely to justify investors allocating $200 billion to an asset class that offers T-bill returns with a standard deviation that is comparable to equities.

We conclude that there is no compelling evidence to justify investing in CTAS in a portfolio context.

5.2 Signaling

It is possible that some CTAs are talented and want to signal their ability, in the face of the lack of credible information and relevant benchmarks. More onerous contract terms for a CTA may signal a more talented fund manager, who is confident in his abilities. This private information may be conveyed contractually by agreeing with the investor to have a high-water mark and no lockup, as opposed to weak contract terms like no high-water mark and a lockup period, for example. A high-water mark (HWM) imposes discipline on the manager’s performance and if this performance is poor, the investor with no lockup can disinvest quickly.

Our data set contains information about the contract terms and fee structure. The contract terms for which we have data are high water mark (HWM) and lockup. It turns out that very few CTAs have lockup provisions, likely reflecting the fact that futures are very liquid markets. CTAs do show variation with respect to whether their contract includes a high water mark and their fees differ, although the fee structure of 2% fee on money-under-management and 20 percent of the gains above the high water market (“2-20’) predominates.

We look at this signaling hypothesis in Table 8, which shows the annualized average net-of-fees returns, the standard deviation of the those returns, and the Sharpe Ratio (excess return/standard deviation) (SR) for an equally-weighted CTA index of the CTAs with and without high water marks and for those with 2-20 fee terms

We restrict attention to the period starting in 2001 because prior to that most CTAs did not have high water marks, an observation discussed further below. The top part of the table shows the results for the period 2001-2007. The relevant comparison is between one of the two categories (ALL and 2-20) with a high water mark (HWM) to the same category with no high water mark (No-HWM). For example, in the case of ALL, the average CTA with a high water market had a Sharpe ratio of 0.48 while those with no high water mark had a Sharpe ratio of 0.41.

Table 8: Returns by Contract Type The table gives the annualized average returns (TR), standard deviations (Vol), and Sharpe Ratio (SR) of CTAs sorted by the presence of contract terms, High water Mark (HWM) and whether there is a 2-20 fee structure. The sample period is 2001/10 to 2007/12 and includes subperiods.

HWM NO-HWM ALL 2-20 ALL 2-20 TR 8.2% 3.4% 7.2% 3.3% 2001-2007 Vol 10.8% 10.1% 10.4% 11.2% SR 0.48 0.06 0.41 0.04

24 205

The table shows that the difference between the average returns on HWM and No HWM are not statistically different, for either ALL of the 2-20 fee structure. Overall, there appears to be no signaling. 25 The table also shows that the funds offering the 2-20 contract terms do not do as well as the population as a whole. It is not clear why this is the case. CTAs increasingly adopted hedge fund-like contracts starting around 2000, but it is not clear why those adopting 2-20 should be worse performers.

We noted above that CTAs have increasingly included high water marks in their contracts. During the period 1994 through 1996 no CTAs in our sample had high water marks in their contracts. Thereafter, the percentage rises almost monotonically to 71 percent by the end of 2007. This period also happens to coincide with an increase in the importance of hedge funds, which predominantly have high water marks in their contracts. One interpretation of the CTA behavior is that they were forced by hedge fund competition to include high water marks. This is consistent with investors receiving slightly better returns during this period. One way to measure this is to look at the performance fee as a percentage of gross returns, that is, what fraction of the gross return that is charged by the CTA as a performance fee? If hedge funds are a source of competitive pressure for CTAs, then this fraction should be going down. Looking only at the years where CTA gross returns are positive (which eliminates the years 1994 and 1999), the performance fee as a percentage of gross return averaged 28 percent for the period 1994 through 2000, and averaged 20 percent for the period 2001 through 2007. It appears that CTAs changed their contract terms to be more like those of hedge funds, and were forced to share a bit more with outside investors.

Without signaling, investors may not be able to distinguish talented CTAs from those without talent. But, the evidence suggests that there is a dearth of talent in the asset class generally. Perhaps talented CTAs do not enter the industry because they cannot differentiate themselves via signaling.

5.3 Information About the Information Problems

Investors may simply be unaware of the poor CTA performance. But, the information setting of investors is different during the period we study compared to the earlier period that Elton, Gruber and Rentzler (EGR) studied, the period from July 1979 to June 1985. EGR attributed the existence and persistence of poorly-performing CTAs to misinformation. Namely, the historical returns series provided in the prospectuses of public commodity funds were misleading and investors believed this misinformation. The past performance was unreasonably upward biased and investors had no other information to rely on.

Combining our evidence with that of EGR suggests that CTAs have been successful in taking money from investors without adding value for about twenty years. How can CTAs persist for so long despite their poor performance? To persist, there must be a demand for CTAs and a supply of CTAs. With regard to demand, we hypothesize that while investors are rational, acquiring information, overcoming the biases and lack of benchmarks, is costly and there is no common

25 A potential difficulty with looking at the period 2001-2007 is that different CTAs are in existence at different times, so an equally-weighted index is reflecting a potentially varying population. To address this we also looked at subperiods, where this issue is mitigated. Looking at the subperiods, there are no real differences and these results are omitted.

25 206

knowledge about the experience of poor returns. Understanding the supply of CTAs is perhaps easier. CTAs can earn fees on money-under-management, even if they have no ability to generate alpha. If they fail, they can easily restart, after erasing their prior history from the data bases. This suggests that they should have very high attrition rates, entering the industry to collect the fees while having difficulty surviving. We examine these hypotheses below.

5.3.1 Costly Information and a Lack of Common Knowledge

The information available to investors via formal offer documents (‘prospectuses’) and publicly available performance databases is fraught with biases and it is difficult to determine the true performance of CTAs. The databases that are available are not uniform, and not all of them allow for backfill bias to be corrected (see footnote 15, above). Evaluation of risk adjusted returns is further complicated by the absence of clear relevant benchmarks. The task of producing risk- adjusted performance evaluation of unbiased returns falls on the investor, a task that is costly, time consuming and requires analytical skills. Thus it is very costly for investors to have a view of CTAs that is different from that portrayed in Figure 1, where their performance looks very attractive. Given the available information, fraught with biases, and lack of relevant benchmarks, investors may simply believe that CTAs are a good investment.

But, costly information production can be only part of the story. For at least two decades investors have, on average, received poor returns. But, individually each investor may believe that his experience is simply bad luck. Since individual investors have no way of learning of the investment experience of other investors, information is never aggregated, so a true picture of the industry never emerges from actual experience. Although the publicity surrounding the EGR studies revealed poor performance, as discussed above, commodity fund mangers changed their organizational form, which may have allowed the industry to continue. The lack of aggregated actual experience of poor performance is necessary to keep investors from revising their view of Figure 1, which they may believe to be true. It seems unlikely that the same set of investors has been involved in investing in CTAs over that period. There must be new investors arriving, so that even when investors experience poor returns and withdraw their money, there are other investors willing to invest.

Table 9: A Comparison of CTA and Hedge Fund Fees. The table summarizes the fixed and variable component of fees for CTAs and Hedge Funds. For each category the table gives the average and the standard deviation expressed in percent per annum.

Management Fee Incentive Fee Average 2.15 19.50 CTAs Standard Deviation 1.22 6.32

Average 1.42 16.33 Hedge Funds Standard Deviation 0.51 6.84

Another, non-mutually exclusive, possible explanation could be the performance sensitivity of investors. Christoffersen and Musto (2002) analyze the money-fund industry and find that poorly- performing funds increase their fraction of performance-insensitive investors over time. They conclude that funds with bad performance should charge more from these investors, as their demand is price inelastic and a reduction in after-fee performance will not result in large outflow of money. Perhaps performance-insensitive investors in hedge funds and CTAs end up disproportionately at CTAs. If so, rational CTAs should charge higher fees. Are there differences

26 207

in fees between hedge funds and CTAs? Table 9 provides a look at the data. CTAs do appear to have higher management fees and slightly higher incentive fees than hedge funds. This suggests that the demand for CTAs is possibly less performance-sensitive and more price inelastic. Investors might be investing in CTAs for perceived diversification benefits and mandates for alternative investments (e.g. pension funds) and end up staying invested even in the face of poor performance.

5.3.2 Entry and Exit of CTAs into Fund Management

We saw above that CTA managers make money, even if investors do not, on average. This creates an incentive for new CTAs to enter the industry. There is little cost to entering the industry, only registration, which costs an almost trivial amount.26 There is no certification or any kind of screening. Poor performance results in liquidation of the fund, at little cost. Moreover, because a nonsurviving CTA can eliminate his history from the data bases retrospectively, there is no stigma to not surviving. It seems clear then that even if CTAs have no talent, in the sense of ability to generate alpha, they should persistently enter the industry because they earn fees on money-under-management, as long as they survive. If a CTA does not survive, he can restart with no history. We can shed some light on this by looking at CTA attrition rates.

Table 10: CTA Entry and Exit. The table summarizes by year-end the number of funds in the Lipper-TASS database (after correction for backfill and survivorship bias), the percentage of firms disappearing in the subsequent 12 and 24 months, the excess returns of the firms exiting in the subsequent 12 months, and the relative performance of the exiting funds compared to the sample average in the year of exit.

Date # Active 12-Month 24-Month ER Exits % Exits funds Attrition % Attrition % Next 12 Month below avg ret 1994 16 38 63 -3.74 66.67 1995 74 34 50 -12.20 60.00 1996 81 23 52 -18.86 52.63 1997 83 33 61 -26.01 70.37 1998 68 41 71 -19.10 42.86 1999 52 50 77 -17.25 50.00 2000 34 44 59 -12.98 46.67 2001 153 21 31 -10.71 56.25 2002 149 11 20 -19.38 76.47 2003 166 10 22 -21.65 56.25 2004 203 18 32 -7.51 58.33 2005 211 19 36 -8.80 57.50 2006 227 19 NA -7.11 40.48

Table 10 shows CTA attrition rates and the excess net returns for the funds that exit. The second column reports the number of active funds as of the date in the first column. The third and fourth columns report the attrition rates for these funds over the next twelve months and twenty four months, respectively. During a year, some funds exit. “ER Exits” is the annualized excess return, with respect to all the reporting funds, for the funds that exit during the year indicated. The

26 Prospective CTAs need to register with the National Futures Association. There is a nonrefundable fee of $200 and a fee of $85 for fingerprinting for each individual principal. See Hhttp://www.nfa.futures.org/registration/cta.aspH .

27 208

returns are calculated on a monthly basis, including all CTAs in existence during that month. The column entitled “% survivors below avg ret” shows the percentage of the exiting funds that had annualized returns below the equally-weighted annualized average of CTA month returns.

Table 10 indicates that CTA attrition rates are high, although they have declined recently. E.g., of the CTAs present on 31-Dec-05 (i.e., during 2006), 36 percent were gone by the end of 2007. Exiting funds are poor performers. E.g., during 2006, the exiting funds averaged an excess return of -8.8 percent annually.

In summary, the evidence is broadly consistent with the view that investors, facing high information costs with regard to evaluating CTA performance, believe that Figure 1 represents the investment opportunity. The poor performance experience of individual investors is not widely known. As a result investors continue to invest with CTAs and, recognizing this, CTAs continue to enter the industry, earning fees on money-under-management even though failure rates are very high.

6. Summary and Conclusions

Consumers and investors need information to rationally allocate their resources. Normally, we think of the price system as guiding these decisions. But, hedge funds are not publically traded, so there are no prices. There is only past performance data. In the case of hedge funds the available vendor data about their performance is biased, and there are few credible benchmarks for performance analysis. For these reasons, it has proven very difficult to evaluate the performance of hedge funds. These issues pose problems for investors as well as researchers as to whether hedge funds are an attractive asset class to invest in. They also potentially pose issues for public policy, to the extent that the hedge fund industry is sufficiently large to pose systemic risks.

We illustrate these issues by narrowing the universe of hedge funds to CTAs, because they are fairly homogeneous, their strategies are better known, and their strategy space is smaller. Using data between 1994 and 2007 from Lipper-TASS, we show that survivorship and backfill bias overstate the reported average return of CTAs by more than 8 percent per annum. Bias-corrected annualized average returns to investors were 4.9 percent, which is merely 85bp over the return on T-bills during this period. However, we estimate that gross average CTA returns (before fees) significantly exceed Tbill returns, which implies that funds retain most of their outperformance by charging fees. We propose simple dynamic futures-based trading strategies for performance evaluation. Because these strategies are in the public domain, they provide a natural hurdle that CTAs ought to overcome. Yet we find that the average CTA exhibits no skill (alpha) relative to these benchmarks. Combining our results with earlier studies by Elton, Gruber, and Rentzler, we conclude that poor CTA performance has persisted for at least twenty years. CTAs are a kind of market failure. Normally, asymmetric information is viewed as leading to an absence of a market. But, in the case of CTAs, the absence of information has led to the persistence of the market.

In Akerlof’s (1970) celebrated work, lemons problems result in market failure, the market does not exist. Our results suggest that CTAs are lemons and that this lemons market can persist. How can the CTA market persist? In Akerlof’s model, the information asymmetry is common knowledge; both sides of the market understand that there is an information asymmetry, namely, that car sellers have private information about the value of their cars. There is no way to signal car quality and because all cars sell at the same price there is an externality, namely, if a used car is sold some of the gains that should accrue to the sellers of good used cars accrue to the sellers of

28 209

bad used cars. This causes sellers of good used cars not to enter the market. Buyers can rationally make this calculation, and they do not buy used cars, knowing that any car in the used car market is a lemon. If there is a trade in the used car market, the price – in the limit –will be the value of a lemon. Like the used car market, CTAs cannot signal ability, but in other respects the situation is fundamentally different for CTAs. There appears to be no common knowledge of an information asymmetry. There are no prices to convey information. Investors appear to believe that Figure 1 represents an accurate portrayal of the performance history relative to a benchmark. As EGR point out, there is also an inability to short CTAs. Somewhat paradoxically the market for CTAs appears to be an example of a persisting lemons market. We argue that CTAs persist as an asset class despite their poor performance, because they face no market discipline based on credible information. There is no required disclosure as with SEC filings for firms or bank Call Reports. There is no regulation like that for mutual funds or banks. There are no private institutions that certify the managers’ competence (like the American Medical Association for doctors), or that certify their performance (like the Good Housekeeping seal of approval), and, as we have seen, no private repository of credible information for comparison purposes. Further, investors’ individual experience of poor performance is not common knowledge. In such a setting, it seems that some people can be fooled all of the time.

29 210

Appendix: Construction of Normative Benchmarks

Foreign Exchange

The literature on the forward bias in currency markets is the among the earliest studying “anomalies” in financial markets, and dates back to Rogoff (1979) and Bilson (1981) – see also Froot and Thaler (1990) for a summary of this early literature. The market for FX is a natural place to look for CTA trading strategies, as the futures market is large, and the surge in market activity since 2001 corresponds to a period when interest rate differentials favored investments in high interest rate currencies financed by short positions in low interest rate currencies (see Galati and Melvin (2004) on the increase in FX trading activity). Or, alternatively said, the environment favored the “carry trade,” in which an investor borrows in the low interest rate currency, and takes a long position in a high interest rate currency, speculating that the exchange rate will not change so as to offset the interest rate differential. Galati and Melvin (2004) show that FX turnover growth increases in interest rate differentials and with the magnitude of prior year’s exchange rate changes.27

In order to construct FX factors we employ the data for spot and one month forward prices against the US dollar for 15 currencies.28 The excess return from the end of the month t to the − FS ttt ++ 1,1 next is calculated as , where F tt +1, is the forward price at time t on a contract that F tt +1, 29 expires at the end of month t+1, and St+1 is the spot price at time t+1. The basis at the end of the month is defined as the difference between the current spot price and the current one month − FS ahead forward, expressed as a ratio to the current spot price: ttt +1, . St At the end of each month, we construct currency basis portfolios by ranking all currencies on their basis (interest rate differential) relative to the US dollar. Currencies in the top half of this ranking are assigned to the high basis portfolio and the bottom half of the currencies to the low basis portfolio. Both portfolios are equally weighted. The positions are rebalanced monthly so that the high (low) basis portfolio represents a dynamically rebalanced portfolio of currencies with the highest (lowest) interest rate differential relative to the US dollar.

Currency momentum portfolios are similarly constructed by ranking currencies by their on prior 3-month excess returns. At the end of each month currencies are assigned to High and Low

27 Also, see Galati and Heath (2007) and Galati, Heath and McGuire (2007). 28 The currencies used are: AUD (Australian Dollar), CAD (Canadian Dollar), CHF (Swiss Franc), DKK (Danish Kroner), DEM (Deutsche Mark), EUR (Euro), FRF (French Frank), GBP (British Pound), IEP (Irish Pound), ITL (Italian Lira), JPY (Japanese Yen), NLG (Netherland Guilder), NZD (NZ Dollar), NOK (Norwegian Kroner), and SEK (Swedish Kroner). 29 For some cases the forward contract trading at time t does not expires at the last day of next month, for example if the last day of next month is a Friday the forward contract might expire on the Monday. In such cases an interpolation rule following interest parity is used to figure out the forward price F tt +1, . Let F , +ktt be the forward price at time t on a contract that expires on t+k, and we are interested in F , + jtt which corresponds to the last day of next month, then: ⎛ j ⎞ ln()ln()SF =− ⎜ ⎟ {ln* ( )− ln()SF }. , + jtt t ⎝ k ⎠ , +ktt t

30 211

currency momentum portfolios, which are equally-weighted and held for one month subsequent to ranking after with time they are rebalanced.

Commodities

The information content of the futures basis for expected risk premia has been documented empirically by Fama and French (1987) and more recently by Erb and Harvey (2006) and Gorton and Rouwenhorst (2006). Commodity price momentum has been documented by Pirrong (2005), Erb and Harvey (2006), Miffre and Riallis (2007), and Gorton, Hayashi and Rouwenhorst ((2007). Our construction of dynamic commodity portfolios mirrors Gorton, Hayashi, and Rouwenhorst (2007) who argue that the excess returns to portfolios sorted by the basis and prior returns in part stems from selecting commodities when inventories are low. Based on the Theory of Storage, GHR show that prior returns (“momentum”) and the futures basis (“backwardation”) are price- based signals of low physical inventory levels, and the excess returns are a compensation for the increased volatility of commodity prices.

At the end of each month, available commodities futures are ranked on the basis, defined as the annualized slope of the futures curve between the nearest and the next-to-nearest to maturity contracts. High and low basis portfolios are constructed from the top and bottom half of the commodities in this ranking. All portfolios are equally weighted and rebalanced monthly. Similar to the Basis portfolios, we construct monthly rebalanced equally-weighted High and Low Commodities momentum portfolios by ranking commodities on prior 1-year returns.

Equities

We construct a momentum and a value factor by sorting country index returns on prior return and book-to-market. Momentum in country equity index returns has been documented by Asness et al. (1997), and Chan et al. (2000), and studied more recently by Bhojraj and Swaminathan (2006). The profitability of value strategies has been documented by Asness (1997). Related papers on performance reversals include Richards (1997) and Balvers et al. (2000).

At the end of each month we sort available country equity index futures by country-wide measures of book-to-market (Value) or prior 12-month return (Momentum). For each of these sorts we construct High and Low Value and Momentum portfolios containing the top and bottom half of constituents of this ranking.

31 212

References

Ackermann, Carl, Richard McEnally, and David Ravenscraft (1999), “The Performance of Hedge Funds: Risk, Return, and Incentives,” Journal of Finance 54(3), p. 833-874.

Akerlof, George (1970), “The Market for ‘Lemons”: Quality Uncertainty and the Market Mechanism,” Quarterly Journal of Economics 84(3), p. 488-500.

Arrow, K. J. (1971), Essays in the Theory of Risk-Bearing (Chicago: Markham Publishing Co.).

Asness, Clifford S. (1997), “The Interaction between Value and Momentum Strategies,” Financial Analyst Journal 53(2), pp 29-36

Asness, Cliff S., John M. Liew, and Ross Stevens (1997), “Parallels Between the Cross-Sectional Predictability of Stock and Country Returns,” Journal of Portfolio Management (Spring), p.79-87.

Balvers, Ronald, Yangru Wu and Erik Gilliland (2000), “Mean Reversion across National Stock Markets and Parametric Contrarian Investment Strategies,” Journal of Finance, 55(2), pp. 745-772.

Barberis, Nicholas and Ming Huang (2007), “Stocks as Lotteries: The Implications of Probability Weighting for Security Prices,” Yale School of Management, working paper; American Economic Review, forthcoming.

Bhojraj, Sanjeev and Bhaskaran Swaminathan (2006), “Macromomentum: Returns Predictability in International Equity Indices,” Journal of Business 79(1), p. 429-451.

Bilson, John (1981), “The Speculative Efficiency Hypothesis,” Journal of Business 54, 433-451.

Brown, S. and W.N. Goetzmann (1997), “Mutual Fund Styles,” Journal of Financial Economics 43(3), p. 373-99.

Brown, Stephen and William Goetzmann (2003), “Hedge Funds with Style,” Journal of Portfolio Management 29(2), p. 101-112.

Brown, Stephen J., William N. Goetzmann, Roger G. Ibbotson (1999), “Offshore Hedge Funds: Survival and Performance 1989-1995,” Journal of Business 72, p. 91-118.

Brown, Stephen J., William N. Goetzmann, Roger G. Ibbotson, and Stephen A. Ross (1992), “Survivorship Bias in Performance Studies,” Review of Financial Studies 5(2), p. 553- 580.

Brown, Stephen J., William N. Goetzmann, and Bing Liang (2004), “Fees on Fees in Funds of Funds,” Journal of Investment Management 2, 39-56.

Chan, Kalok, Allaudeen Hameed and Wilson Tong (2000), “Profitability of Momentum Strategies in the International Equity Markets,” The Journal of Financial and Quantitative Analysis 35(2), pp. 153-172.

32 213

Christoffersen, Susan and David Musto (2002), “Demand Curves and the Pricing of Money Management,” Review of Financial Studies 15, 1499-1524.

Cremers, Martijn, Antti Petajisto, and Eric Zitzewitz (2008), “Should Benchmark Indices Have Alpha? Revisiting Performance Evaluation,” working paper Yale School of Management.

DeRoon, Frans, Jenke TerHorst, and Theo Nijman (2003), “Evaluating Style Analysis,” Journal of Empirical Finance vol. 11, no. 1 (January), p. 29-53.

Edwards, Franklin and M.O. Caglayan (2001), “Hedge Fund Performance and Manager Skill,” Columbia University, working paper.

Edwards, Franklin and Jimmy Liew (1999), “Managed Commodity Funds,” Journal of Futures Markets Vol. 19, No. 4, p. 377-411.

Elton, Edwin, Martin Gruber, and Joel Rentzler (1990), “The Performance of Publicly Offered Commodity Funds,” Financial Analysts Journal (July/August), p. 23-30

Elton, Edwin, Martin Gruber, and Joel Rentzler (1989), “New Public Offerings, Information, and Investor Rationality: The Case of Publicly Offered Commodity Funds,” Journal of Business 60(2), p. 175-199.

Elton, Edwin, Martin Gruber, and Joel Rentzler (1987), “Professionally Managed, Publicly Traded, Commodity Funds,” Journal of Business 60(2), p. 175-199.

Erb, Claude and Campbell Harvey (2006), “The Strategic and Tactical Value of Commodity Futures,” Financial Analysts Journal 62: 69-97.

Fama, Eugene F. and Kenneth R. French (1993), “Common Risk Factors in the Returns on Stocks and Bonds,” Journal of Financial Economics 33, p. 3-56.

Fama, Eugene and Kenneth French (1987), “Commodity Futures Prices: Some Evidence on Forecast Power, Premiums, and the Theory of Storage,” Journal of Business 60: 55-73. Fama, Eugene F. and Kenneth R. French (1992), “The Cross-Section of Expected Stock Returns,” Journal of Finance 47(2), p. 427-465.

Farhi, Emmanuel and Xavier Gabaix (2008), “Rare Disasters and Exchange Rates,” NBER Working Paper No. 13805.

French, Kenneth (2008), “The Cost of Active Investing,” Dartmouth College, working paper, Journal of Finance, forthcoming.

Froot, Kenneth A., and Richard H. Thaler (1990), “Anomalies: Foreign Exchange,” The Journal of Economic Perspectives, 4(3), 179-192.

Fung, William and David Hsieh (1997), “The Information Content of Performance Track Records: Investment Style and Survivorship Bias in the Historical Returns of Commodity Trading Advisors,” Journal of Portfolio Management, 24(1), p. 30-41.

33 214

Fung, William and David Hsieh (2000), “Performance Characteristics of Hedge Funds and Commodity Funds: Natural vs. Spurious Biases,” Journal of Financial and Quantitative Analysis vol. 35(3), p. 291-307.

Fung, William and David Hsieh (2001), “The Risk in Hedge Fund Strategies: Theory and Evidence from Trend Followers,” Review of Financial Studies 14(2), p. 313-341.

Fung, William and David Hsieh (2002a), “Asset-Based Style Factors for Hedge Funds,” Financial Analysts Journal 58(5), p. 16-27.

Fung, William and David Hsieh (2002b), “Hedge Funds: An Industry in Its Adolescence,” Federal Reserve Bank of Atlanta Economic Review (Fourth Quarter), p. 1 - 34.

Fung, William and David Hsieh (2004), “Hedge Fund Benchmarks: A Risk-Based Approach”, Financial Analysts Journal, 60(5) p. 65-80.

Galati, Gabriele and Michael Melvin (2004), “Why has FX Trading Surged? Explaining the 2004 Triennial Survey,” Bank for International Settlements, Quarterly Review (December), p. 67 - 74.

Galati, G, A Heath and P McGuire (2007): “Evidence of Carry Trade Activity”, Bank for International Settlements, Quarterly Review (September), pp 27–41.

Galati, Gabriele and Alexandra Heath (2007), “What Drives the Growth in FX Activity? Interpreting the 2007 Triennial Survey,” Bank for International Settlements Quarterly Review (December), p. 63 - 72.

Gorton, Gary and K. Geert Rouwenhorst (2006) “Facts and Fantasies about Commodity Futures,” Financial Analysts’ Journal 62(2), p. 47-68.

Gorton, Gary, Fumio Hayashi and K. Geert Rouwenhorst (2007), “The Fundamentals of Commodity Futures Returns,” http://papers.ssrn.com/sol3/papers.cfm?abstract_id=996930, working paper.

Hall, Dustin (2008), “The Elephants in the Room: Dangers of Hedge Funds in Our Financial Markets,” Florida Law Review 60, p. 183-227.

Hasanhodzic, Jasmina and Andrew Lo (2007), “Can Hedge Fund Returns Be Replicated? The Linear Case,” Journal of Investment Management 5(2), p. 5-45.

Jensen, Michael (1968), “The Performance of Mutual Funds in the Period 1945-1964,” Journal of Finance 23(2), p. 389-416.

Levy, Haim and Sarnat, Marshall (1984), Portfolio and Investment Selection: Theory and Practice (Prentice-Hall International).

Liang, Bing (2000), “Hedge Funds: The Living and the Dead,” Journal of Financial and Quantitative Analysis 35, p. 309-326.

Malkiel, Burton and Atanu Saha (2005), “Hedge Funds: Risk and Return,” Financial Analysts Journal 61(6), p. 80-88.

34 215

Markowitz, H. (1952), “The Utility of Wealth,” Journal of Political Economy 60(2), p. 152–158.

Miffre, Joëlle and Georgios Rallis (2007), “Momentum in Commodity Futures Markets,” Journal of Banking and Finance 31(6), p. 1863- 1886.

Mulvey John., S.S.N. Kaul and Koray Simsek (2004), “Evaluating a Trend-Following Commodity Index for Multi-Period Asset Allocation,” Journal of Alternative Investments 7(1), p. 54-69.

Park, James (1995), “Managed Futures as an Investment Asset,” PhD. Dissertation, Columbia University.

Pirrong, Craig (2005), “Momentum in Futures Markets,” Bauer College of Business, University of Houston, working paper.

Polkovnichenko, V. (2005), “Household Portfolio Diversification: A Case for Rank-Dependent Preferences,” Review of Financial Studies 18, 1467–1502.

Posthuma, Nalke and Pieter Jelle van der Sluis (2003), “A Reality Check on Hedge Fund Returns,” available at SSRN: http://ssrn.com/abstract=438840 .

Richards, Anthony J. (1997), “Winner-Loser Reversals in National Stock Market Indices: Can They be Explained?,” Journal of Finance 52(5), pp 2129-2144.

Rogoff, Kenneth S. (1979), “Essays on Expectations and Exchange Rate Volatility,” PhD. Dissertation, Massachusetts Institute of Technology.

Rosenthal, Leonard, and Colin Young (1990), “The Seemingly Anomalous Price Behavior of Royal Dutch/Shell and Unilever N.V./PLC,” Journal of Financial Economics 26, 123-141.

Sharpe, William (1992), “Asset Allocation: Management Style and Performance Measurement,” Journal of Portfolio Management (Winter), p. 7-19.

Siri, Erik and Peter Tufano (1998), “Costly Search and Mutual Fund Flows,” Journal of Finance 53(5), 1589-1622.

Waksman, Sol (2000), “Commodity Trading Advisor Survey: Adding Equities to A Managed Futures Portfolio,” Journal of Alternative Investments (Fall), p. 43-44.

35 216 OxfordResearchGroup | September 2008

International Security Monthly Briefing – September 2008

THE FINANCIAL CRISIS AND SUSTAINABLE SECURITY Paul Rogers

Introduction

Oxford Research Group’s International Security Monthly Briefings focus primarily on issues such as the conflicts in Iraq, Afghanistan and Pakistan, the evolution of western counter-terrorism policies and the development of the al-Qaida movement. On occasions they also cover matters such as energy security, climate change and world food prospects. In view of the serious financial situation that has developed in recent months, this briefing provides an initial analysis of the possible impact of the crisis on security.

This is undertaken in the context of ORG’s work on sustainable security which, in turn, is predicated on an underlying analysis of the security issues that are likely to be most prominent in the next two to three decades. This assesses that there are four main trends that are particularly salient.

Firstly, global socio-economic divisions are widening, with most of the benefits of the past three decades of economic growth being concentrated in the hands of a trans-global elite community of about 1.2 billion people, mainly in the countries of the Atlantic community and the West Pacific, but with elite communities in the tens of millions in countries such as China, India and Brazil. Improvements in education, literacy and communications in recent decades have increased the awareness of many marginalised people of this unjust distribution of wealth. In extreme circumstances this can lead to the rise of violent and extreme social movements such as the Naxalites in India.

Secondly, climate change is expected to have profound effects on that majority of the world’s population living in the tropical and sub-tropical regions but without the economic resources to respond to severe storms, rising sea levels and drastic changes in rainfall distribution. Increased migration and social and political unrest are likely consequences.

Thirdly, resource competition, especially over energy resources in the Persian Gulf region and elsewhere, will, on present trends, be an increasing source of tension and conflict.

Lastly, the strong tendency of powerful elites to maintain security, by military force if necessary, is expected to be counter-productive, as has already been seen by many of the consequences of the war on terror.

Countering such trends involves a fundamental commitment to emancipation and socio-economic justice. This includes fair trade, debt cancellation, assistance for sustainable development, a radical cut in carbon emissions, rapidly increased use of renewable energy resources and the development of conflict prevention and conflict resolution policies that avoid the use of force.1

The Financial Crisis and Historical Experience

The current crisis has three main characteristics:

• It is global. While most emphasis was initially on the sub-prime market in the United States, the crisis has spread rapidly through the UK and across Europe, has resulted in a 60% fall in the Shanghai stock market in a year, steep stock market falls across much of Latin America and bank crises in Australia and New Zealand.

1 Chris Abbott, Paul Rogers and John Sloboda, Beyond Terror: The Truth About the real Threats to Our World (London: Random House, 2008).

1 217

OxfordResearchGroup | September 2008

• It has initially been a crisis of liquidity and confidence in the financial sector rather than a decline in industrial and retail activities, but it is expected to have a substantial effect on industrial and commercial output as sources of investment finance diminish. • It is likely to last at least two years, with several more years for recovery.

Prior to the sub-prime issue, the international economy was already affected by rapid oil price rises and a more general bull market of rising prices for primary commodities. One early effect was a substantial increase in food prices that had a particular impact on poorer communities across the world.2

The 2008 crisis is not directly comparable to the 1929 Wall Street crash, which was not truly global, nor to the 1987 stock market problems in Europe nor even the widespread Asian downturn of the 1990s, even though these had some global ramifications. The only comparable previous global crisis was that of 1973-74. Then, unilateral moves by Arab oil producers during the Yom Kippur/Ramadan War of October 1973 precipitated a remarkable 450% increase in oil prices within ten months, resulting in an unusual combination of economic stagnation and inflation. A parallel issue was a rapid increase food prices.

The worst excesses of the food crisis were averted partly by a decline in food prices because of the onset of the recession, together with some emergency aid coming from some of the newly-rich oil producers. The experience of stagflation in industrial countries resulted less in a move towards mixed economies with a higher level of state planning, and more to the development of free-market ideas, not least in the form of what was later termed “Reaganomics”.

The 25 years from 1980 saw the rapid development of free market globalisation that stimulated substantial economic growth but with a notable lack of socio-economic justice as wealth-poverty divisions widened. Towards the end of the 2000s, the combination of oil price rises and economic overextension, especially in sub-prime markets, resulted in a transnational banking crisis.

Impacts on Poorer Communities

Some aspects of the current crisis will have relatively little impact on poorer communities. For example, oil prices are unlikely to maintain their current levels as demand falls due to a decline in economic activity. It would be possible for oil producers to act together to maintain higher prices but this is unlikely for two reasons. One is that the Organisation of Petroleum Exporting Countries (OPEC) does not have sufficient political unity to exert control over the market, unlike the 1973-74 period. The second is that sovereign wealth funds and other investment vehicles of some oil-rich states depend on buoyant stock markets in North America, Europe and East Asia. Maintaining very high oil prices would therefore tend to damage such investments.

Lower oil prices will be of some help to poorer countries struggling to meet the higher costs of their oil imports. Furthermore, the oil price decline is also likely to have some impact on food prices, resulting in price decreases that will be of some help to the poorest communities.

These, though, are among the few aspects of the current economic environment that might have some limited advantage for poorer people. In almost every other respect the outcome is less positive. For example, weak economies in major countries that normally attract migrant labour mean that remittances from labourers to their home countries will diminish as jobs become scarcer and wages decline. Such remittances are not just important across South and Southeast Asia but are also important for several Latin American economies. Furthermore, increases in unemployment in countries that are destinations for migrant workers will lead to a reaction against such workers, as has been seen recently in South Africa. This can readily extend to increased support for far-right political parties.

2 See the May 2008 briefing, Food Poverty and Security.

2 218

OxfordResearchGroup | September 2008

Another early consequence of an economic downturn will be a decline in international tourism and travel. Whatever the damaging impact that tourism can have on poorer communities, it is still the case that there is some monetary transfer involved, and some poorer states depend heavily in such earnings.

More importantly, a decline in economic activity will have a substantial impact on commodity prices, affecting the export earnings for poorer countries for a wide range of commodities including copper, tin, coffee, tea, sugar, cotton and hardwoods. Even now, many southern countries depend on such commodity exports for the substantial majority of all their export earnings. It is here that previous experience is particularly relevant to the current crisis and its effects on the majority world.

A New International Economic Order?

In the early 1970s, a substantial increase in commodity prices put the economies of most industrialised countries under some strain, especially as they were also experiencing the oil price rises. The UN Conference on Trade and Development (UNCTAD) had already been attempting to bring a degree of planning into world commodity markets, encouraging individual commodity agreements for products such as coffee and tin that were designed to provide stability along with some slow but steady increases in prices. Such agreements were seen as helping to alter the terms of trade between third world and industrialised states in a manner that would greatly improve the development prospects of the former.

By early 1974 the wild fluctuations in commodity markets were so marked that industrialised countries such as the US, UK, France and Japan were ready to accept the need for international market planning, and a Declaration on a New International Economic Order was agreed at a special session of the UN General Assembly in April 1974. The core of this proposal was the Integrated Programme for Commodities (IPC) which would bring in a series of linked commodity agreements backed by a Common Fund to finance the setting up of the necessary commodity buffer stocks. At the time many development economists believed that the IPC could provide a really valuable boost to the development prospects of many poorer countries, helping to bring in a new era of fair trade. However, that it was even proposed was mainly due to the temporary problems being faced by the world’s wealthy economies.

In the event, the increase in oil prices resulted in a decrease in industrial activity, a fall in primary commodity prices and, almost immediately, a general loss of interest in the IPC by the major industrial powers. What was eventually established, later in the 1970s, was a pale shadow of the original programme, and this had little impact as the era of the free market unfolded in the 1980s. The loss of that programme is a reminder that, in times of economic downturn, the prospects for poorer communities rarely loom large in the recovery policies of the world’s wealthy states.

Such behaviour has reached almost grotesque proportions in the current crisis, with wealthy states willing to commit more than a trillion dollars to rescue their own banking system in crisis. These are financial outlays that are enormous when compared with those that are being committed to achieving the United Nations Millennium Development Goals.

Responding to the Current Crisis

Although the current crisis does not have direct historical parallels, the marked tendency will therefore be for the most powerful economies to engage primarily in responding to their own problems. Much of this will be at the level of individual states, such as recent US Government intervention in the mortgage and insurance markets and numerous interventions across Europe. There is also likely to be some degree of cooperation between the more wealthy states of the North Atlantic community, drawn mainly from the members of the OECD.

Previous experience indicates that the emphasis will be almost entirely on domestic concerns rather than the wider global community. While this might provide some relief among the poorer sectors of the

3 219

OxfordResearchGroup | September 2008

populations of wealthy states, it will do nothing to help the much larger numbers of impoverished people of the majority world. Moreover, further action to limit third world debt is unlikely and there will almost certainly be pressure on aid budgets. Even key issues such as climate change and the risk of resource conflict are likely to slip down the political agenda.

The implications of this are serious, in that any hindrance to facilitating sustainable development across the countries of the South will increase human insecurity and suffering. Furthermore, any limitation in addressing the urgent issue of climate change will just add to the problems of the South as the damaging effects of climate change increase their impact. Some of the most fragile of the world’s economies, from much of Africa through to Southwest Asia, will suffer most from economic recession and the impact of climate change. More generally, the bitterness that already exists across continents will be reinforced by a perception that the dominant economies have little or no interest in the majority of the world’s people.

Even so, it is just possible that the current crisis will be seen to necessitate a serious reconsideration of how the world economy has evolved in the past three decades. In essence, the nature of the globalised free market is being called into question amidst demands for considerable reforms. The reason for this may well be the manner in which the free market has allowed the current crisis of liquidity and confidence to develop in the wealthy economies, rather than that the free market has increased socio- economic divisions. The extent to which the reforms will be instituted will depend to some extent on the depth and duration of the current crisis but at the time of writing (early October) there are indications that it could well be severe and prolonged.

What could come out of this might be reforms that not only respond to the crisis in the western banking system but also address the deeper global inequalities that have developed in recent years. For that to happen there will need to be a degree of political wisdom on the part of some national governments, accompanied by visionary proposals by intergovernmental agencies, such as some of the specialised agencies of the United Nations. There are notable past examples of this, not least the Prebisch Plan on trade and development in 1963 that prompted UNCTAD’s early work, the UN Environment Programme’s work on ozone depletion in the mid-1980s and recent intergovernmental work on climate change.

However, there is little prospect of effective change if it is left solely to governments and intergovernmental agencies. The richer states will look to their own predicaments, and their influence in intergovernmental organisations may limit new proposals. What is essential is the sustained action of nongovernmental organisations as part of a wider civil society. Responding to the current crisis can either be a process limited to the narrow domestic concerns of the most powerful states or it can be seen as an opportunity for reform of the world’s economic system that will benefit the majority world. The timescale is the next two to five years, the likely duration of the current crisis, and the stakes are high.

Paul Rogers is Professor of Peace Studies at the University of Bradford and Global Security Consultant to Oxford Research Group (ORG). His international security monthly briefings are available from the ORG website at www.oxfordresearchgroup.org.uk, where visitors can sign-up to receive them via email each month. These briefings are circulated free of charge for non-profit use, but please consider making a donation to ORG if you are able to do so.

Copyright © Oxford Research Group, 2008 Some rights reserved. This briefing is licensed under a Creative Commons Attribution-No nCommercial-NoDerivs 3.0 Licence. For more information please visit http://creativecommons.org/licenses/by-nc-nd/3.0/.

4 220

Federal Reserve Bank of New York Staff Reports

Credit Derivatives and Bank Credit Supply

Beverly Hirtle

Staff Report no. 276 February 2007 Revised March 2008

This paper presents preliminary findings and is being distributed to economists and other interested readers solely to stimulate discussion and elicit comments. The views expressed in the paper are those of the author and are not necessarily reflective of views at the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the author. 221

Credit Derivatives and Bank Credit Supply Beverly Hirtle Federal Reserve Bank of New York Staff Reports, no. 276 February 2007, revised March 2008 JEL classification: G21, G32

Abstract

Credit derivatives are the latest in a series of innovations that have had a significant impact on credit markets. Using a micro data set of individual corporate loans, this paper explores whether use of credit derivatives is associated with an increase in bank credit supply. We find evidence that greater use of credit derivatives is associated with greater supply of bank credit for large term loans—newly negotiated loan extensions to large corporate borrowers—though not for (previously negotiated) commitment lending. This finding suggests that the benefits of the growth of credit derivatives may be narrow, accruing mainly to large firms that are likely to be “named credits” in these transactions. Further, the impact is primarily on the terms of lending—longer loan maturity and lower spreads—rather than on loan volume. Finally, use of credit derivatives appears to be complementary to other forms of hedging by banks.

Key words: credit derivatives, risk management, credit supply, bank lending

Hirtle: Federal Reserve Bank of New York (e-mail: [email protected]). The author thanks Matt Botsch and Sarita Subramanian for their assistance in gathering the data for the paper; Adam Ashcraft, Mark Flannery, James Vickery, Philip Strahan, and seminar participants at the Federal Reserve Banks of New York and San Francisco for helpful comments and suggestions; and William English for making data from the Survey of Terms of Business Lending and the Senior Loan Officers Opinion Survey available. The views expressed in this paper are those of the author and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. 222

Credit Derivatives and Bank Credit Supply

I. Introduction

The market for credit derivatives has grown enormously in recent years. Notional amounts of credit derivatives reached $45.5 trillion as of mid-2007, a 50-fold increase from the level at mid-year 2001 (International Swaps and Derivatives Association 2007). The development of these instruments is an important innovation, the latest of a series of innovations such as loan sales in the 1980s and securitizations in the 1990s that have had a significant impact on the nature and operation of credit markets. Like these earlier innovations, a key property of credit derivatives is that they separate the origination of credit, the funding of credit, and the holding and management of credit risk.

This separation has implications for the distribution of credit risk across the financial system and, in turn, for the supply of credit. Banks that originate credit to corporate borrowers need no longer hold the credit risk associated with these loans, while other financial firms can hold credit risk without having to originate or fund the underlying credit.

The risk diversification potential of credit derivatives has been widely discussed and acknowledged. But a follow-on question is whether greater risk diversification, both within the banking industry and between banks and other financial institutions, has led to an increase in the supply of credit. To what extent has the ability to spread credit risk outside the banking system allowed banks to originate and hold more credit? Have banks used the diversification potential of credit derivatives to reduce their overall risk exposures, or have they “undone” diversification- related risk reductions by expanding their loan portfolios? If credit supply has expanded, which borrowers have benefited?

Research examining earlier credit market innovations such as loan sales and securitizations has generally found that banks have used opportunities to diversify credit risk exposures to increase lending (Cebenoyan and Strahan 2004, Franke and Krahnen 2005, Goderis 223 2 et al. 2006). This paper extends this previous work by examining the impact of banks’ use of credit derivatives on their credit supply. Using a confidential data set of thousands of commercial and industrial (C&I) loans made by a sample of U.S. banks between 1997 and 2006, we look at how the volume of new loans changes as a bank purchases more protection through the credit derivatives market. A key advantage of this data set is that we can look not only at the volume of credit provided, but also at lending terms and the characteristics of the new lending, including credit spreads and the maturity of loans. We can also separate the loans by borrower size

(proxied by the size of the loan) and by the type of lending arrangement (term lending versus lending under previously negotiated commitments).

We find evidence that banks increase the supply of credit as they obtain additional credit protection through credit derivatives, but only for certain types of loans and borrowers. The evidence is strongest for large term loans: average maturities increase, new loan flows (weakly) increase, and spreads fall as credit derivatives protection rises. Further, the impact on loan maturities and spreads is economically meaningful, with a one standard deviation increase in credit derivatives protection implying a six-month increase in average loan maturity and a 21 basis point drop in spreads. For small term loans and lending done under loan commitment, however, the results are decidedly more mixed, with offsetting positive and negative effects of increased credit derivatives protection across new loan amounts, maturities, and spreads. Even in the cases where the results indicate a positive effect on supply of these types of loans, the impact is economically small.

The term loan results may be the most relevant for assessing the impact of credit derivatives on new credit supply, since lending terms on commitment loans may reflect arrangements that were negotiated months, or even years, prior to the loan extension, and the timing and size of extensions is largely determined by borrowers. Further, since large firms are more likely to be “named credits” in the credit derivatives market, the findings suggest that the benefits of credit derivatives may accrue mainly to these firms, rather than being spread more 224 3 broadly across the business sector. It is also interesting to note that the most significant impact comes on the terms of lending – loan maturity and loan spreads – rather than on the volume of lending.

Finally, the extent to which a bank engages in other forms of hedging – particularly, its holdings of financial (non-credit) derivatives – affects the impact of credit derivatives on credit supply. Hedging via credit derivatives appears to be complementary to other forms of hedging in that credit supply tends to increase more in response to an increase in credit derivatives protection for banks that are active hedgers in other arenas.

The remainder of the paper is organized as follows. The next section reviews previous work on credit derivatives and other credit market innovations on the supply and terms of credit.

Section III discussed the confidential loan data and empirical specification used in the analysis, while Section IV presents the key results. Section V concludes.

II. Credit Market Innovation and Bank Credit Supply

In the traditional model of bank lending, the bank performs all aspects of the credit process: originating the loan, holding it on the balance sheet (funding it), and holding and managing the associated credit risk. Credit market innovations in the 1980s and 1990s altered this model in some significant ways. In particular, innovations such as loans sales, syndications, and securitizations separated the process of loan origination – establishing a relationship with the borrower, gathering and analyzing information about the borrower’s creditworthiness, and establishing the terms of the loan – from funding the loan. These arrangements sometimes also removed the credit risk associated with the loan, though the originating bank frequently provides credit guarantees or holds a first-loss or other recourse position that retains some portion of the credit risk exposure.

Credit derivatives provide another way to de-couple the various aspects of the credit process. The key feature of these contracts as compared to earlier credit risk transfer arrangements is that credit derivatives separate credit risk from both origination and funding. 225 4

Credit derivatives are financial contracts that allow one party – the guarantor – to assume the credit risk associated with a particular debt obligation or with a group of debt obligations from another party – the beneficiary. The debt obligations in question are largely those of commercial and industrial firms, particularly large, investment grade corporate borrowers.1 The risk transfer can involve contingent payments from the guarantor to the beneficiary related to a credit event such as a bankruptcy, default or rating downgrade (as in a credit default swap), or transfer of the cash flows from a reference asset such as a loan or bond in return for interest payments (as in a total return swap).2 In theory, credit derivatives allow parties to take on credit risk without doing any actual lending, or to do lending without assuming any credit risk.3

Credit derivatives allow credit risk transfer within the banking system and also between banks and non-bank financial institutions. This risk transfer is frequently cited as a stabilizing factor in the financial system, reducing concentrations of exposures at individual banks and spreading credit risk more widely to those parties best able to hold it (Geithner 2006, Greenspan

2005). A recent study by an international group of banking, securities, and insurance regulators argues that risk transfer at some individual firms may be substantial, though it suggests that the overall extent of risk transfer is relatively small, at least in comparison to total credit risk in the banking system (Joint Forum 2005).

For individual banks, the ability to transfer or assume credit risk via credit derivatives facilitates risk management and the optimal use of bank capital. The key idea is that in managing their loan portfolios, banks make a series of related decisions about how much to lend, the terms on which loans will be made, and how the risk of those loans will be managed (Froot and Stein

1998, Duffee and Zhou 2001). As the discussion above suggests, innovations in credit markets

1 Some credit derivatives are also written on sovereign creditors and, increasingly, on credit market indices. 2 See Lopez (2001) for a more detailed description of particular types of credit derivatives. 3 This is an overstatement, even in theory, since banks assume counterparty credit risk when entering into a credit derivative contract. Counterparty credit risk is the risk that the party on the other side of the derivative will fail to perform on the terms of the contract. For a broader discussion of the operational risks to banks of using credit derivatives, see Gibson (2007). 226 5 have considerably widened the options available for managing credit risk. Duffee and Zhou

(2001), for instance, argue that credit derivatives allow banks to reduce credit risk exposures in a more flexible, dynamic way over the life of a loan as compared to loan sales, which are for the loan’s full term. Instefjord (2005) suggests that banks may be willing to take on credit risk because liquid credit derivatives enable them to “offload” this risk if necessary.4

Decisions about whether to lend to a particular borrower or group of borrowers, how much to lend, and the terms of that lending thus may be made jointly with the decision about how to manage the risk of the resulting loan: to keep it on the books and accept the risk, to sell it or securitize it, or to keep it and hedge some or all of the credit risk. All of these decisions are made in the context of the bank’s overall appetite and capacity for additional risk exposure, in particular, with regard to the extent of economic and regulatory capital capacity to absorb additional risk (Froot and Stein 1998). Risk-reducing mechanisms such as credit derivatives can result in lower allocated capital for a given scale of business or can allow a business to expand its activities for a given amount of capital.

Thus, a bank’s decision to lend and on what terms depends on a series of factors, including the strength of loan demand and the quality of investment opportunities, the bank’s capital position, and its ability to hedge or disperse risk. Regarding the last of these, it is important to note that a can bank determine and execute its hedging strategy in several ways. For instance, a bank could make lending and hedging decisions on a flow basis relative to individual borrowers – so that it might be willing to lend to that particular borrower because it can hedge that borrower’s risk using a credit derivative (Gibson 2007, Bomfim 2005). The bank could also

4 The growth of the market for credit derivatives may alter conditions in markets for alternative credit risk management mechanisms such as loan sales and syndications. For instance, Duffee and Zhou (2001) argue that banks’ use of credit derivatives could alter the characteristics of the pool of loans offered in the loan sales market in ways that worsen the adverse selection problem and undercut the effectiveness of loan sales in reducing the cost of distress to banks. Morrison (2005) argues that because banks’ use of credit derivatives is less transparent than their loan sales activity – in the sense of knowing which borrowers are affected – that the certification value of bank loans could be undercut, resulting in disintermediation from the banking system, as borrowers choose to issue low quality bonds rather than take on bank debt. 227 6 engage in ex ante hedging, essentially “keeping its powder dry” to be able to expand lending when good opportunities arise. In the first scenario, firm benefit from credit derivatives only if they are “named credits,” while in the second, the impact is felt by a much wider range of borrowers. But in both scenarios, the ability to hedge the risks associated with some borrowers has a positive impact on the supply of bank loans.

The question that emerges in practice is whether credit derivatives have had a tangible impact on bank credit supply and, if so, for which borrowers? This is, of course, primarily an empirical question. To begin addressing this question, we review aggregate data on U.S. commercial banks’ use of credit derivatives. Table 1 reports credit derivative usage for all U.S. commercial banks, based on data in bank regulatory reports (the “Call Reports”). The Call

Reports have collected data on banks’ credit derivatives holdings since 1997. These data include the notional principal amounts on for credit derivatives contracts for which the bank is the

“beneficiary” (purchases credit protection) and for which the bank is the “guarantor” (sells credit protection). We use these notional principal amounts as the measure of credit derivatives activity.

As previously reported in Minton, Stulz, and Williamson (2006), very few U.S. banks actually hold credit derivatives. At the end of 2006, just 42 banks reported any credit derivatives holdings. While the number has risen over time – from 20 banks in 1997 to 30 or more in the

2000s – this total still represents just a tiny fraction of all U.S. banks. Usage is clustered among the largest banks, however, with nearly all banks with $100 billion or more in assets using credit derivatives. Some of these large banks are dealers who manage large portfolios of customer- related positions, as well as positions held for their own internal credit risk management purposes.

Figure 1 provides some perspective on how credit derivatives use by U.S. banks has evolved over time. The figure plots the notional principal amounts of credit derivatives held by

U.S. banks as a share of commercial and industrial loans held on the balance sheet, separated into contracts on which banks have bought protection from contracts on which they have sold protection. Dealer banks – defined as those with average notional principal amounts of $10 228 7 billion and above for the sample period – are excluded to give a sense of the extent to which these contracts are being used for risk management purposes. While both credit protection bought and credit protection sold have increased since the late 1990s, the increase in protection bought has been significantly sharper. Even at the end of the sample period, however, aggregate notional amounts of protection bought represented just 5 percent of aggregate C&I loans, a relatively small share.

The specific types of derivatives used by banks might provide some insight into the likely impact of credit derivatives on bank loan supply. The Call Report data indicate that the vast majority of credit derivatives held by U.S. banks are credit default swaps5, but for the questions of interest in this paper, differences in the type of derivatives instrument may be of less significance than differences in the creditors covered by the instrument. In particular, significant use of index-based or multi-name credit derivatives – derivatives on which the payoff is derived from the default behavior of more than one creditor – would be consistent with banks’ doing portfolio-level hedging, suggesting that a wide set of borrowers might benefit from credit derivatives. In contrast, if single-name derivatives dominate, then the impact may be more narrow, limited to just those borrowers named in the derivatives. For instance, Gibson (2007) argues that banks use single-name credit derivatives primarily to hedge large exposure risk from individual corporate borrowers, a view echoed by the chief risk officer of a major U.S. commercial bank (Truslow 2007).

Unfortunately, the Call Reports data do not provide information about single- versus multi-name positions, so we cannot examine this question on a bank-by-bank basis. However, survey data from the Bank for International Settlements (BIS) indicate that between 2004 and

2006, about two-thirds to three-quarters of credit derivatives held by non-dealer banks and

5 In 2006, the first year in which the Call Reports collected information on credit derivatives by type of instrument, 97 percent of all credit derivatives held by U.S. commercial banks were credit default swaps. Among non-dealer banks, this share drops to about 80 percent, with most of the remainder being total return swaps. 229 8 securities firms were single-name instruments. The survey data thus present some prima facie evidence that the impact from banks’ credit derivatives use is likely to be narrow, though of course it is possible that banks could use single-name credit derivatives to hedge the risk of one borrower and use the resulting freed-up risk capacity to lend to someone else. One of the goals of the empirical work that follows is to shed some light on this question.

Previous empirical work on the impact of credit derivatives on credit supply is limited.

However, there are several papers suggesting a positive relationship between bank credit supply and active credit risk management more generally. Credit market innovations such as loan sales

(Cebenoyan and Strahan 2004), collateralized debt obligations (Franke and Krahnen 2005), and collateralized loan obligations (Goderis et al. 2006) all seem to be associated with increases in lending by banks. In a related vein, Brewer et al. (2000) find that banks that use interest rate derivatives have faster growth of C&I loans.

The work that has directly examined the impact of credit derivatives presents a more mixed picture. Minton, Stulz, and Williamson (2006) find that banks with higher shares of C&I lending are more likely to be net purchasers of credit derivatives protection and are also more likely to use other credit risk transfer techniques. However, their focus is more on explaining whether or not banks purchase credit derivatives protection rather than on the impact of credit derivatives on bank loan supply. Ashcraft and Santos (2007) do tackle this question directly by examining the effect of the onset of credit default swap trading on the spreads that the underlying firms pay on their bonds and loans, and find mixed results. Spreads for risky and opaque firms rise at the onset of trading, while spreads on safer and more transparent firms fall somewhat. The beneficial effects increase as credit derivatives trading becomes more liquid.

The main goal of this paper is to provide empirical evidence about the existence and size of the impact of credit derivatives on banks’ credit supply. In particular, we ask how the volume of new C&I loans originated and the terms of those loans change as banks increase their use of credit derivatives to reduce credit exposures. We also examine how the impact of credit 230 9 derivatives varies across borrower types, loan types, and bank types. An important contribution relative to earlier work is that we examine several aspects of credit supply aside from the volume of lending held on the balance sheet, providing a more complete picture of the impact of credit derivatives on the credit process.

III. Data and Empirical Approach

The data used in this paper are a combination of several regulatory and confidential data sets. Information on banks’ use of credit derivatives is derived from the Call Reports, which contain quarterly balance sheet, off-balance sheet, and income statement information for all U.S. commercial banks.6 We link the Call Report data on credit derivatives and other bank-specific characteristics to two confidential data sets about banks’ lending activities. The first of these is loan-level data from the Federal Reserve’s Survey of Terms of Business Lending (STBL).7 The

STBL collects data on new commercial and industrial loans extended to U.S. borrowers by a sample of approximately 250 banks during one-week periods in February, May, August, and

November of each year.8 The survey collects information on loan amount, maturity, risk rating, interest rate, and selected loan characteristics (e.g., whether the loan is secured and whether it is made under a commitment). The data set does not identify or contain information about individual borrowers, however.

The basic empirical approach is to do panel regressions relating banks’ supply of credit to commercial and industrial borrowers to their use of credit derivatives to hedge credit exposures.

Specifically, we estimate the following equation:

:)1( Credit Supply ti = β1, Credit UsagesDerivative − X titi − +Γ+ ε ,1,1, ti ,

6 These data are available at http://www.chicagofed.org/economic_research_and_data/commercial_bank_data.cfm. 7 The data collected in the survey are aggregated in public reports (the Federal Reserve E.2 report, available at http://www.federalreserve.gov/releases/e2/). 8 Note that the STBL collects information about loans extended, not about new credit facilities such as loan commitments. 231 10 where i refers to a bank, t is the month of the STBL survey, and t-1 refers to the quarter-end prior to the STBL survey month. Credit Supply is one of a series of variables (described below) that capture various aspects of the banks’ supply of credit to commercial borrowers, and Credit

Derivatives Usage is a measure of the extent of credit protection obtained through credit derivatives.

X is a vector of bank characteristics, which includes log asset size and log asset size squared, the ratio of C&I loans held on the balance sheet to total assets (to capture the extent to which a given bank is active in C&I lending), the ratio of unused commercial loan commitments to total assets, the total risk-based capital ratio, the ratio of risk-weighted assets to total assets, the deposits-to-assets ratio, the ratio of assets held in the trading account to total assets, the ratio of non-performing C&I loans to total C&I loans, and the ratio of the notional principal amount of other derivatives (that is, all derivatives other than credit derivatives) to total assets. We also include the ratio of non-credit-risk derivatives held for purposes other than trading (“hedging derivatives”) to total assets as a proxy measure of the bank’s general propensity to hedge its exposures.9 The regressions also include bank-specific fixed effects. We adjust for the impact of significant mergers by treating the post-merger bank as a different entity from the pre-merger bank.10

A key issue in this specification is identifying the direction of feedback between a bank’s decision to lend and its decision to use credit derivatives to hedge the risk of that lending. The concern is that a positive correlation between credit derivatives use and lending might reflect shocks in loan demand that simultaneously cause the bank to increase its lending (or alter loan

9 Bank regulatory reports contain information on financial derivatives – such as interest rate, foreign exchange, commodity, and equity derivatives – held for purposes other than trading, but do break out credit derivatives in this way. 10 We define a significant merger as one in which the assets of the acquired bank equal 20 percent or more of the assets of the acquiring bank. We drop any observations where the merger occurred between the STBL survey month and the previous quarter-end. 232 11 terms) and to take on more credit derivatives protection, rather than indicating a positive impact on loan supply from hedging via credit derivatives.

Our strategy is to try to isolate changes in a bank’s propensity to use credit derivatives for risk management from any short-term changes in derivatives use due to shocks to loan demand.

The idea is that increased familiarity with credit derivatives – such as having on-going relationships with derivatives dealers, familiarity with contract form, and legal expertise in house

– would facilitate a bank hedging via these instruments. That is, changes in a bank’s propensity to use credit derivatives on an on-going basis could affect its agility in using credit derivatives and its ability to take advantage of good lending opportunities when they arise.

To begin, the credit derivatives and other Call Report control variables are lagged relative to the new loan data, so that they reflect a bank’s position prior to the period covered by the loan data. Assuming that credit derivatives use is persistent, lagged values would capture a bank’s propensity to use credit derivatives, but should not be affected by shocks to demand over the subsequent period when the lending is done. Note that because the regressions include bank-level fixed effects, the control variables capture changes in a bank’s use of credit derivatives over time, rather than simple cross-sectional variation.

The specification also includes variables intended to reflect changes in loan demand over time. First, the regressions include quarterly dummy variables to control for macroeconomic and industry-wide factors that may vary over time, and which capture economy-wide changes in loan amounts and terms.11 Second, the ratio of unused loan commitments to total assets is included in the specification as a means of controlling for the potential for future loan originations. Increases in unused commitment amounts may cause the bank to anticipate greater future loan draw-downs

11 The results are quite similar if we omit the quarterly time dummy variables and use variables that directly aim to capture economic and financial market conditions. These variables include the three month Treasury bill rate, the spread between the ten-year and three month Treasury rates (the yield curve slope), the spread between the AAA and BBB bond indices (credit spread) and the 4-quarter real GDP growth rate. 233 12 and to increase credit derivatives protection in expectation of those new loans being made. The unused commitment variable controls directly for this source of feedback.

Finally, the regressions include a direct measure of loan demand conditions faced by each bank over each lending period, based on confidential bank-level information from the Federal

Reserve’s Senior Loan Officers Opinion Survey on Bank Lending Practices (SLOOS). Four times a year, the Federal Reserve conducts a survey of senior loan officers of a sample of about

60 large U.S. commercial banks to ask about credit terms and standards and loan demand.

Results of this survey are published in aggregate form.12 Loan officers are asked to rank on a 1- to-5 scale whether demand from commercial borrowers has increased or decreased (lower numbers mean increased demand), providing separate assessment for large and small commercial borrowers.13 We include the 1-to-5 responses of the loan officers on the two questions relating to loan demand.14 We map responses from the SLOOS survey so that the responses cover the period during which the banks were extending the credit captured in the STBL as much as possible.15

Including the loan demand variables, the regression equation is:

:)2( Credit Supply ti = β1, Credit UsagesDerivative ti −1,

+ β 2 arg eLDemand ti + β 3, Demand XSmall titi − +Γ+ ε ,1,, ti ,

The final data is thus constructed by mapping confidential bank-level responses to the

SLOOS to the STBL data on new C&I loan extensions, and in turn to Call Report data on credit derivatives and other bank characteristics. There is significant overlap in the banks in SLOOS

12 These surveys are available at: http://www.federalreserve.gov/boarddocs/SnLoanSurvey/. 13 A response of “1” means that demand is “substantially stronger”, “2” means “moderately stronger”, “3” means “about the same”, “4” means “moderately weaker”, and “5” means “substantially weaker”. See http://www.federalreserve.gov/boarddocs/SnLoanSurvey/. 14 Using alternative forms of the SLOOS information – for instance, converting the 1-to-5 responses into indicator variables for whether the bank faced increased or decreased loan demand – does not alter the results. 15 While the SLOOS was conducted roughly once a quarter during the 1997 to 2006 sample period, the precise months the survey was taken vary, so the timing of the STBL-SLOOS mapping also varies over the sample. In the first half of the sample, the SLOOS and STBL are generally from the same month, while in the second half of the sample, the STBL takes place during the early part of the three-month period covered by the SLOOS. 234 13 and STBL samples, so that we are able to match about two-thirds of the observations in the STBL sample with information from the SLOOS.

Table 2 presents information about the combined STBL-SLOOS sample. As the table illustrates, the sample banks include about half of all credit derivatives users, including most of the very largest. In November 2005, for instance, the combined sample included 3 of the 5 largest credit derivatives users (measured by total notional principal of credit derivatives held), and 11 of the 20 largest. Like the overall population, credit derivatives usage is clustered among the largest banks in the combined sample.

Given this clustering, we focus on large banks in our analysis. In particular, we limit the final data sample to banks with average real assets of $10 billion or more over the quarters they are in the sample. We also exclude foreign-owned banks, since their U.S. credit derivative holdings may reflect lending done outside the United States. Finally, we also exclude a small number of dealer banks because their positions include customer transactions unrelated to their internal credit risk management process.16 The final sample consists of 979 observations for 58 banks from Q2 1997 to Q4 2006. These 979 bank-level observations are based on data for nearly

550,000 individual commercial and industrial loans. Table 3 contains basic statistics of the regression data set.

The key variables of interest are those that measure credit supply and credit derivatives usage. Our main credit supply variable is quarterly bank-aggregate loan principal amounts from the STBL data scaled by total C&I loans held on the balance sheet. This “new loan” variable measures the flow of new loans extended to commercial borrowers. We also examine loans extended to different cohorts of commercial borrowers, where we use loan size to proxy for borrower size. Specifically, we calculate new loan supply to small and to large corporate

16 Dealer banks are defined as those (1) having at least $10 billion in credit derivatives notional principal at some point between 1997 and 2005 and (2) being among the 4 largest credit derivatives users during that period. 235 14 borrowers (measured as the aggregate of all loans with principal amounts less than/greater than

$1 million17).

Most of the previous studies of the relationship between credit risk management and credit supply have focused on the volume of C&I lending held on the balance sheet, so using the

STBL data on credit flows is a complementary extension of that earlier work. As a benchmark and to aid comparison to earlier work, we also examine data from the Call Reports on levels and changes in C&I loans held on the balance sheet.

Beyond looking at credit flows, the information in the STBL also allows us to examine other dimensions of credit supply. These dimensions include the maturity of the loans and the interest rates charged. Banks’ supply of credit could change not only in terms of the volume of lending, but also in terms of the duration of the credit relationship the bank is willing to assume or the price of credit offered to borrowers. The STBL data provide a rich opportunity to examine these aspects of the credit supply relationship.

The STBL data contain information about loan maturity, measured in months. We calculate average maturity of the new loans as the weighted average maturity of the individual loans, using loan principal as the weight. A complicating factor is that a significant portion of loans – about 25 percent of the STBL sample – are open-ended loans with no stated maturity.

Since the maturity of these loans is undefined, we drop them from the bank-average figures. To be sure that the bank-averages are meaningful, we drop observations where more than 90 percent of loans have undefined maturities.

The final credit supply variable is the loan spread, calculated as the effective interest rate on the loan minus the 3-month LIBOR rate. Unlike the other credit supply variables, which are bank-level aggregates, we examine spreads at the individual loan level. These regressions include additional control variables reflecting loan characteristics collected in the STLB,

17 As noted, this definition is based on the size of the loan, not the size of the borrower. However, this approach is consistent with the Call Report definition of “small business loans”, which is all commercial and industrial loans with original principal amounts of less than $1 million. 236 15 including loan size, risk rating, maturity, whether the loan is secured, whether there is a prepayment penalty, and whether the loan has been made under a commitment. As noted above, the STBL data do not include borrower characteristics other than risk rating.

The key control variable is the measure of credit derivatives usage. In particular, we want to measure the extent of credit protection obtained by the bank through the credit derivatives market. Similar to Minton, Stulz, and Williamson (2006), we calculate net credit derivatives protection as the notional amount of credit derivatives on which the bank is the beneficiary minus the notional amount on which the bank is the guarantor, divided by total C&I loans. This is an imperfect proxy for the net protection obtained, as notional amounts capture the scale of credit derivatives coverage, but not the terms of the contracts or the risk of the underlying credits.

Equal amounts of notional principal need not translate into equal amount of credit risk bought or sold. Nevertheless, this measure is the best proxy variable using available information. As a check, we also estimate regressions with separate variables for derivatives on which the bank is a beneficiary (“credit protection bought”) and for which the bank is the guarantor (“credit protection sold”).

As illustrated in Table 3, the amount of net credit protection purchased through the credit derivatives market is relatively small. Net credit protection purchased averages just 1.6 percent of C&I lending; even on a gross basis (that is, not netting out protection sold), credit protection bought represents just 2.7 percent of C&I lending on average for the banks in the sample. In part, these low figures reflect that relatively few banks use any credit derivatives, as noted above. But even among those banks that use credit derivatives, net protection purchased averages just 4.6 percent of C&I loans held on the balance sheet (about 25 percent of credit derivatives users are net sellers of protection).

237 16

IV. Credit Derivatives and Credit Supply

Credit Derivatives and the Volume of C&I Lending

Table 4 presents the results of the estimates relating the flow of new C&I loans to banks’ purchase of credit protection through credit derivatives. The first two columns of the table contain results for all new C&I loans in the aggregate. Two specifications are reported: the first, with net credit protection (that is, the difference between credit protection bought and credit protection sold) as the credit derivatives measure, and the second with credit protection bought and credit protection sold entered separately.

In both specifications, the results suggest that as banks increase the amount of credit protection purchased through the credit derivatives market, the volume of new C&I lending decreases. The coefficient on the net credit protection variable is negative and statistically significant (column 1). When the credit derivative variables are entered separately (column 2), the results suggest that this effect comes through the purchase of credit protection. The coefficient on credit protection bought is negative and significant, while the amount of credit protection sold has no significant effect on the flow of new C&I loans extended.

The remaining columns of the table report results for different cohorts of C&I borrowers, proxied by the size of loan. The middle columns of the table report results for small commercial borrowers (loans less than $1 million), while the last two columns report results for large commercial borrowers (loans greater than $1 million). It is interesting to examine the impact of credit derivatives on lending to these different cohorts of borrowers, since larger borrowers are much more likely to be “named credits” in specific credit derivatives contracts. Thus, the impact of credit derivatives on credit supply for these borrowers may be significantly different than for smaller commercial firms.

The direct effect of increased credit derivatives protection is negative and significant for both large and small C&I borrowers. For small borrowers, both protection bought and protection sold have a significant effect on the flow of new lending: as with the overall results, new lending 238 17 appears to decline (increase) as the amount of credit protection purchased (sold) increases. The flow of new lending to large corporate borrowers also decreases as net credit derivatives protection rises, though in this case the effect is solely through credit protection purchased.

While the direction of the impact is similar across the two cohorts of borrowers, the economic impact of increases in credit derivatives protection is much larger for small corporate borrowers.

A one-standard-deviation increase in net credit derivatives protection implies a 0.3 standard deviation decrease in the flow of credit in small corporate loans, as compared to less than a 0.1 standard deviation decrease in the flow of larger loans.

The results suggest differences in the impact of credit derivatives protection across types of borrowers. It is also informative to ask whether there are differences across types of banks. In particular, some banks may be more active and efficient than other banks in managing risk, and these differences may alter the impact of their credit derivatives use. Such a finding would be consistent with earlier work on the impact of risk management, such as Cebenoyan and Strahan

(2004), who found that banks that were active risk managers had lower risk and higher profits than other banks.

To explore this idea, we use the ratio of hedging derivatives to total assets as a proxy for the extent of active risk management at banks. As noted above, this variable captures the extent to which a bank uses other, non-credit derivatives (derivatives based on interest rates, equity prices, foreign exchange rates, and commodity prices) for hedging, as opposed to trading, purposes. We interact this hedging ratio with the credit derivatives variables to see whether the impact of credit protection varies with a bank’s overall propensity to hedge. The coefficient on the main credit derivatives variable captures the impact for banks that do no other hedging, while the impact for banks that are “hedgers” depends both on the size and signs of the coefficients and on the extent of other hedging activities:

Hedging sDerivative )3( ImpactNet = β + β * . Credit ctionProte Credit HedgingXctionProte Assets 239 18

The results of this interacted specification are reported in Table 5. The coefficients on the main credit derivatives variables continue to suggest that as credit derivatives protection increases, the flow of new lending to both large and small corporate borrowers decreases. As in the initial results, this effect comes mainly from credit protection bought for large borrowers, and from both credit protection bought and credit protection sold for smaller borrowers.

A bank’s propensity to hedge does not appear to affect the relationship between credit derivatives and the flow of new loans to large C&I borrowers (see the last two columns of the table), as the coefficients on the interacted terms are at best only weakly statistically significant.

In contrast, the negative impact of credit derivatives on the flow of new C&I lending to small borrowers is reduced as a bank’s hedging activity increases (see the middle two columns of the table). The coefficient on the cross term between credit derivatives protection and hedging propensity is positive and significant. The estimates suggest that net impact of credit derivatives use on new C&I lending is positive for banks that are very active hedgers.18

Thus far, the results suggest differences in the impact of credit derivatives protection across different types of corporate borrowers and across different types of banks. The STBL data allow us to do one further important decomposition across different types of C&I loans. In particular, the STBL data identifies loans made under commitment and those made without a broader lending agreement (“term loans”). About three-quarters of the loans in the STBL sample are made under commitments. This distinction could be important because the terms of a commitment loan may have been negotiated months, or even years, before the actual extension of funds by the bank.19 The bank has less control over the flow of lending from previously negotiated commitments than it does with newly negotiated term credit arrangements. By

18 The estimates suggest that the net impact of credit derivatives protection is positive for banks with values of the hedging variable greater than 95.6 percent (0.431/0.451), about 6 percent of the sample. 19 The STBL has collected information on commitment date since the end of 2003. Nearly 25 percent of commitment loans made between 2004 and 2006 were made under commitments negotiated a year or more before the loan extension. The median commitment age was about six months. 240 19 separating the sample into commitment and term lending, we may be better able to identify the impact of credit derivatives protection on (new) bank credit supply. Previous research (Drucker and Puri 2007) has found differences in the way term loans and loan commitments are treated in the loan sales market, further suggesting that this distinction could be important.

Tables 6 and 7 report the results of the new lending regressions on the commitment lending and term lending sub-samples. Different patterns appear for the two sub-samples. The results for commitment lending (Table 6) mirror those for the overall sample. Increases in credit derivatives protection are associated with reduced future flows of new C&I loans, and this negative impact is offset for small C&I borrowers as a bank’s overall propensity to hedge increases. Since commitment loans represent about three quarters of all loans in the STBL dataset, this finding is perhaps not surprising.

The relationship between credit derivatives and the flow of new term loans differs somewhat, however (Table 7). For small term borrowers, increased credit derivatives protection is once again associated with reduced future lending, mainly through the impact of credit protection purchased. A bank’s overall propensity to hedge has no significant impact on this result; the coefficients on the interacted terms are essentially zero.

For large term borrowers, increased credit derivatives protection appears to have no significant impact on the flow of new loans. The coefficients on both the net credit protection and credit protection bought variables are not statistically significant, nor is there a significant impact for banks that are active hedgers. Interestingly, the flow of new loans does appear to decrease as credit protection sold increases, suggesting that there may be some substitution between these two different means of taking on credit risk. Put another way, as a bank reduces the amount of credit protection it sells through credit derivatives, the flow of new lending to large term borrowers increases.

Overall, the results thus far provide little support for the idea that increased use of credit derivatives is associated with increases in the supply of bank credit. The direct effect of 241 20 increased credit derivatives protection on the flow of new C&I loans is nearly uniformly negative.

Large term borrowers are an exception to this general result, with either no significant relationship or a positive one through the (inverse) impact of sales of credit protection. The generally prevailing negative effect does appear to be reversed at banks that are active hedgers using other types of derivatives, but this reversal is significant only for small commitment borrowers.

These results are not consistent with previous research that has found a positive relationship between active risk management and bank credit supply (for instance, Cebenoyan and Strahan 2004, Goderis et al. 2006). However, as noted above, the STBL data allow us to examine credit supply through flows of new loans, while past studies of the relationship between credit market innovations and credit supply have focused on loans held on the balance sheet. It may be that differences in both the definition of credit supply (on-balance sheet holdings versus new loans originated) and data sources account for the difference in findings. Thus, for comparison, we also examine the on-balance sheet holdings of C&I loans to see if the negative relationship between credit derivatives protection and credit supply is evident there. These results are reported in Table 8.

These estimates have the same basic structure as those based on the STBL data in that a measure of credit supply is regressed on lagged bank characteristics, including credit derivatives positions. In this case, we use Call Report data on C&I loans in quarter t regressed on bank characteristics in quarter t-1, using the same set of control variables as in the regressions using the

STBL data. We look both at the level of balance sheet C&I lending (scaled by assets) and the quarterly percent change in C&I loans held on the balance sheet.20 We include all banks with

20 Note that the control variables include the lagged level of C&I loans to assets, so the loan levels equations include a lagged dependent variable. Since the specification includes bank-level fixed effects, the loan levels equations are estimated with the dynamic panel data framework developed by Arellano and Bond (1991). This framework – which uses lagged values as instruments – means that some observations are dropped. 242 21 average assets of $10 billion or more, whether or not they are in the STBL sample, so our overall sample expands to approximately 1200 observations.

Consistent with the estimates using the STBL data, these results provide little evidence of a positive impact of credit derivatives usage on on-balance sheet amounts of C&I lending. While the coefficients on net credit protection and credit protection bought are positive (and the coefficients on credit protection sold are negative), none are statistically significant either individually or jointly (see the bottom row of the table). There is some evidence of a more positive effect at banks that are active hedgers, but only in one specification of the loan growth equation. Overall, then, the on-balance sheet findings are more consistent with the findings for the flow of new loans than they are with previous research examining the impact of earlier credit market innovations on bank credit supply, in the sense that they do not suggest a positive relationship between credit derivatives protection and the volume of commercial lending.

Credit Derivatives, Maturity, and Spreads

The results thus far provide at best weak support for the idea that banks that purchase more credit derivatives protection increase their credit supply. However, we have explored this question only by examining the amount of credit extended, whereas “credit supply” arguably encompasses loan terms, such as maturity and loan interest rates, as well as loan volume.

Corporate borrowers may benefit from banks’ use of credit derivatives through lower loan spreads or longer loan maturities, even if the amount of new loans originated does not increase.

As noted above, the key advantage of the STBL data is that they allow us to examine additional aspects of credit supply other than the volume of lending. In particular, the data allow us to examine the relationship between banks’ use of credit derivatives and the maturity and spreads charged on new C&I loans.

Tables 9 and 10 present regression results examining the link between credit derivatives and the average maturity of new C&I loans extended by banks for commitment and term loans, 243 22 respectively.21 One way that banks could increase loan supply is to be willing to enter into longer-maturity loans and thus have a more extended credit relationship with their borrowers.

For commitment loans (Table 9), the primary evidence for this effect is among small borrowers.

The direct impact of increased credit derivatives protection is negative – and once again comes entirely through credit protection purchased. As other hedging activity at a bank increases, however, the impact reverses, such that the net impact of increased credit derivatives protection on average loan maturity is positive for those banks that are very active hedgers. However, the economic impact is fairly small: for a bank at the 95th percentile in hedging activity in the sample, a one-standard-deviation increase in credit derivative protection implies just a 0.1 standard deviation increase in average loan maturity (about one month). For large commitment loans, changes in credit derivatives protection appear to have no impact on average loan maturity.

In contrast to the results for commitment loans, the results for term loans suggest an overall positive relationship between credit derivatives protection and average loan maturity for banks that are active hedgers (Table 10). While the direct impact of increases in credit derivatives protection is negative, the coefficient estimates for the interacted variables are positive and significant for both large and small borrowers. Further, the estimates suggest that net impact of increased credit protection is positive for a relatively large portion of the banks in the sample, especially for large term loans.22 The economic impact is also fairly large: for a bank at the 90th percentile in hedging activity, a one-standard-deviation increase in net credit derivatives protection implies an increase of 5.7 months in the average maturity of new large term loans, against an typical average maturity of 22 months.

The final set of results concerns the relationship between credit derivatives and loan spreads. Here we regress loan-level spreads over the 3-month LIBOR rate against the loan

21 Since the results for the flow of new loans suggest there are significant differences between term and commitment loans, we focus on these separate results – rather than the aggregate sample – in this section. 22 The estimates suggest that the net impact of credit derivatives protection on the average maturity of large term loans is positive for banks whose hedging ratio is in the range of 0.09 to 0.12, approximately the top 30 to 40 percent of the sample. 244 23 characteristics included in the STBL data – loan size, risk rating, maturity, and whether the loan is secured or has a pre-payment penalty – as well as the bank characteristics included in the previous regressions, including the loan demand variables from the SLOOS. The regressions include bank-level fixed effects and the residuals are clustered at the bank level.

Table 11 contains the results for commitment loans. Given the large number of loans in the sample – more than 500,000 – most of the loan-specific control variables are statistically significant and most have the anticipated sign (riskier loans have higher spreads and secured loans have higher spreads). Fewer of the bank-specific control variables are statistically significant – reflecting the impact of bank fixed effects and clustering the residuals – but the results do indicate that banks that rely more heavily on deposits for funding tend to have loans with higher spreads and that banks that are active overall hedgers tend to have lower spreads.

The key variables of interest are the ones that capture credit derivatives exposures (the bottom panel of the table). The results suggest that credit derivatives have no impact on spreads for small commitment loans (the coefficients are uniformly not statistically significant), but that spreads on large commitment loans decrease as credit derivatives protection increases. This effect comes from credit protection bought. The economic importance of this effect is small, however: a one-standard deviation increase in credit derivatives protection implies just a 3 to 7 basis point reduction in the loan spread (less than one-tenth of a standard deviation). The extent of other hedging at the bank appears to have no impact on spreads.

The results for term lending, reported in Table 12, are much stronger. The coefficient estimates suggest that increases in credit derivatives protection are associated with lower loan spreads. For small term borrowers, this impact is offset for banks that engage in greater overall hedging activity, such that for active hedgers (the top 10 percent of the sample), increased credit derivatives protection is associated with higher spreads. For large term borrowers, however, a bank’s overall propensity to hedge does not have an impact and higher credit derivatives protection leads to lower loan spreads. This effect is economically meaningful: a one-standard 245 24 deviation increase in credit protection is associated with a 21 basis point reduction in the loan spread.

Summary and Assessment

What do all these results imply about the impact of credit derivatives on bank credit supply? The answer depends on the type of borrower, the type of loan, and the type of bank. For commitment lending, there is little to suggest that increased use of credit derivatives leads to a significant increase in loan supply. For large commitment borrowers, spreads fall a small amount as credit derivatives protection increases, but the flow of new lending also falls and maturities are unchanged. For small commitment borrowers, spreads are unchanged as credit derivatives protection varies and both the flow of new loans and average loan maturity decrease as credit derivatives protection increases. These results reverse at banks that are very active hedgers, though the economic impact is small even for banks near the top of hedging activity distribution.

Thus, the impact of impact of credit derivatives on loan supply to commitment borrowers is at best mixed (lower spreads but less lending) or weakly positive, while in some cases, the results seem more consistent with an outright decrease in loan supply to commitment borrowers.

In contrast, the evidence is more consistent with an increase in credit supply to large term borrowers. Increases in credit derivatives protection have a weakly positive impact on the volume of new large term loans (when protection sold decreases), and average maturity increases

(especially at active hedgers) and spreads fall. The impact on maturity and spreads is economically meaningful, especially at active hedgers. For smaller term borrowers, the picture is more mixed. For banks that are not active hedgers, increased credit derivatives protection is associated with lower amounts of new term lending to small borrowers and shorter maturities, but also lower spreads. At active hedgers, maturity increases as banks purchase more credit protection, but so do spreads.

For purposes of understanding the net impact of credit derivatives on bank credit supply, the term loan results may be the most relevant. Term loans, by definition, are newly negotiated 246 25 and reflect current credit conditions. Commitment loans, in contrast, have terms that may have been negotiated months or even years previously. Banks have less control over the volume and timing of commitment loan extensions, since these are largely up to the discretion of the borrower, subject to any “material adverse change” clauses in the commitment contract. The flow of new term lending may thus more closely capture a bank’s current ability and desire to create new credit supply.

It is also interesting to note that the strongest evidence of increased credit supply comes for large, rather than small, term borrowers. Large borrowers are much more likely to be “named credits” in credit derivative transactions – that is, to have single-name credit default swaps written on their debt. While the evidence is indirect, the findings concerning large term borrowers suggest that much of the benefit of the growth in the credit derivatives market accrues to these borrowers, rather than being spread more broadly across firms. This is consistent with BIS survey data indicating that non-dealer banks predominantly hold single-name credit derivatives.

Overall, the impact of credit derivatives use by banks appears to be narrow.

One final aspect of the results worth highlighting concerns the interaction of credit derivatives protection and a bank’s overall propensity to hedge. In several of the equations, the impact of credit derivatives protection differed significantly as the extent of hedging via financial derivatives increased. For measures of extent of credit granted (volume of new lending and average maturity), the difference was generally in the direction of increased credit supply, though in terms of price, increased overall hedging was associated with decreased supply (higher spreads). One interpretation of these findings is that hedging via credit derivatives is complementary to other forms of hedging. Banks that are active hedgers in one arena – through their use of financial derivatives – appear to be using credit derivatives to increase credit supply more than banks who are less active hedgers. This finding is consistent with Minton, Stulz and

Williamson (2006), who find that banks using credit derivatives are more likely to use other credit risk mitigation techniques. 247 26

V. Summary and Conclusions

This goal of this paper is to examine the relationship between banks’ use of credit derivatives and the supply of bank credit. Credit derivatives represent an important credit market innovation that, in theory, allows banks to originate and fund loans without holding the associated credit risk. More broadly, credit derivatives are the latest in a series of innovations that have facilitated credit risk management and made it easier for banks to diversify their credit risk exposures.

The key question is whether banks have used these instruments primarily to diversify and thus reduce their risk exposures, or whether banks have undone the diversification by expanding their lending. Research on earlier credit market innovations has found that activities such as loans sales and securitizations have not resulted in overall reductions in bank risk, but rather an expansion of lending. Such an increase is credit supply would be an important consequence of the recent rapid growth of the market for credit derivatives.

We find limited evidence supporting the idea that banks increase the supply of credit as they obtain additional credit protection through credit derivatives, with the strongest results for large term borrowers. Commitment borrowers do not appear to benefit greatly from increased use of credit derivatives by their lenders. These results suggest that the benefits of increased credit derivatives protection are relatively narrow, in the sense that they accrue mainly to the type of borrower most likely to be a “named credit” in a credit derivatives transaction.

It is further interesting to note that the impact of credit derivatives protection for these borrowers is primarily on the terms of lending – longer loan maturity and lower spreads – rather than on loan volume. This is a key advantage of working with the STBL data rather than focusing exclusively on on-balance sheet holdings of commercial loans. Had we focused exclusively on loan amounts – either on-balance sheet holdings or the flow of new originations – we would have missed the most economically meaningful impact of credit derivatives protection, on maturities and loan spreads for large term loans. 248 27

References

Arellano, Manuel and Stephen Bond. 1991. “Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations.” Review of Economic Studies. 58: 277-297.

Ashcraft, Adam B. and João A.C. Santos. 2007. “Has the Credit Default Swap Market Lowered the Costs of Corporate Debt?” Federal Reserve Bank of New York Staff Report no. 290. July 2007.

Bomfim, Antulio N. 2005. Understanding Credit Derivatives and Related Instruments. San Diego: Elsevier Academic Press. 2005.

Brewer III, Elijah, Bernadette A. Minton, and James T. Moser. 2000. “Interest-rate Derivatives and Bank Lending.” Journal of Banking and Finance. 24: 353-79.

Cebenoyan, A. Sinan and Philip E. Strahan. 2004. “Risk Management, Capital Structure and Lending at Banks.” Journal of Banking and Finance. 28: 19-43.

Drucker, Stephen and Maju Puri. 2007. “On Loan Sales, Loan Contracting, and Lending Relationships.” FDIC Center for Financial Research Working Paper no. WP 2007-04. Available at SSRN: http://ssrn.com/abstract=920877.

Duffee, Gregory R. and Chunsheng Zhou. 2001. “Credit Derivatives in Banking: Useful Tools for Managing Risk?” Journal of Monetary Economics. 48: 25-54.

Franke, Gunter and Jan Pieter Krahnen. 2005. “Default Risk Sharing Between Banks and Markets: The Contribution of Collateralized Debt Obligations.” National Bureau of Economic Research Working Paper 11741. November 2005.

Froot, Kenneth A. and Jeremy C. Stein. 1998. “Risk Management, Capital Budgeting, and Capital Structure Policy for Financial Institutions: An Integrated Approach.” Journal of Financial Economics. 47: 55-82.

Geithner, Timothy F. 2006. “Implications of Growth in Credit Derivatives for Financial Stability.” Remarks at the New York University Stern School of Business Third Credit Risk Conference. New York City. May 16, 2006.

Gibson, Michael. 2007. “Credit Derivatives and Risk Management.” Board of Governors of the Federal Reserve System Finance and Economics Discussion Series paper 2007-47.

Goderis, Benedikt, Ian W. Marsh, Judit Vall Castello, and Wolf Wagner. 2006. “Bank Behavior with Access to Credit Risk Transfer Markets.” Available at SSRN: http://ssrn.com/abstract=937287.

Greenspan, Alan. 2005. “Risk Transfer and Financial Stability.” Remarks at the Federal Reserve Bank of Chicago’s Forty-first Annual Conference on Bank Structure. Chicago, IL. May 5, 2005.

Instefjord, Norvald. 2005. “Risk and Hedging: Do Credit Derivatives Increase Bank Risk?” Journal of Banking and Finance. 29: 333-345. 249 28

International Swaps and Derivatives Association. 2007. “ISDA Market Survey.” http://www.isda.org/statistics/pdf/ISDA-market-survey-historical-data.pdf.

Joint Forum. 2005. “Credit Risk Transfer.” Basel: Bank for International Settlements. March 2005.

Lopez, Jose A. 2001. “Financial Instruments of Mitigating Credit Risk.” Federal Reserve Bank of San Francisco Economic Letter. 2001-34.

Minton, Bernadette A., René Stulz, and Rohan Williamson. 2006. “How Much Do Banks Use Credit Derivatives to Reduce Risk?” Fisher College of Business Working Paper 2006- 03-001. June 2006.

Morrison, Alan D. 2005. “Credit Derivatives, Disintermediation, and Investment Decisions.” Journal of Business. 78: 621-647.

Truslow, Don. 2007. “Effectiveness of Credit Risk Management in Allocating Risk.” Remarks at the Federal Reserve Bank of Richmond 2007 Credit Market Symposium. Charlotte, NC, March 22, 2007. http://www.richmondfed.org/news_and_speeches/conferences/pdf/cms_risk%20allocatio n_truslow.pdf

250 29

Figure 1

Credit Derivatives Bought and Sold As Share of C&I Loans Weighted Average for U.S. Commercial Banks (Excl. Dealers) .05

Protection Bought .04 .03 .02

.01 Protection Sold 0

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Source: Call Reports on Condition and Income

251 30

Table 1: Number of U.S. Commercial Banks Using Credit Derivatives by Asset Size Category (Number of Banks in Category) Asset Size 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Less than $1 4 1 1 1 3 4 6 5 5 5 Billion (8764) (8454) (8162) (7935) (7703) (7482) (7348) (7183) (7054) (6952)

$1 to $10 Billion 1 3 3 3 2 2 4 4 4 8 (355) (358) (358) (331) (342) (343) (359) (371) (387) (394)

$10 to $100 8 9 12 14 17 11 10 12 16 20 Billion (72) (73) (78) (81) (74) (80) (79) (78) (74) (74)

More than $100 7 7 7 8 9 8 10 10 8 9 Billion (8) (8) (7) (9) (10) (9) (10) (11) (8) (9)

20 20 23 26 31 25 30 31 33 42 TOTAL (9199) (8896) (8605) (8356) (8129) (7914) (7796) (7643) (7523) (7429) Source: Call Reports for September 30 of each year. Assets in 2006 dollars.

Table 2: Combined STBL and SLOOS Data Set Number of Banks Using Credit Derivatives by Asset Size Category (Number of Banks in Category) Asset Size 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Less than $1 0 0 0 0 0 0 0 0 0 0 Billion (0) (0) (0) (0) (0) (0) (0) (0) (0) (0)

$1 to $10 Billion 0 0 0 0 0 0 0 0 0 0 (7) (3) (4) (6) (6) (5) (6) (3) (3) (4)

$10 to $100 8 7 10 9 10 7 5 7 8 11 Billion (32) (31) (30) (27) (25) (27) (26) (23) (22) (26)

More than $100 5 3 3 5 6 7 7 7 7 8 Billion (6) (4) (3) (6) (7) (8) (7) (7) (7) (8)

13 10 13 14 16 14 12 14 15 19 TOTAL (45) (38) (37) (39) (38) (40) (39) (33) (32) (38) Source: Federal Reserve Survey of the Terms of Business Lending, Federal Reserve Senior Loan Officers Opinion Survey, and Call Reports. Call Report data as of September 30; STBL sample as of November. Assets in 2006 dollars. 252 31

Table 3: Basic Statistics of the Regression Data Set Standard Number of Variable Mean Median Deviation Minimum Maximum Observations

Loan Supply Variables

New Loans/ C&I Loans 0.066 0.037 0.095 0.0001 0.709 979

C&I Loans On Balance Sheet/ 0.168 0.158 0.084 0.026 0.511 1038 Total Assets

Change in C&I Loans/ Total C&I 0.020 0.015 0.085 -0.549 1.093 1218 Loans

Average Maturity 18.643 17.285 11.969 0.236 99.941 974

Spread 3.302 3.167 1.368 -5.656 14.945 537,441

Credit Derivatives Variables Net Credit Derivatives Protection/ C&I Loans 0.016 0.000 0.085 -0.279 0.698 979

Credit Protection Bought/ C&I Loans 0.027 0.000 0.090 0.000 0.811 979

Credit Protection Sold/ C&I Loans 0.011 0.000 0.033 0.000 0.295 979

Bank Characteristics

Asset Size 62.529 38.573 67.007 8.298 415.859 979

C&I Loans/ Assets 0.187 0.179 0.085 0.027 0.529 979

Unused Commitments/ Assets 0.273 0.228 0.182 0.001 1.066 979

Total Risk Based Capital Ratio 0.115 0.112 0.013 0.093 0.220 979

Risk Based Assets/ Assets 0.841 0.826 0.133 0.389 1.421 979

Deposits/ Assets 0.626 0.620 0.119 0.236 0.920 979

Trading Assets/ Assets 0.015 0.003 0.039 0.000 0.236 979

Nonperforming C&I Loans/ C&I 0.011 0.009 0.009 0.000 0.059 979 Loans

Other Derivatives/ Assets 1.238 0.257 2.848 0.000 25.002 979

Hedging Derivatives/ Assets 0.227 0.063 0.614 0.000 5.116 979

Loan Demand

Demand from Large Borrowers 3.069 3 0.721 1 5 979

Demand from Small Borrowers 3.025 3 0.641 1 5 979 Source: Federal Reserve Survey of Terms of Business Lending, Federal Reserve Senior Loan Officers Opinion Survey, and Call Reports.

253 32 Table 4: Credit Derivatives and New C&I Loan Extensions Bank Characteristics All Loans Small Loans Large Loans Log (Asset Size) 0.0047 -0.0315 -0.0152*** -0.0187*** 0.0199 -0.0128 (0.0419) (0.0422) (0.0048) (0.0049) (0.0405) (0.0407)

Log (Asset Size) Squared -0.0026 0.0024 0.0018*** 0.0023*** -0.0044 0.0001 (0.0061) (0.0062) (0.0007) (0.0007) (0.0059) (0.0061)

C&I Loans/ Assets -0.5018*** -0.5267*** -0.0153 -0.0177 -0.4865*** -0.5090*** (0.1812) (0.1817) (0.0143) (0.0144) (0.1697) (0.1702)

Unused Commitments/ Assets 0.0455 0.0258 0.0012 -0.0007 0.0443 0.0265 (0.0466) (0.0469) (0.0038) (0.0040) (0.0461) (0.0464)

Total Risk-Based Capital Ratio 0.2437 0.2589* 0.0227 0.0242 0.2210 0.2348 (0.1530) (0.1543) (0.0151) (0.0153) (0.1465) (0.1476)

Risk-Based Assets/ Total Assets 0.1341** 0.1559** -0.0165*** -0.0144** 0.1506** 0.1703** (0.0675) (0.0696) (0.0060) (0.0061) (0.0643) (0.0663)

Deposits/ Assets -0.2603* -0.2185 0.0013 0.0054 -0.2616* -0.2239 (0.1569) (0.1579) (0.0118) (0.0113) (0.1524) (0.1538)

Trading Assets/ Assets -0.3520 -0.3840 0.0513** 0.0482** -0.4034 -0.4322 (0.2917) (0.2905) (0.0234) (0.0235) (0.2828) (0.2817)

Nonperforming C&I Loans/ C&I Loans -0.0436 -0.0483 0.0071* 0.0067* -0.0507 -0.0549 (0.0480) (0.0478) (0.0039) (0.0040) (0.0471) (0.0469)

Other Derivatives/ Assets -0.0214*** -0.0215*** 0.0003* 0.0003* -0.0217*** -0.0219*** (0.0033) (0.0034) (0.0002) (0.0002) (0.0033) (0.0033)

Hedging Activity 0.0243*** 0.0238*** -0.0003 -0.0004 0.0246*** 0.0242*** (0.0051) (0.0051) (0.0004) (0.0004) (0.0050) (0.0050) Loan Demand Demand from Large Borrowers 0.0043 0.0049 -0.0002 -0.0001 0.0044 0.0050 (0.0041) (0.0041) (0.0003) (0.0003) (0.0040) (0.0040)

Demand from Small Borrowers 0.0022 0.0014 0.0004 0.0003 0.0019 0.0011 (0.0039) (0.0039) (0.0003) (0.0003) (0.0037) (0.0038) Credit Derivatives Net Credit Protection / C&I Loans -0.1177*** -0.0344*** -0.0833** (0.0412) (0.0050) (0.0397)

Credit Protection Bought/ C&I Loans -0.1493*** -0.0374*** -0.1118*** (0.0417) (0.0055) (0.0400)

Credit Protection Sold/ C&I Loans -0.0188 0.0211*** -0.0399 (0.0544) (0.0051) (0.0532)

Number of Observations 979 979 979 979 979 979

R-Squared 0.193 0.197 0.199 0.205 0.199 0.202

P-value: Credit Derivatives Variable Equal 0? 0.004 0.000 0.000 0.000 0.036 0.001 The dependent variable is the sum of principal amounts of new C&I loans extended by each bank in the sample in each quarter, divided by previous quarter- end C&I loans. Small loans are those with principal amounts of $1 million or less; large loans are those with principal amounts exceeding $1 million. Net credit protection is the notional principal of credit derivatives on which the bank is the beneficiary (credit protection bought) minus the notional principal of credit derivatives on which the bank is the guarantor (credit protection sold), divided by C&I loans. Loan Demand data are from the Federal Reserve’s Senior Loan Officers Opinion Survey (SLOOS). Bank characteristics are from the Call Reports, while new loan data is from the Federal Reserve’s Survey of Terms of Business Lending (STBL). The sample consists of all banks in both the STBL and SLOOS panels with average real assets greater than $10 billion. Data are from Q2 1997 to Q4 2006. The regression included bank-specific fixed effects and quarterly dummy variables. The symbols ***, **, and * mean that the coefficient is statistically different from zero at the 1%, 5% and 10% levels, respectively. 254 33 Table 5: Credit Derivatives and New C&I Loan Extensions with Hedging Interaction Bank Characteristics All Loans Small Loans Large Loans Log (Asset Size) 0.0045 -0.0041 -0.0155*** -0.0157*** 0.0200 0.0116 (0.0420) (0.0421) (0.0048) (0.0053) (0.0405) (0.0405)

Log (Asset Size) Squared -0.0025 -0.0015 0.0019*** 0.0019** -0.0044 -0.0034 (0.0061) (0.0062) (0.0007) (0.0008) (0.0059) (0.0060)

C&I Loans/ Assets -0.5009*** -0.5189*** -0.0137 -0.0154 -0.4872*** -0.5035*** (0.1816) (0.1833) (0.0143) (0.0145) (0.1701) (0.1716)

Unused Commitments/ Assets 0.0454 0.0248 0.0010 -0.0011 0.0443 0.0259 (0.0467) (0.0469) (0.0038) (0.0039) (0.0461) (0.0464)

Total Risk-Based Capital Ratio 0.2429 0.2465 0.0211 0.0212 0.2218 0.2252 (0.1532) (0.1543) (0.0151) (0.0153) (0.1467) (0.1477)

Risk-Based Assets/ Total Assets 0.1332* 0.1476** -0.0181*** -0.0168*** 0.1514** 0.1643** (0.0679) (0.0703) (0.0060) (0.0062) (0.0647) (0.0670)

Deposits/ Assets -0.2645* -0.2263 -0.0067 -0.0027 -0.2578* -0.2235 (0.1604) (0.1604) (0.0125) (0.0117) (0.1561) (0.1567)

Trading Assets/ Assets -0.3525 -0.3442 0.0505** 0.0524** -0.4030 -0.3966 (0.2921) (0.2925) (0.0233) (0.0232) (0.2831) (0.2838)

Nonperforming C&I Loans/ C&I Loans -0.0431 -0.0350 0.0080** 0.0091** -0.0511 -0.0442 (0.0478) (0.0481) (0.0039) (0.0041) (0.0470) (0.0473)

Other Derivatives/ Assets -0.0214*** -0.0220*** 0.0004** 0.0003* -0.0218*** -0.0223*** (0.0033) (0.0034) (0.0002) (0.0002) (0.0033) (0.0034)

Hedging Activity 0.0243*** 0.0179*** -0.0004 -0.0012*** 0.0247*** 0.0191*** (0.0051) (0.0048) (0.0004) (0.0004) (0.0050) (0.0047) Loan Demand Demand from Large Borrowers 0.0043 0.0048 -0.0002 -0.0001 0.0045 0.0049 (0.0041) (0.0041) (0.0003) (0.0003) (0.0040) (0.0040)

Demand from Small Borrowers 0.0022 0.0012 0.0004 0.0003 0.0019 0.0010 (0.0039) (0.0039) (0.0003) (0.0003) (0.0037) (0.0038) Credit Derivatives Net Credit Protection / C&I Loans -0.1223** -0.0431*** -0.0792* (0.0475) (0.0061) (0.0455)

Net Credit Protection X Hedging 0.0239 0.0451*** -0.0211 (0.0839) (0.0126) (0.0815)

Credit Protection Bought/ C&I Loans -0.1772*** -0.0493*** -0.1279*** (0.0463) (0.0056) (0.0452)

Credit Protection Bought X Hedging 0.1744** 0.0640*** 0.1104 (0.0841) (0.0134) (0.0831)

Credit Protection Sold/ C&I Loans -0.0456 0.0248*** -0.0703 (0.0594) (0.0065) (0.0583)

Credit Protection Sold X Hedging 0.1456 -0.0228* 0.1684* (0.0943) (0.0123) (0.0916)

Number of Observations 979 979 979 979 979 979

R-Squared 0.193 0.201 0.210 0.225 0.199 0.205

P-value: Credit Derivatives Variable Equal 0? 0.016 0.000 0.000 0.000 0.088 0.000 The dependent variable is the sum of principal amounts of new C&I loans extended by each bank in the sample in each quarter, divided by previous quarter- end C&I loans. Small loans are those with principal amounts of $1 million or less; large loans are those with principal amounts exceeding $1 million. Net credit protection is the notional principal of credit derivatives on which the bank is the beneficiary (credit protection bought) minus the notional principal of credit derivatives on which the bank is the guarantor (credit protection sold), divided by C&I loans. Loan Demand data are from the Federal Reserve’s Senior Loan Officers Opinion Survey (SLOOS). Bank characteristics are from the Call Reports, while new loan data is from the Federal Reserve’s Survey of Terms of Business Lending (STBL). The sample consists of all banks in both the STBL and SLOOS panels with average real assets greater than $10 billion. Data are from Q2 1997 to Q4 2006. The regression included bank-specific fixed effects and quarterly dummy variables. The symbols ***, **, and * mean that the coefficient is statistically different from zero at the 1%, 5% and 10% levels, respectively. 255 34 Table 6: Credit Derivatives and New C&I Loan Extensions: Commitment Loans Bank Characteristics All Loans Small Loans Large Loans Log (Asset Size) -0.0643* -0.0529 -0.0191*** -0.0178*** -0.0452 -0.0351 (0.0331) (0.0327) (0.0043) (0.0048) (0.0330) (0.0327)

Log (Asset Size) Squared 0.0068 0.0051 0.0024*** 0.0022*** 0.0043 0.0029 (0.0049) (0.0049) (0.0006) (0.0007) (0.0049) (0.0049)

C&I Loans/ Assets 0.0239 0.0206 0.0438*** 0.0432*** -0.0199 -0.0225 (0.0467) (0.0472) (0.0088) (0.0089) (0.0467) (0.0470)

Unused Commitments/ Assets 0.0337 0.0257 0.0037 0.0024 0.0300 0.0233 (0.0352) (0.0356) (0.0042) (0.0043) (0.0357) (0.0362)

Total Risk-Based Capital Ratio -0.1087 -0.1131 -0.0182 -0.0187 -0.0905 -0.0944 (0.1098) (0.1100) (0.0136) (0.0137) (0.1081) (0.1083)

Risk-Based Assets/ Total Assets -0.0442 -0.0426 -0.0330*** -0.0326*** -0.0113 -0.0100 (0.0391) (0.0403) (0.0048) (0.0050) (0.0388) (0.0400)

Deposits/ Assets -0.2239* -0.2099* -0.0030 -0.0007 -0.2209* -0.2092* (0.1166) (0.1202) (0.0105) (0.0102) (0.1164) (0.1197)

Trading Assets/ Assets -0.2707 -0.2463 0.0185 0.0217 -0.2891 -0.2679 (0.1937) (0.1943) (0.0205) (0.0204) (0.1936) (0.1944)

Nonperforming C&I Loans/ C&I Loans -0.0952** -0.0857** 0.0056* 0.0070** -0.1008** -0.0926** (0.0411) (0.0414) (0.0033) (0.0035) (0.0412) (0.0415)

Other Derivatives/ Assets -0.0160*** -0.0165*** 0.0007*** 0.0006*** -0.0167*** -0.0171*** (0.0031) (0.0031) (0.0002) (0.0002) (0.0031) (0.0031)

Hedging Activity 0.0091** 0.0037 -0.0014*** -0.0021*** 0.0105*** 0.0058 (0.0038) (0.0037) (0.0003) (0.0004) (0.0038) (0.0037) Loan Demand Demand from Large Borrowers 0.0003 0.0004 -0.0006** -0.0006** 0.0009 0.0010 (0.0024) (0.0024) (0.0003) (0.0003) (0.0023) (0.0024)

Demand from Small Borrowers 0.0021 0.0017 0.0004* 0.0004 0.0017 0.0013 (0.0022) (0.0023) (0.0002) (0.0003) (0.0022) (0.0023) Credit Derivatives Net Credit Protection / C&I Loans -0.1160*** -0.0372*** -0.0788** (0.0389) (0.0055) (0.0379)

Net Credit Protection X Hedging 0.0509 0.0456*** 0.0053 (0.0634) (0.0126) (0.0603)

Credit Protection Bought/ C&I Loans -0.1464*** -0.0418*** -0.1046*** (0.0386) (0.0053) (0.0384)

Credit Protection Bought X Hedging 0.1749*** 0.0633*** 0.1116* (0.0589) (0.0131) (0.0584)

Credit Protection Sold/ C&I Loans 0.0360 0.0249*** 0.0110 (0.0508) (0.0062) (0.0497)

Credit Protection Sold X Hedging 0.1109 -0.0228* 0.1336** (0.0680) (0.0130) (0.0646)

Number of Observations 979 979 979 979 979 979

R-Squared 0.231 0.238 0.312 0.324 0.222 0.228

P-value: Credit Derivatives Variable Equal 0? 0.009 0.000 0.000 0.000 0.064 0.000 The dependent variable is the sum of principal amounts of new C&I loans made under commitment extended by each bank in the sample in each quarter, divided by previous quarter-end C&I loans. Small loans are those with principal amounts of $1 million or less; large loans are those with principal amounts exceeding $1 million. Net credit protection is the notional principal of credit derivatives on which the bank is the beneficiary (credit protection bought) minus the notional principal of credit derivatives on which the bank is the guarantor (credit protection sold), divided by C&I loans. Loan Demand data are from the Federal Reserve’s Senior Loan Officers Opinion Survey (SLOOS). Bank characteristics are from the Call Reports, while new loan data is from the Federal Reserve’s Survey of Terms of Business Lending (STBL). The sample consists of all banks in both the STBL and SLOOS panels with average real assets greater than $10 billion. Data are from Q2 1997 to Q4 2006. The regression included bank-specific fixed effects and quarterly dummy variables. The symbols ***, **, and * mean that the coefficient is statistically different from zero at the 1%, 5% and 10% levels, respectively.

256 35 Table 7: Credit Derivatives and New C&I Loan Extensions: Term Loans Bank Characteristics All Loans Small Loans Large Loans Log (Asset Size) 0.0688** 0.0488 0.0037 0.0021 0.0652** 0.0467 (0.0316) (0.0331) (0.0026) (0.0028) (0.0296) (0.0308)

Log (Asset Size) Squared -0.0093** -0.0066 -0.0006 -0.0003 -0.0087** -0.0062 (0.0047) (0.0050) (0.0004) (0.0004) (0.0044) (0.0047)

C&I Loans/ Assets -0.5249*** -0.5396*** -0.0575*** -0.0586*** -0.4674*** -0.4810*** (0.1824) (0.1839) (0.0156) (0.0157) (0.1672) (0.1686)

Unused Commitments/ Assets 0.0117 -0.0009 -0.0026 -0.0035 0.0143 0.0026 (0.0295) (0.0293) (0.0030) (0.0030) (0.0268) (0.0266)

Total Risk-Based Capital Ratio 0.3516*** 0.3596*** 0.0393*** 0.0399*** 0.3123*** 0.3197*** (0.1090) (0.1100) (0.0110) (0.0111) (0.0997) (0.1006)

Risk-Based Assets/ Total Assets 0.1775*** 0.1902*** 0.0149*** 0.0158*** 0.1626*** 0.1744*** (0.0582) (0.0600) (0.0051) (0.0052) (0.0534) (0.0550)

Deposits/ Assets -0.0406 -0.0164 -0.0037 -0.0020 -0.0369 -0.0143 (0.1200) (0.1162) (0.0074) (0.0073) (0.1141) (0.1106)

Trading Assets/ Assets -0.0818 -0.0980 0.0321* 0.0307 -0.1139 -0.1287 (0.2380) (0.2397) (0.0194) (0.0196) (0.2205) (0.2220)

Nonperforming C&I Loans/ C&I Loans 0.0520* 0.0506* 0.0023 0.0022 0.0497** 0.0485* (0.0267) (0.0269) (0.0022) (0.0022) (0.0248) (0.0250)

Other Derivatives/ Assets -0.0054*** -0.0055*** -0.0003*** -0.0003*** -0.0051*** -0.0052*** (0.0017) (0.0017) (0.0001) (0.0001) (0.0016) (0.0016)

Hedging Activity 0.0152*** 0.0142*** 0.0010*** 0.0010*** 0.0142*** 0.0133*** (0.0037) (0.0036) (0.0003) (0.0003) (0.0034) (0.0033) Loan Demand Demand from Large Borrowers 0.0040 0.0044 0.0004 0.0005 0.0036 0.0039 (0.0036) (0.0036) (0.0003) (0.0003) (0.0033) (0.0033)

Demand from Small Borrowers 0.0001 -0.0004 -0.0001 -0.0001 0.0002 -0.0003 (0.0034) (0.0034) (0.0003) (0.0003) (0.0031) (0.0031) Credit Derivatives Net Credit Protection / C&I Loans -0.0063 -0.0058*** -0.0005 (0.0230) (0.0017) (0.0218)

Net Credit Protection X Hedging -0.0269 -0.0005 -0.0265 (0.0579) (0.0053) (0.0537)

Credit Protection Bought/ C&I Loans -0.0308 -0.0075*** -0.0233 (0.0226) (0.0017) (0.0214)

Credit Protection Bought X Hedging -0.0006 0.0007 -0.0012 (0.0609) (0.0056) (0.0564)

Credit Protection Sold/ C&I Loans -0.0815** -0.0002 -0.0814** (0.0376) (0.0024) (0.0359)

Credit Protection Sold X Hedging 0.0348 -0.0000 0.0348 (0.0728) (0.0065) (0.0675)

Number of Observations 979 979 979 979 979 979

R-Squared 0.132 0.135 0.156 0.158 0.131 0.133

P-value: Credit Derivatives Variable Equal 0? 0.787 0.013 0.000 0.000 0.859 0.018 The dependent variable is the sum of principal amounts of new C&I loans not made under commitment extended by each bank in the sample in each quarter, divided by previous quarter-end C&I loans. Small loans are those with principal amounts of $1 million or less; large loans are those with principal amounts exceeding $1 million. Net credit protection is the notional principal of credit derivatives on which the bank is the beneficiary (credit protection bought) minus the notional principal of credit derivatives on which the bank is the guarantor (credit protection sold), divided by C&I loans. Loan Demand data are from the Federal Reserve’s Senior Loan Officers Opinion Survey (SLOOS). Bank characteristics are from the Call Reports, while new loan data is from the Federal Reserve’s Survey of Terms of Business Lending (STBL). The sample consists of all banks in both the STBL and SLOOS panels with average real assets greater than $10 billion. Data are from Q2 1997 to Q4 2006. The regression included bank-specific fixed effects and quarterly dummy variables. The symbols ***, **, and * mean that the coefficient is statistically different from zero at the 1%, 5% and 10% levels, respectively. 257 36 Table 8: Credit Derivatives and C&I Loans Held on the Balance Sheet Bank Characteristics Loan Growth Loan Levels Log (Asset Size) -0.0475 -0.0297 0.1083*** 0.0865** (0.0795) (0.0915) (0.0358) (0.0361)

Log (Asset Size) Squared 0.0034 0.0009 -0.0086** -0.0058 (0.0127) (0.0144) (0.0040) (0.0039)

C&I Loans/ Assets -0.4146*** -0.4220*** 0.8026*** 0.8392*** (0.1515) (0.1514) (0.0399) (0.0367)

Unused Commitments/ Assets 0.0294 0.0270 0.0219 0.0189 (0.0433) (0.0438) (0.0137) (0.0120)

Total Risk-Based Capital Ratio -0.5036* -0.5186* 0.0177 -0.0133 (0.2952) (0.2963) (0.0904) (0.0898)

Risk-Based Assets/ Total Assets 0.0256 0.0232 -0.0037 -0.0116 (0.0780) (0.0801) (0.0269) (0.0237)

Deposits/ Assets 0.1931 0.1935 0.0053 0.0203 (0.2178) (0.2173) (0.0490) (0.0396)

Trading Assets/ Assets -0.2154 -0.1847 -0.1259 -0.1822 (0.5348) (0.5368) (0.1370) (0.1251)

Nonperforming C&I Loans/ C&I Loans 0.0283 0.0413 -0.0335* -0.0332* (0.0610) (0.0614) (0.0176) (0.0170)

Other Derivatives/ Assets -0.0023 -0.0028 0.0006 0.0003 (0.0032) (0.0033) (0.0018) (0.0017)

Hedging Activity 0.0009 -0.0051 -0.0051* -0.0033 (0.0082) (0.0073) (0.0029) (0.0021) Loan Demand Demand from Large Borrowers -0.0127** -0.0127** 0.0005 -0.0001 (0.0053) (0.0053) (0.0009) (0.0008)

Demand from Small Borrowers 0.0048 0.0043 -0.0006 -0.0002 (0.0049) (0.0049) (0.0009) (0.0008) Credit Derivatives Net Credit Protection / C&I Loans 0.0533 0.0256 (0.0880) (0.0180)

Net Credit Protection X Hedging 0.1961 -0.0140 (0.1764) (0.0342)

Credit Protection Bought/ C&I Loans 0.0251 0.0231 (0.0956) (0.0192)

Credit Protection Bought X Hedging 0.3273* -0.0120 (0.1818) (0.0362)

Credit Protection Sold/ C&I Loans -0.1212 -0.0257 (0.1023) (0.0295)

Credit Protection Sold X Hedging -0.0123 -0.0025 (0.2258) (0.0427)

Number of Observations 1218 1218 1038 1038

R-Squared 0.134 0.136

P-value: Credit Derivatives Variables Equal 0? 0.261 0.102 0.249 0.619 The dependent variables are the percent change in C&I loans held on the balance sheet (loan growth) and the level of C&I loans held on the balance sheet divided by assets (loan levels). Net credit protection is the notional principal of credit derivatives on which the bank is the beneficiary (credit protection bought) minus the notional principal of credit derivatives on which the bank is the guarantor (credit protection sold), divided by C&I loans. Bank information comes from the Call Reports, and information on loan demand comes from the Senior Loan Officers Opinion Survey (SLOOS). The sample consists of all banks with average real assets greater than $10 billion. Data are from Q2 1997 to Q4 2006. The regression included bank-specific fixed effects and quarterly dummy variables. The symbols ***, **, and * mean that the coefficient is statistically different from zero at the 1%, 5% and 10% levels, respectively. 258 37 Table 9: Credit Derivatives and Average Loan Maturity: Commitment Loans Bank Characteristics All Loans Small Loans Large Loans Log (Asset Size) 7.0798 3.0234 8.2191 3.8804 -2.2387 -10.3980 (12.2651) (13.0385) (9.3734) (9.9330) (17.8799) (19.7679)

Log (Asset Size) Squared -0.5744 -0.0074 -1.5069 -0.9022 1.0745 2.2296 (1.6269) (1.7353) (1.2354) (1.3141) (2.3157) (2.5799)

C&I Loans/ Assets -4.7023 -7.1102 22.2756* 18.5344 -11.6928 -15.6883 (16.9621) (16.9752) (11.6010) (11.6157) (22.5235) (22.4904)

Unused Commitments/ Assets -3.7485 -5.3665 -4.8031 -7.8862* 3.0647 2.2169 (5.7565) (6.0470) (4.3850) (4.3939) (8.0385) (8.3387)

Total Risk-Based Capital Ratio 92.7921 94.8835* 121.2012** 123.8245** -2.9790 4.1471 (57.2712) (57.4236) (53.8471) (54.0253) (46.2619) (46.5014)

Risk-Based Assets/ Total Assets 10.0748 12.2409 -1.2869 1.9488 4.8564 7.8329 (8.8330) (9.1156) (6.6512) (6.6959) (11.8730) (12.0986)

Deposits/ Assets -7.4932 -8.3515 -21.1082*** -21.2855*** 5.6549 3.2049 (6.5561) (6.6833) (5.2250) (5.3405) (9.4747) (9.6791)

Trading Assets/ Assets -19.5145 -15.6146 9.1286 16.2553 -27.4420 -26.2344 (33.1236) (32.9834) (22.3900) (22.6468) (38.8347) (39.0903)

Nonperforming C&I Loans/ C&I Loans 18.6804 14.7219 -66.6779* -69.4962* -17.7860 -32.8531 (51.0794) (51.2728) (38.2638) (38.3127) (87.0575) (88.1953)

Other Derivatives/ Assets 0.4952 0.5028 0.4546* 0.4116 0.7427 0.8379 (0.4644) (0.4682) (0.2581) (0.2640) (0.5923) (0.6075)

Hedging Activity -0.3488 -0.1576 -1.0914** -1.4178*** 0.1449 1.1435 (0.7975) (0.8371) (0.4766) (0.4996) (1.2477) (1.2818) Loan Demand Demand from Large Borrowers -0.0448 0.0060 -0.2628 -0.1760 0.1485 0.2093 (0.5501) (0.5520) (0.4378) (0.4384) (0.7545) (0.7573)

Demand from Small Borrowers 0.5026 0.4364 0.6094 0.4663 0.1021 0.0601 (0.5774) (0.5827) (0.4477) (0.4489) (0.7012) (0.7134) Credit Derivatives Net Credit Protection / C&I Loans -6.0061 -14.1503** 10.4524 (7.4474) (5.5286) (10.2187)

Net Credit Protection X Hedging 13.5223 21.0866 5.5268 (18.7854) (14.6840) (43.6146)

Credit Protection Bought/ C&I Loans -8.2126 -20.5032*** 11.2033 (7.8554) (5.5842) (11.1411)

Credit Protection Bought X Hedging 10.5524 30.6694** -13.0553 (17.5300) (15.1751) (42.0725)

Credit Protection Sold/ C&I Loans -3.8064 -8.0784 -11.9090 (8.9168) (7.0568) (11.4115)

Credit Protection Sold X Hedging -23.7039 -18.1326 -45.5191 (25.3219) (17.7897) (49.3189)

Number of Observations 964 964 961 961 892 892

R-Squared 0.117 0.119 0.159 0.167 0.075 0.080

P-value: Credit Derivatives Variable Equal 0? 0.665 0.510 0.038 0.000 0.448 0.167 The dependent variable is the weighted average maturity (in months) of new C&I loans made under commitment extended by a bank in each quarter. Loan principal amounts are used as weights. Observations where more than 90% of new loans have no stated maturity are dropped. Net credit protection is the notional principal of credit derivatives on which the bank is the beneficiary (credit protection bought) minus the notional principal of credit derivatives on which the bank is the guarantor (credit protection sold), divided by C&I loans. Loan information comes from the Survey of Terms of Business Lending (STBL), bank information comes from the Call Reports, and information on loan demand comes from the Senior Loan Officers Opinion Survey (SLOOS). The sample consists of all banks in the STBL panel with average real assets greater than $10 billion. Data are from Q2 1997 to Q4 2006. The regression included bank-specific fixed effects and quarterly dummy variables. The symbols ***, **, and * mean that the coefficient is statistically different from zero at the 1%, 5% and 10% levels, respectively. 259 38 Table 10: Credit Derivatives and Average Loan Maturity: Term Loans Bank Characteristics All Loans Small Loans Large Loans Log (Asset Size) -57.1047 -48.1507 -63.2009** -45.7444 -27.7168 -22.6619 (35.7623) (40.4665) (31.6002) (35.3828) (68.4241) (75.1018)

Log (Asset Size) Squared 7.5117 6.3444 8.5217** 6.2502 4.2847 3.6455 (4.6751) (5.2783) (4.1886) (4.6738) (9.1980) (10.0055)

C&I Loans/ Assets -72.2650** -69.0835* 15.8270 24.3664 -57.7401 -53.6524 (34.9624) (36.4101) (34.2385) (35.2430) (49.4724) (51.7954)

Unused Commitments/ Assets 66.2058*** 68.2879*** 66.5321*** 71.0202*** 49.2678** 50.3909** (18.7823) (18.7276) (20.1171) (19.8526) (24.0538) (24.9256)

Total Risk-Based Capital Ratio -186.1293 -181.4319 -79.1437 -68.0293 -354.2852 -348.7122 (135.9765) (138.3766) (107.1052) (108.7892) (238.4258) (244.9968)

Risk-Based Assets/ Total Assets -31.7163 -34.7889 -38.0791 -45.0857* -31.2165 -33.4008 (26.8271) (26.5822) (23.8501) (23.0318) (36.4790) (37.1090)

Deposits/ Assets 83.7664*** 85.4813*** 78.8472*** 81.6333*** 102.7455** 102.1701** (23.2372) (23.3663) (21.3609) (21.3247) (42.2026) (42.4595)

Trading Assets/ Assets -25.2037 -29.1761 43.8797 34.3291 -10.0732 -13.6519 (52.2403) (53.4929) (45.7044) (48.9473) (70.9751) (71.2307)

Nonperforming C&I Loans/ C&I Loans -28.6948 -23.9259 55.5207 63.5473 -373.4457 -374.2493 (158.5378) (159.6562) (131.9114) (131.1738) (267.5737) (269.5267)

Other Derivatives/ Assets 0.3938 0.3119 0.2927 0.1472 0.0820 0.0765 (1.0812) (1.0860) (0.8922) (0.8925) (2.0627) (2.0681)

Hedging Activity 0.0552 -0.2634 1.6406 1.2141 -1.6726 -1.7113 (1.9312) (2.1268) (1.6927) (1.8868) (4.1808) (5.8244) Loan Demand Demand from Large Borrowers -2.7645* -2.8472* -2.3355* -2.5240** -3.8831* -3.9178* (1.4561) (1.4557) (1.1936) (1.1742) (2.3524) (2.3703)

Demand from Small Borrowers 1.0623 1.1853 1.1546 1.4594 -0.2966 -0.2296 (1.7330) (1.6886) (1.5301) (1.3997) (2.2438) (2.2869) Credit Derivatives Net Credit Protection / C&I Loans -44.4591* -44.8888** -22.4773 (24.0026) (18.6274) (35.8420)

Net Credit Protection X Hedging 141.3672*** 87.9614** 188.9500** (49.0727) (37.3066) (93.1472)

Credit Protection Bought/ C&I Loans -39.7948 -31.2850 -17.4912 (31.3128) (25.7593) (49.2901)

Credit Protection Bought X Hedging 146.6871** 91.3038** 187.3799* (57.1877) (44.5652) (112.2652)

Credit Protection Sold/ C&I Loans 58.3403 81.1810** 32.7119 (35.8463) (35.2183) (36.7730)

Credit Protection Sold X Hedging -122.2727** -54.1454 -185.3415* (51.8411) (45.8358) (110.6563)

Number of Observations 766 766 762 762 502 502

R-Squared 0.156 0.157 0.175 0.181 0.168 0.169

P-value: Credit Derivatives Variable Equal 0? 0.015 0.062 0.033 0.049 0.122 0.346 The dependent variable is the weighted average maturity (in months) of new C&I loans not made under commitment extended by a bank in each quarter. Loan principal amounts are used as weights. Observations where more than 90% of new loans have no stated maturity are dropped. Net credit protection is the notional principal of credit derivatives on which the bank is the beneficiary (credit protection bought) minus the notional principal of credit derivatives on which the bank is the guarantor (credit protection sold), divided by C&I loans. Loan information comes from the Survey of Terms of Business Lending (STBL), bank information comes from the Call Reports, and information on loan demand comes from the Senior Loan Officers Opinion Survey (SLOOS). The sample consists of all banks in the STBL panel with average real assets greater than $10 billion. Data are from Q2 1997 to Q4 2006. The regression included bank-specific fixed effects and quarterly dummy variables. The symbols ***, **, and * mean that the coefficient is statistically different from zero at the 1%, 5% and 10% levels, respectively. 260 39

Table 11: Credit Derivatives and Loan Spreads: Commitment Loans All Loans Small Loans Large Loans Bank Characteristics Log (Asset Size) 1.2273 1.3033 1.1920 1.2617 -0.3003 -0.6526 (0.9553) (0.9716) (0.8787) (0.8968) (1.2037) (1.2599)

Log (Asset Size) Squared -0.1199 -0.1306 -0.1192 -0.1288 0.0284 0.0787 (0.0954) (0.0974) (0.0876) (0.0904) (0.1324) (0.1422)

C&I Loans/ Assets -0.8378 -0.7855 -0.6955 -0.6542 1.0080 0.7124 (1.1890) (1.1496) (1.1772) (1.1509) (1.0979) (1.1326)

Unused Commitments/ Assets -0.9651** -0.9746** -0.9988** -1.0128** 0.0083 -0.0544 (0.4756) (0.4707) (0.4894) (0.4834) (0.3791) (0.3625)

Total Risk-Based Capital Ratio -1.7904 -1.5972 -1.2991 -1.0636 -3.7350 -3.6862 (2.2219) (2.2818) (2.2364) (2.3224) (2.2833) (2.3450)

Risk-Based Assets/ Total Assets 0.9590* 0.9677* 1.1775** 1.1951** -1.6429*** -1.5682** (0.5682) (0.5707) (0.5534) (0.5607) (0.6202) (0.6265)

Deposits/ Assets 1.7969*** 1.7688*** 1.7605*** 1.7327*** 1.5769** 1.6826*** (0.5730) (0.5671) (0.5850) (0.5837) (0.6011) (0.5927)

Trading Assets/ Assets 0.4641 0.4255 1.1760 1.1013 -2.8028** -2.8659** (1.4688) (1.5040) (1.7306) (1.8018) (1.1988) (1.1561)

Nonperforming C&I Loans/ C&I Loans -2.2692 -2.0925 -1.5034 -1.3265 -0.3197 -0.5367 (3.3744) (3.3803) (3.7224) (3.7194) (4.3569) (4.1150)

Other Derivatives/ Assets 0.0105 0.0100 0.0159 0.0155 0.0605*** 0.0621*** (0.0162) (0.0163) (0.0248) (0.0250) (0.0160) (0.0148)

Hedging Activity -0.0813** -0.1022 -0.0877** -0.1138* -0.0271 -0.0348 (0.0329) (0.0681) (0.0376) (0.0664) (0.0670) (0.1114) Loan Characteristics Log (Loan Size) -0.0555 -0.0555 -0.1310 -0.1310 -0.9801*** -0.9769*** (0.0705) (0.0705) (0.1799) (0.1798) (0.2789) (0.2781)

Log (Loan Size) Squared -0.0082** -0.0082** -0.0043 -0.0043 0.0257*** 0.0256*** (0.0035) (0.0035) (0.0093) (0.0092) (0.0089) (0.0089)

Unrated 0.9849*** 0.9878*** 0.9731*** 0.9767*** 0.8078*** 0.8042*** (0.2173) (0.2167) (0.2341) (0.2332) (0.1112) (0.1108)

Risk Rating 2 0.1844 0.1858 0.2201 0.2220 0.1705*** 0.1685*** (0.2101) (0.2100) (0.2327) (0.2327) (0.0612) (0.0615)

Risk Rating 3 0.7712*** 0.7732*** 0.7556*** 0.7582*** 0.6942*** 0.6913*** (0.2430) (0.2431) (0.2757) (0.2760) (0.0647) (0.0649)

Risk Rating 4 0.8543*** 0.8564*** 0.8164*** 0.8191*** 1.2077*** 1.2043*** (0.1513) (0.1502) (0.1746) (0.1738) (0.0806) (0.0811)

Risk Rating 5 1.4480*** 1.4500*** 1.4004*** 1.4030*** 1.7573*** 1.7531*** (0.1412) (0.1396) (0.1634) (0.1620) (0.1538) (0.1538)

Secured 0.3379*** 0.3376*** 0.2735*** 0.2733*** 0.6656*** 0.6677*** (0.0621) (0.0622) (0.0651) (0.0653) (0.0673) (0.0674)

Pre-payment Penalty -0.0490 -0.0492 0.0610 0.0609 -0.7767*** -0.7787*** (0.1518) (0.1520) (0.1671) (0.1671) (0.1294) (0.1290)

Maturity 0.0090*** 0.0090*** 0.0076*** 0.0076*** 0.0138*** 0.0138*** (0.0015) (0.0015) (0.0017) (0.0017) (0.0015) (0.0015)

Maturity Squared -0.0000** -0.0000** -0.0000 -0.0000 -0.0000*** -0.0000*** (0.0000) (0.0000) (0.0000) (0.0000) (0.0000) (0.0000) 261 40

Table 11 (Continued): Credit Derivatives and Loan Spreads: Commitment Loans All Loans Small Loans Large Loans Loan Demand Demand from Large Borrowers 0.0267 0.0256 0.0173 0.0163 0.0475** 0.0492** (0.0175) (0.0168) (0.0177) (0.0170) (0.0186) (0.0191)

Demand from Small Borrowers -0.0163 -0.0167 -0.0030 -0.0039 -0.0474* -0.0515** (0.0281) (0.0283) (0.0277) (0.0279) (0.0253) (0.0243) Credit Derivatives Net Credit Protection / C&I Loans -0.2245 -0.2560 -0.3609** (0.2364) (0.2464) (0.1603)

Net Credit Protection X Hedging 1.4077 1.2300 0.5740 (0.9495) (0.9421) (1.7998)

Credit Protection Bought/ C&I Loans -0.2060 -0.2757 -0.8165** (0.3128) (0.3087) (0.3853)

Credit Protection Bought X Hedging 1.4946 1.4161 2.0551 (1.0183) (0.9713) (2.4926)

Credit Protection Sold/ C&I Loans 0.2847 0.2867 0.0392 (0.2867) (0.3026) (0.1771)

Credit Protection Sold X Hedging -1.0381 -0.8521 -1.9592 (1.4262) (1.4262) (2.1084)

Number of Observations 502,772 502,772 466,691 466,691 36,081 36,081

R-Squared 0.243 0.243 0.180 0.180 0.361 0.362

P-value: Credit Derivatives Variables Equal 0? 0.338 0.470 0.417 0.539 0.084 0.242 The dependent variable is the effective interest rate on the loan minus the 3-month LIBOR rate. Observations where the calculated spread exceeds 15 percent have been dropped. Net credit protection is the ratio of the notional principal on derivatives contracts where the bank purchases credit protection (“beneficiary”) minus the notional amount of contracts on which the bank sells credit protection (“guarantor”), divided by C&I loans. The regressions include bank fixed effects and quarterly dummy variables and are estimated with residuals clustered at the bank level. Data are from Q2 1997 to Q4 2006. Loan information comes from the Survey of Terms of Business Lending (STBL), bank information comes from the Call Reports, and information on loan demand comes from the Senior Loan Officers Opinion Survey (SLOOS). The symbols ***, **, and * indicate statistical significance at the 1%, 5% and 10% levels, respectively. 262 41 Table 12: Credit Derivatives and Loan Spreads: Term Loans All Loans Small Loans Large Loans Bank Characteristics Log (Asset Size) 0.9997 1.6519 1.3974 2.1347* 1.6304 1.1579 (1.3174) (1.1934) (1.3608) (1.2384) (1.3709) (1.3154)

Log (Asset Size) Squared -0.1759 -0.2529* -0.2435 -0.3307** -0.2644 -0.2137 (0.1654) (0.1425) (0.1743) (0.1525) (0.1725) (0.1702)

C&I Loans/ Assets -3.4915** -2.2559 -3.1511 -1.5423 0.8922 0.1699 (1.7472) (1.7815) (2.1899) (2.4007) (1.1421) (1.0212)

Unused Commitments/ Assets 0.7707 0.6652 0.8780 0.8167 0.4014 0.0732 (1.1423) (1.0636) (1.2192) (1.1174) (0.6852) (0.7161)

Total Risk-Based Capital Ratio -8.2599 -7.2317 -9.1679* -7.8296* 6.6831 5.3892 (5.6134) (4.7795) (5.4385) (4.5583) (4.5832) (4.4637)

Risk-Based Assets/ Total Assets 0.5481 -0.1777 0.7087 -0.1066 -1.7668* -1.5450* (0.9616) (0.9645) (0.9509) (0.9720) (0.8954) (0.8073)

Deposits/ Assets 0.9971 0.9300 0.8575 0.7789 -1.1169* -1.1315* (1.5090) (1.3409) (1.5430) (1.3544) (0.5903) (0.5974)

Trading Assets/ Assets -0.2711 -0.7277 -2.6580 -3.7276 -1.2035 -0.7437 (1.9003) (2.1873) (3.2473) (3.6681) (1.4630) (1.3459)

Nonperforming C&I Loans/ C&I Loans 2.5423 3.7704 2.6008 3.3159 2.3047 5.5994 (7.2089) (7.0790) (8.5342) (8.4319) (7.0792) (7.9568)

Other Derivatives/ Assets -0.0126 -0.0220 -0.0801 -0.0866 -0.0058 -0.0093 (0.0619) (0.0631) (0.0867) (0.0928) (0.0393) (0.0395)

Hedging Activity 0.4988*** 0.2949** 0.5877*** 0.3816*** 0.2962* 0.2561 (0.1524) (0.1273) (0.1593) (0.1440) (0.1532) (0.1616) Loan Characteristics Log (Loan Size) -0.3425** -0.3390** -0.3180 -0.3155 -0.9342** -0.8700** (0.1318) (0.1314) (0.4798) (0.4784) (0.3899) (0.3943)

Log (Loan Size) Squared -0.0030 -0.0032 -0.0034 -0.0035 0.0257** 0.0236* (0.0050) (0.0050) (0.0231) (0.0230) (0.0121) (0.0123)

Unrated 0.5084 0.5233 0.4923 0.5135 0.5996*** 0.5906*** (0.3457) (0.3423) (0.4701) (0.4661) (0.1591) (0.1576)

Risk Rating 2 -0.0905 -0.0768 -0.1627 -0.1492 0.2115** 0.2101** (0.3624) (0.3582) (0.4949) (0.4899) (0.0915) (0.0900)

Risk Rating 3 0.3301 0.3406 0.2826 0.2939 0.4128*** 0.4124*** (0.3453) (0.3414) (0.4653) (0.4606) (0.1257) (0.1245)

Risk Rating 4 0.3233 0.3393 0.1662 0.1831 1.0097*** 1.0145*** (0.3520) (0.3478) (0.4712) (0.4660) (0.1630) (0.1641)

Risk Rating 5 0.7979** 0.8236** 0.6200 0.6476 1.8029*** 1.8120*** (0.3789) (0.3738) (0.4834) (0.4776) (0.2543) (0.2508)

Secured -0.0521 -0.0468 -0.1345 -0.1285 0.6764*** 0.6688*** (0.1339) (0.1331) (0.1347) (0.1342) (0.0804) (0.0801)

Pre-payment Penalty -0.1806** -0.1931** -0.1715** -0.1869** -0.1769 -0.2076 (0.0843) (0.0851) (0.0703) (0.0719) (0.1770) (0.1844)

Maturity 0.0172*** 0.0171*** 0.0156*** 0.0155*** 0.0136*** 0.0137*** (0.0041) (0.0041) (0.0041) (0.0041) (0.0023) (0.0023)

Maturity Squared -0.0001*** -0.0001*** -0.0001*** -0.0001*** -0.0000*** -0.0000*** (0.0000) (0.0000) (0.0000) (0.0000) (0.0000) (0.0000) 263 42

Table 12 (Continued): Credit Derivatives and Loan Spreads: Term Loans All Loans Small Loans Large Loans Loan Demand Demand from Large Borrowers 0.0146 0.0228 0.0021 0.0106 0.0148 0.0225 (0.0445) (0.0442) (0.0493) (0.0482) (0.0331) (0.0331)

Demand from Small Borrowers -0.0040 -0.0121 0.0100 0.0005 -0.0056 -0.0110 (0.0411) (0.0425) (0.0408) (0.0410) (0.0239) (0.0242) Credit Derivatives Net Credit Protection / C&I Loans -1.0027* -0.8661 -0.5025 (0.5387) (0.5779) (0.8024)

Net Credit Protection X Hedging 1.1019 0.7332 1.7986 (1.1418) (1.2881) (4.8274)

Credit Protection Bought/ C&I Loans -1.9362*** -1.7830** -2.3001*** (0.6690) (0.7892) (0.8114)

Credit Protection Bought X Hedging 5.2476** 4.9346** 4.0424 (2.2778) (2.4553) (6.4326)

Credit Protection Sold/ C&I Loans 0.8632 0.8620 -0.6301 (0.8422) (0.8867) (0.5714)

Credit Protection Sold X Hedging 0.8397 1.1751 -2.2092 (0.9361) (0.9998) (3.2466)

Number of Observations 34,669 34,669 31,022 31,022 3,647 3,647

R-Squared 0.292 0.294 0.186 0.188 0.390 0.393

P-value: Credit Derivatives Variables Equal 0? 0.162 0.010 0.265 0.009 0.814 0.004

The dependent variable is the effective interest rate on the loan minus the 3-month LIBOR rate. Observations where the calculated spread exceeds 15 percent have been dropped. Net credit protection is the ratio of the notional principal on derivatives contracts where the bank purchases credit protection (“beneficiary”) minus the notional amount of contracts on which the bank sells credit protection (“guarantor”), divided by C&I loans. The regressions include bank fixed effects and quarterly dummy variables and are estimated with residuals clustered at the bank level. Data are from Q2 1997 to Q4 2006. Loan information comes from the Survey of Terms of Business Lending (STBL), bank information comes from the Call Reports, and information on loan demand comes from the Senior Loan Officers Opinion Survey (SLOOS). The symbols ***, **, and * indicate statistical significance at the 1%, 5% and 10% levels, respectively.

264

THE FAILURE OF PRIVATE EQUITY

* STEVEN M. DAVIDOFF

DRAFT Some citations may be absent or incomplete Please do not cite without author’s permission

Word Count: 26,647

Abstract

Throughout the Fall 2007 and into the new year 2008 private equity firms repeatedly attempted to terminate pending acquisitions. The litigation surrounding these purported terminations and heightened scrutiny directed upon the terms of private equity agreements opened a revealing window on a number of supposed “flaws” in the private equity structure. This Article seeks to understand whether these failures existed, and if so, what caused them. It does so by examining the forces driving the construct and evolution of private equity and the rationale for private equity’s structure and specific contractual terms. I find that the private equity structure to be a rich, textured environment. The terms of the contractual relationship between the private equity firm and the acquired company are analogous to an iceberg; they form only the publicly available view of a much deeper understanding between the parties. In the non-public sphere, parties to private equity contracts utilize norms, conventions, reputational constraints, language and relational bonding to fill contractual gaps, override explicit contractual terms, and achieve a negotiated solution beyond the four corners of the contract. The attorney as transaction cost engineer in the private equity context consequently structures the private equity contract by paying heed both to contractual terms and law, contractually created forces and non- legal factors. But attorney reliance on these extra-contractual factors and forces makes the private equity structure path dependent and resistant to change. In light of these findings, the failures of the pre-Fall 2007 private equity structure were particularly a failure by attorneys for acquired companies to innovate and negotiate terms in full contemplation of such events. Reliance upon extra-legal forces permitted these attorneys to negotiate facially flawed private equity contracts and otherwise justified

* Associate Professor of Law, University of Connecticut School of Law; Visiting Professor of Law, The Ohio State University Michael E. Moritz College of Law; B.A., University of Pennsylvania; J.D., Columbia University School of Law; M.S., Finance, London Business School. The author would like to thank Rachel Arnow- Richman, Derek Bambauer, Alan Fishbein, Noah Hall, Claire Hill, David Skeel, Faith Stevelman and the faculty work-shop and panel attendees at [Arizona, Harvard, Illinois, Ohio, Widener]. This paper was prepared as part of the author’s appointment as corporate scholar-in-residence at Widener University School of Law.

Electronic copy available at: http://ssrn.com/abstract=1148178 265

2 82 S. Cal. L. Rev. __ [8-Oct-08 sloppy and ambiguous drafting.

INTRODUCTION ...... 2 I. THE STRUCTURE OF PRIVATE EQUITY...... 8 A. The Origins of the Private Equity Structure...... 8 B. The Shifting Structure of Private Equity...... 13 II. THE FAILURE OF PRIVATE EQUITY ...... 18 A. The Failure of Contract...... 18 B. The Failure of Norms...... 21 C. The Failure of Specific Performance ...... 24 D. The Failure of Financing...... 29 III. NEGOTIATING THE STRUCTURE OF PRIVATE EQUITY...... 31 A. Optionality and the Private Equity Structure ...... 33 B. Other Facets of the Private Equity Structure...... 40 IV. THE PATH DEPENDENCY OF THE PRIVATE EQUITY STRUCTURE ...... 43 A. The Path Dependency of the Private Equity Structure...... 44 B. The Stickiness of the Private Equity Structure...... 46 C. Law Firm Centrality and the Private Equity Structure...... 52 D. The Optimality of the Private Equity Structure ...... 55 V. THE FUTURE OF PRIVATE EQUITY...... 55 CONCLUSION ...... 59 APPENDIX A ...... 61

INTRODUCTION

The Fall of 2007 was a tumultuous time in the U.S. capital markets. A conflux of factors, including the implosion of the subprime mortgage market, created a financial maelstrom disrupting the economy and causing the credit markets to “dry-up” and become increasingly illiquid.1 Almost overnight credit became both more expensive and difficult to obtain as financial institutions such as investment and depositary banks became increasingly unwilling to extend new credit.2 The credit securitization market was particularly affected leaving these financial institutions with pending financing obligations and existing loans they could sell, if at all, only at a large loss.3

1 See Roben Farzad, et al., Not So Smart, In an era of easy money, the pros forgot that the party can't last forever, BUSINESSWEEK, Sep. 3, 2007; Phil Izzo, Economists in Poll Expect Credit Turmoil to Continue, THE WALL ST. J., Nov. 15, 2007, at A. 2 See Antony Currie & Dwight Cass, Ominous Crunching, THE WALL ST. J., Nov. 2, 2007, at C; Eric Dash, Debt Markets Looks to Feds to Ease Fear, THE N.Y. TIMES, Sep. 18, 2007, C. 3 Leveraged loans securitized by CLOs together with high yield debt were the primary tools utilized by private equity firms to finance the debt component of their acquisitions. See infra Part I.A. In the credit markets at the time, large loans were typically securitized and sold to third parties as collateralized loan obligations (CLOs). This allowed financial institutions to transfer these loans off their balance sheets and make new loans with the proceeds from the sale. If the CLO market was unavailable it hampered the ability of these financial institutions to extend new loans. In addition, the closing of the CLO market left these financial institutions with loans extended and agreed to be extended that could either not be sold or sold only at a loss. See Floyd Norris, Sickly Credit Markets Heal a Little as Leveraged Loans Rebound, THE N.Y. TIMES, Oct. 6, 2007, C3.

Electronic copy available at: http://ssrn.com/abstract=1148178 266

8-Oct-08] Failure of Private Equity 3

Facing large losses, these institutions began to balk at funding pre-agreed private equity acquisitions.4 This sudden, unexpected turn of events and the general revaluation and decline in stock prices it wrought led private equity firms to reassess their pending acquisitions agreed to in more stable times.5 The private equity firms’ reevaluations were often unkind. Throughout the Fall and into the new year 2008 private equity firms in a number of previously agreed acquisitions repeatedly attempted to terminate their contractual obligations to acquire companies.6

In such instances, the private equity firms largely successfully relied upon the negotiated language in their contracts to terminate their pending acquisitions or agree a settlement to the same effect.7 In many instances litigation ensued before such disposition. The litigation surrounding these terminations and heightened scrutiny directed upon the terms of private equity agreements opened a revealing window on the practices of parties and attorneys in negotiating and structuring private equity transactions. Under the public glare, the intricate structures of these transactions appeared to be fundamentally flawed or otherwise the product of “suboptimal” structuring and contracting.8 Observers particularly criticized acquirees for agreeing to optional takeover structures; the inclusion of reverse termination fee provisions which permitted private equity firms to terminate the acquisition for any reason simply by paying a flat fee of approximately 3% of the transaction value.9 A blame-game unfolded and fault for these failures was alternatively pinned on financial institutions, investment bankers, private equity firms, acquiree boards and acquiree attorneys.10

This Article is an examination of the structure of these private equity transactions and the role of lawyers, particularly those representing acquirees, in its negotiation and documentation. It seeks to understand the forces driving the evolution of the private equity structure, the rationale for

4 The term private equity as used in this Article refers to the acquisition of both public and private companies by investment entities utilizing a leveraged financing structure. The term is sometimes used on a more general basis to refer to an investment in any private company including venture capital investments. See generally Gail Marmorstein et al., Hidden Treasure: A Look Into Private Equity's History, Future, And Lure, J. WEALTH MGMT, at 2 (Sum 1999). 5 See Ken MacFadyen, Avoiding the Messy Breakup: Amid the turmoil produced by the credit crunch, some PE investors are dealing with a case of acquirers' remorse, INV. DEALERS' DIG., Sept. 10, 2007. 6 I discuss this wave of private equity transaction terminations infra at Part II. 7 See infra at Part II. 8 I discuss these notions further infra at Part IV.D. 9 See, e.g., Steven M. Davidoff, Where Do Breakup Fees Go From Here?, N.Y. TIMES DEALBOOK, Apr 7, 2008, available at http://dealbook.blogs.nytimes.com/2008/04/07/where-do-breakup-fees-go-from-here/. 10 [This author participated on one such panel entitled “Who is To Blame?” before the Merger, Acquisitions & Split-offs class taught by Professor Robert Clark and Vice Chancellor Strine at Harvard Law School.] 267

4 82 S. Cal. L. Rev. __ [8-Oct-08 the structure and its contractual terms and whether the pre-August 2007 private equity structure was indeed a “suboptimal” or “flawed” one. I look to answer these questions through an examination of litigation transcripts, interviews with industry participants and research on a database of 192 private equity contracts and other public records.

I find that the private equity bargain, including the contract implementing the private equity structure, is a rich, textured environment. The terms of the contractual relationships between the private equity firm and the acquired company are partly analogous to an iceberg; they form only the publicly available view of a much deeper understanding between the parties. In the non-public sphere, parties to private equity contracts and their lawyers utilize norms, conventions and outside constraints to fill contractual gaps, override explicit contractual terms, and achieve a negotiated solution beyond the four corners of the contract. In this world, negotiating attorneys have their own discourse on contract language, particularly with respect to open terms, which oftentimes differ from judicial interpretation of the terms they utilize.11 The contract is merely the starting point for analyzing the negotiated understanding of the parties. In the private equity structure though, the contract still has a very real and valid purpose. It outlines the relationship of the parties, and documents their agreement to the extent feasible creating a bonded relationship to affect future conduct. This is a world where parties and their attorneys do not necessarily leave terms ambiguous for future litigation or negotiating purposes.12 Rather, the contract functions as a fulcrum for further private bonding, interpretation, negotiation and agreement after execution of the acquisition agreement.13

In the private equity structure, though, contract terms do play an important, flexible role throughout the relationship. Flipping the iceberg on its head – the contract is not the endpoint but rather the foundation and start for an understanding among the parties. Contract matters from the beginning in the private equity structure. The picture painted has much to offer in terms of further understanding and judicial interpretation of complex contracts; a topic I will explore in a companion paper to this one.14

11 See Lawrence M. Solan, Contract as Agreement, 83 NOTRE DAME LAW REVIEW 353 (2007). 12 See B. Douglas Bernheim & Michael D. Whinston, Incomplete Contracts and Strategic Ambiguity, 88 AM. ECON REV. 902 (1998); Claire Hill, Bargaining in the Shadow of the Lawsuit (2008) (unpublished manuscript on file with author). I discuss this further infra at notes 209-210 and accompanying text. 13 See also Sergio G. Lazzarini et al., Order with Some Law: Complementarity Versus Substitution of Formal and Informal Arrangements, 20 J. L. ECON. ORG. 261, 261 (2004) (“[B]y enforcing contractible exchange dimensions, contracts facilitate the self-enforcement of noncontractable dimensions.”) 14 This companion paper will examine the implications of my findings for current contract theory. Ultimately, the failure and structure of private equity lend support to relational contract theory, but also show that the contract terms and the negotiation thereof have important, key roles in contract. This has follow-on 268

8-Oct-08] Failure of Private Equity 5

This is a world that Professor Lisa Bernstein and Professor Stewart Macaulay, who have extensively written on social norms and their role in establishing extra-legal business relationships, would not find surprising.15 But the findings of this article’s study differ from their research and other research on relational and private contracting which at best has viewed the contract as relevant merely in the “end game” after the relationship has failed and the parties are in dispute.16

The attorney as transaction cost engineer in the private equity context consequently structures the private equity contract by paying heed both to contractual terms and law, contractually created forces and non-legal factors.17 The attorney measures the weight of each of these and adjusts contracts terms and transaction structure in response. However, these extra contractual forces can incentivize lawyers to avoid innovating. The structure of private equity is thus path dependent.18 It results from “good enough” structuring decisions made by transactional attorneys who are under- incentivized to innovate.19 Change in the private equity structure is intermittent, caused by externalized forces and results in piece-meal

implications for the neoformalists and those who advocate for less “law” in judicial interpretation of contracts. See inter alia Eric A. Posner, Law, Economics, and Inefficient Norms, 144 U. PA. L. REV. 1697 (1996); Eric A. Posner, A Theory of Contract Law Under Conditions of Radical Judicial Error, 94 NW. U. L. REV. 749 (2000); Robert E. Scott, The Case for Formalism in Relational Contract, 94 NW. U. L. REV. 847 (2000). The neoformalists have argued that the judiciary is constitutionally incapable in a complex business environment of providing useful default rules, determining the true meaning of parties or otherwise imposing appropriate judgment ex post facto. Accordingly, courts should interpret contracts literally. As support for this the neoformalists have claimed that parties can simply adjust their conduct after any such judgment. See Alan Schwartz, Incomplete Contracts, in THE NEW PALGRAVE DICTIONARY OF ECONOMICS AND THE LAW (Peter Newman ed. 1998); Alan Schwartz, Relational Contracts in the Courts: An Analysis of Incomplete Agreements and Judicial Strategies, 21 J. LEGAL STUD. 271 (1992). This case study of private equity lends support to the neoformalist view. I discuss this further at infra note 291. 15 See inter alia Lisa Bernstein, Opting Out of the Legal System: Extralegal Contractual Relations in the Diamond Industry, 21 J. LEGAL STUD. 115 (1992); Lisa Bernstein, Private Commercial Law in the Cotton Industry: Creating Cooperation Through Rules, Norms, and Institutions, 99 MICH. L. REV. 1724 (2001); Stewart Macaulay, Non-Contractual Relations in Business: A Preliminary Study, 28 AM. SOCIO. REV. 1 (1963). They would not be alone – the two have inspired a raft of further scholarship on this topic. 16 See Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code’s Search for Immanent Business Norms, 144 U. PA L. REV. 1765, 1800-01 (1996) (“the terms of a written contract are viewed as relevant primarily when transactors have decided not to deal again, that is, when their relationship is at an end-game”); Claire A. Hill & Christopher King, How Do German Contracts Do as Much with Fewer Words?, 79 CHI.-KENT L. REV. 889 (2004). Neoformalist contract doctrine has similarly focused on the contract as primarily relevant in the end-game of the contractual relationship. See Robert E. Scott & Paul B. Stephan, Self-Enforcing International Agreements and the Limits of Coercion, 2004 WIS. L. REV. 551, 597. 17 The concept of the transactional attorney as a transaction cost engineer, negotiating transaction structure and terms to create value, net of legal fees, was first prominently put forth by Professor Ronald J. Gilson. See Ronald J. Gilson, Value Creation by Business Lawyers: Legal Skills and Asset Pricing, 94 YALE L.J. 239, 243 (1984). 18 I discuss the meaning of path dependency further infra at notes 223-225 and accompanying text. 19 This proposition that forces push lawyers towards “good enough” decisions is one first brought to my attention in the scholarship of Professor Claire A. Hill. See Claire A. Hill, Why Contracts are Written in "Legalese," 77 CHI.-KENT L. REV. 59, 71 (2001). See also John C. Coates, IV, Explaining Variation in Takeover Defenses: Blame the Lawyers, 89 CAL. L. REV. 1301, at 1308-24 (2001) (examining the forces which affect attorney decisions at the IPO stage which provide “lawyers with sufficient autonomy that they determine their clients' pre-IPO defenses, largely unconstrained by market forces or ethical rules”.). 269

6 82 S. Cal. L. Rev. __ [8-Oct-08 revisions premised upon the pre-existing structure.

Moreover, the private equity legal universe is a small, self-contained one. From January 1, 2004 through August 1, 2007, the same 22 law firms represented a acquiree or private equity acquirer in 91.06% of all public private equity deals.20 Given that these law firms repeatedly represent private equity firms but only represent acquirees on a one-off basis, these firms may not be fully incentivized to negotiate innovative, beneficial provisions for acquirees. While this may not have been the determining factor in the failure of the private equity structure, it likely contributed to the forces working against innovation and reinforcing path dependency.21 Reliance on extra-contractual legal forces additionally allowed attorneys to hide this possible bias.

Notions of transactional “optimality” in the private equity structure are thus relative: the structure of private equity cannot be characterized as an efficient one due to these constraints which create path dependency and hamper innovation. These are not Coasian bargains.22 Illustratively, in the wake of the events of Fall 2007, attorneys reweighed the balance of contract terms and law versus non-legal forces and the structure of private equity and private equity contracts terms rapidly shifted to accommodate the host of litigation and failed transactions.23 The path dependency of private equity, though, still hampered creative solutions that would permit private equity to actively compete in the transformed marketplace.

This Article speaks to the failure of private equity contract and its structure. But there were multiples failures of private equity in this past year. The private equity boom of 2004 through July 2007 occurred against the backdrop of a vast credit bubble. During that time credit was freely available at extraordinarily low interest rates. Financial institutions rushed head-long to finance private equity acquisitions at rates that in hindsight were severely mispriced.24 In this environment private equity firms themselves took advantage of this easy and available credit to pay attractive

20 See Chart IV.C. infra. 21 I discuss the underpinnings of my belief further infra at Part IV.C. 22 Of course, Coase himself acknowledges that transaction costs can inhibit or forestall Coasian bargaining. See Ronald H. Coase, The Problem of Social Cost, 3 J. OF LAW AND ECON. 1 (Oct. 1960). 23 See infra at Part V.A. 24 In other words, the interest rates charged by these financial institutions did not reflect the possible risks associated with a future default by the acquired company on the credit extended. Moreover, the banks extended too much financing resulting in over-leverage of these acquisitions. See, e.g., Great Phrases in History: Irrational Exuberance. ‘Incontinent’ Investing?, WSJ DEAL JOURNAL, http://blogs.wsj.com/deals/2008/10/07/great-phrases-in-history-irrational-exuberance-incontinent-investing/, Oct. 7, 2008. 270

8-Oct-08] Failure of Private Equity 7 prices and make a record number of acquisitions.25

In hindsight, the companies who agreed to be acquired by private equity firms and who were so acquired made the right decision. The shareholders of these acquired companies received a price far above the current trading level of the stock market. Meanwhile, private equity firms acquired a vast portfolio of companies at low interest rates, flexible credit terms and with minimum money invested. The bulk of the funds used for these acquisitions was again provided by banks who were willing to accept smaller equity investments from private equity firms.26 Against this macro-picture, the failure of private equity was a failure of financial institutions to properly price their loans and financial instruments.27 This particular failure is part of a broader credit one that has had astounding economic implications for our world economy.

This Article touches on these issues and provides an explanatory history of the private equity market and its mechanisms, but this Article’s focus is on a more particular failure: the actual private equity transactions which failed to complete. As such, it is a story of the failure of lawyers, particularly lawyers for acquirees, and private equity contracts. And it is a wider case study of how lawyers negotiate contracts, contract functions in a particular industry, path dependency can exist even in sophisticated environments and extra-legal norms and conventions can fail. Here, this Article aims to fill an increasingly noted gap in the contracts literature.28 There has been much theoretical discourse but little research into how clients and their attorneys negotiate and agree to complex contracts in either continuing or discrete relationships.29 Finally, this study has importance for other studies on contract evolution and boiler-plate which have tended to label terms as efficient or suboptimal based on the pure text of contracts without looking to extra-legal understandings and practices of the parties.30 If indeed the contract is only part of the agreement of the parties, these studies may be incomplete.

25 See Justin Baer & Edward Evans, Wall Street bankers wonder how big leveraged buyouts can get, Bloomberg, Apr 3, 2007. 26 For example, the amount of debt provided to Kohlberg Kravis Roberts to purchase First Data exceeded its market capitalization in the week prior to the acquisition. 27 See Speech of Jane Wheeler, Penn M&A Institute, Nov. 25, 2008 (copy on file with author). 28 See generally Brayden King and Gordon D. Smith, Contracts as Organizations (Mar. 2007). University of Wisconsin Legal Studies Research Paper No. 1037 available at SSRN: http://ssrn.com/abstract=969816. See also Nicholas S. Argyres et al., Complementarity and Evolution of Contractual Provisions, 18 Org. Sci. 3, 3 (2007); Stewart Macaulay, Contracts, New Legal Realism, and Improving the Navigation of the Yellow Submarine, 80 TUL L. REV. 1161, 1186-82 (2006). 29 See generally King and Smith, supra note 28 (surveying the limited number of contract studies). 30 See infra notes 218-216 and accompany text. 271

8 82 S. Cal. L. Rev. __ [8-Oct-08

Part I of this Article examines the historical evolution of the private equity contract and structure. Part II details the very public Fall 2007 failure of these contracts. Part III examines more particularly the structure of private equity and its alleged individual failures as well as the negotiation process for these transactions. Part IV details the path dependency of the private equity structure and the role of attorneys, particularly those for acquirees, in perpetuating the structure. Part V concludes with a discussion of the response of market participants to the failures of private equity and their implication for the future structure of these transactions.

I. THE STRUCTURE OF PRIVATE EQUITY

The structure of private equity is largely a product of its unique financing. Private equity pervasively utilizes substantial leverage to purchase public and private corporations. Private equity firms typically borrow 60%-80% of the required purchase price and obtain the remaining necessary capital from pre-committed investors who provide equity for this purpose.31 In order to enhance their returns and increase the number of purchased companies, private equity seeks to place as much debt and as little equity as feasible into the acquisition capital structure.32 Private equity is therefore significantly dependent upon the nature of debt financing and the availability of credit. If credit is more freely available and at lower interest rates, this permits private equity firms to borrow more money to make increased acquisitions at higher prices paid to acquired companies. Because of this, the structure of private equity over the years has evolved, driven in large measure by the type and availability of financing.

A. The Origins of the Private Equity Structure

The foundations of today’s private equity structure were laid in the 1970s and 1980s.33 A beginning occurred in 1976 when Jerome Kohlberg and first cousins Henry Kravis and George Roberts created the first true

31 These pre-committed equity investments are often supplemented by additional side-by-side equity investments by co-investors. See JOSHUA LERNER, ET AL., VENTURE CAPITAL AND PRIVATE EQUITY: A CASEBOOK, at [●] (4th Ed. 2008). 32 Placing more debt on a company enhances the possible returns to the equity investors because any subsequent profit from selling the acquired company is returned on a smaller equity investment. Private equity consistently places significantly more debt on its acquisitions than ordinary public companies. The reasons for this differential are uncertain but can possibly be attributed to the uncertainty of cash flows and the heightened risk increased leverage places on a company, a risk that public companies are not as willing to bear. See Ronald J. Gilson & Charles K. Whitehead, Deconstructing Equity: Public Ownership, Agency Costs, and Complete Capital Markets, 108 COLUM. L. REV. 231 (2008). 33 See GEORGE P. BAKER & GEORGE DAVID SMITH, THE NEW FINANCIAL CAPITALISTS: KOHLBERG KRAVIS ROBERTS AND THE CREATION OF CORPORATE VALUE 53-56 (1998); BRYAN BURROUGH AND JOHN HELYAR, BARBARIANS AT THE GATE: THE FALL OF RJR NABISCO 133-36 (1990). 272

8-Oct-08] Failure of Private Equity 9 private equity firm Kohlberg, Kravis & Roberts Co.34 The trio at KKR raised the industry’s first equity fund in 1978.35 This provided a pre- arranged source of committed equity capital. However, at this time debt for acquisitions was still raised on an ad hoc basis and was largely limited to secured credit financing from bank lenders.36 If private equity was to grow, a steady source of debt financing was required which would permit a larger amount of debt to be incurred for acquisitions.

This source would be pioneered by the brilliant and infamous Michael Milken and the firm he worked for, Drexel Burnham Lambert. Throughout the 1970s and 1980s, Michael Milken and his colleagues at Drexel had been working to create a larger market for high yield debt, often derogatorily known as junk bonds.37 This debt was often referred to as junk because this debt was either unrated or rated below investment grade and was subordinated to other senior more highly rated debt.38 Historically, high yield debt was shunned by investors and utilized by small issuers who had fewer financing choices.39 Milken had studied this market and found that investors in this debt had historically realized extraordinary returns.40 He popularized this finding and soon convinced many institutional and other investors to purchase the high yield debt offerings Drexel underwrote.41 Milken needed an even larger supply of issuers of these securities to fulfill the demand he had largely created.

In private equity Milken found a large source; from the mid-1980s private equity acquisitions became one of the principal issuers of high yield securities.42 Private equity firms during this time would use traditional senior secured loans together with high yield and other debt-type securities to increase the debt level on individual acquisitions.43 The additional funds provided by this high yield financing allowed private equity to make larger and more frequent company purchases.44 It would be the nature of this debt

34 See Allen Kaufman and Ernest J. Englander, Kohlberg Kravis Roberts & Co. and the Restructuring of American Capitalism, 67 Bus. Hist.Rev. 52, 67-68 (1993). 35 Id. at 71. See also GEORGE ANDERS, MERCHANTS OF DEBT: KKR AND THE MORTGAGING OF AMERICAN BUSINESS (1992). 36 See Brian Cheffins & John Armour, The Eclipse of Private Equity, at 17 (Apr 2007), available at http://ssrn.com/abstract=982114. 37 DANIEL R. FISCHEL, PAYBACK: THE CONSPIRACY TO DESTROY MICHAEL MILKEN AND HIS FINANCIAL REVOLUTION (1995). 38 See PATRICK A. GAUGHAN, MERGERS, ACQUISITIONS, AND CORPORATE RESTRUCTURINGS 330 (3rd ed., 2002); Scott P. Mason and Bill Hildebolt, History and Analysis of the High-Yield Debt Market. 39 FISCHEL, supra note 37, at [•]. 40 See CONNIE BRUCK, THE PREDATOR’S BALL: HOW MICHAEL MILKEN AND HIS JUNK BOND MACHINE STAKED THE CORPORATE RAIDERS (1988). 41 Id. at [•]. 42 See BURROUGH & HELYAR, supra note 33, at [•]; Cheffens & Armour, supra note 36, at 17. 43 Id. 44 See Kaufman & Englander, supra note 34, at 76-79. 273

10 82 S. Cal. L. Rev. __ [8-Oct-08 financing, and the needs of the investment banks underwriting or originating it, that would drive the structure of private equity acquisitions. The structure most commonly used in the 1980s through the 1990s can be diagramed as follows:

274

8-Oct-08] Failure of Private Equity 11

Chart 1A45

Financing Private Bank(s) Private Equity Equity Debt Financing Fund Structure (Commitment letter 100% w/market out or Ownership interest Circa 1985- “highly confident” early 1990s letter)

Parent Equity Infusion (no agreement)

Financing Condition

Merger Target Subsidiary

In the above structure the private equity buy-out was effected by thinly capitalized shell subsidiaries set up specifically for this purpose by the private equity firm (Parent and Merger Subsidiary in the above diagram).46 The shells had no substantial assets of their own. Instead, the acquisition agreement required that the shells use a measure of “best efforts” to complete the transactions contemplated by the agreement.47 Since the shells had no real assets, companies to be acquired (Acquiree in the above diagram) demanded assurances that the financing would be available. So, these arrangements were also typically accompanied by a debt financing commitment letter from an investment and possibly commercial bank (Financing Bank in the above diagram).48 The banks would provide senior bank credit facilities, but would also act as underwriters for selling any high yield debt in the market and for any other offering.49

Importantly, the debt commitment letter was not a binding arrangement

45 See RJR Nabisco Holdings Corp. Registration Statement on Form S-1, dated 1989. 46 The fund uses separate shell subsidiaries to effect this acquisition in order limit its liability among other reasons. 47 See, e.g., id at [•]. 48 See BURROUGH & HELYAR, supra note 33, at [•]. 49 See BRUCK, supra note , at [•]. 275

12 82 S. Cal. L. Rev. __ [8-Oct-08 to provide funds, rather the debt commitment letter was an agreement to negotiate definitive financing arrangements on the terms set forth in the commitment letter. In addition, the commitment letter was executed at the time the acquisition agreement was executed. The final documentation was not signed until the transaction completed some months later. It was at this time that the banks would extend any loans and attempt to sell the high yield debt to finance the acquisition.

However, because there was this period between the signing of the acquisition agreement and completion of the transaction, there was substantial risk for the banks. The banks had agreed to extend this credit under terms set forth in the commitment letter. If market conditions changed or interest rates fluctuated in the wrong direction, the banks would still be obligated to fund under the old terms set forth in the commitment letter. In such a case, when the banks went to sell the debt issued in connection with the transaction they might have to charge a lower price for it than expected when the agreements were first signed, thereby incurring a loss. In extreme circumstances they might be unable to sell the debt entirely leaving them stuck holding the entire financing package.

To address this issue, the banks typically negotiated commitment letters which contained a “market out” clause, a clause which permitted the banks to terminate their financing obligations if market conditions deteriorated or otherwise impeded placement or incurrence of the debt.50 Due to the high leverage on these transactions, banks were often unwilling to even provide this level of commitment. In such circumstances, the banks would issue a “highly confident” letter.51 These letters were pioneered by Drexel Burnham in financings where the success of the debt issuance was too uncertain to provide any firm written commitments.52 The financing banks would instead opine that they were “highly confident” that the debt could be raised in the markets but provide no contractual agreement to do so.53

In either case, though, the private equity fund itself was not liable if the transaction failed to close. Due to the uncertainty of the debt financing, private equity firms refused to commit themselves to entirely fund the

50 See David J. Sorkin & Eric M. Swedenburg, Recent Developments in Financing-Related Provisions in Leveraged Buyouts, at 1-2 (Jan. 2006), available at http://lawprofessors.typepad.com/mergers/files/simpson_jan_2006_client_memo.pdf) 51 See BRYAN BURROUGH & JOHN HELYAR, BARBARIANS AT THE GATE: THE FALL OF RJR NABISCO (2003). 52 See BRUCK, supra note , at [•]. 53 See BRUCE WASSERSTEIN, BIG DEAL : 2000 AND BEYOND (2000). 276

8-Oct-08] Failure of Private Equity 13 acquisition if the debt financing failed.54 Acquirees typically agreed to this demand. Since the private equity firms had no contractual obligation to fund the acquisition, this effectively provided private equity firms with an ability to exit from the buy-out any time before consummation of the acquisition for any reason even beyond failure of the debt financing. The acquisition agreement also permitted the shell to terminate the agreement if financing was unavailable. This was accomplished by placing a financing condition in the acquisition agreement, conditioning the shell’s obligation to acquire the acquiree on the shell having obtained sufficient financing to do so.55

B. The Shifting Structure of Private Equity

In the 1990s the private equity structure continued to evolve in response to market forces. Acquirees began to contractually bind the private equity firms themselves. They did this by demanding and receiving equity commitment letters from the private equity firms’ sponsoring fund.56 These letters obligated the fund to supply the shells with the necessary equity to complete the transaction. This filled the equity gap in financing these transactions, providing a contractual commitment for the shell subsidiaries to access the necessary equity component of their financing. The debt commitment letter was thus paired with an equity commitment letter. Notably, this new mechanism placed the first real limitation on the ability of private equity firms to exit transactions – the equity commitment letter now contractually bound the private equity firm’s fund to provide the shell subsidiary the equity investment in the acquisition. During this time period, the terms of debt commitment letters also shifted. The principal change in these letters was the inclusion of bridge financing.57 Bridge financing is interim financing between the completion of the transaction and the placement of any permanent debt financing.58 The addition of bridge financing thus provided increased certainty to the acquiree that the transaction would be completed if the offering of any of the permanent debt was delayed.

54 The equity capital was the share ownership stake and together with debt financing typically consisted of the bulk of the funds used to purchase the acquiree. 55 See Sorkin & Swedenburg, supra note 50, at 2. 56 Id. The prior absence of this equity commitment is attributable to the origins of private equity. Initially in the late 1970s and early 1980s private equity firms largely funded the equity component of the transaction through syndication begun after the transaction announcement. See, e.g., Transaction Documents for Gibbons Greeting; Coca Cola Bottling of New York. An equity commitment letter was inappropriate at that time since there yet was any equity to be committed and the parties who would commit the equity were still unknown. 57 See Bridge Loans Make A Comeback, CORPORATE GROWTH REP., Aug. 14, 1995, at 7987; A New Generation of Bridge Lending, MERGERS & ACQUISITIONS, Sept./Oct. 1994, at 6. 58 See Nancy H. Wolitas, et al, Financing Options for Issues Related to Leveraged Buy-Outs in FOURTH ANNUAL PRIVATE EQUITY FORUM: LEGAL & FINANCIAL STRATEGIES FOR DEALMAKING IN THE CURRENT MARKET , at 456-57 (PLI Instit. Oct.-Dec. 2002). 277

14 82 S. Cal. L. Rev. __ [8-Oct-08

The structure of private equity deals further evolved in the new millennium.59 A significant shift in transaction structure was triggered by the March 2005 $11.3 billion buy-out of SunGard Data Systems by a private equity consortium.60 The SunGard transaction was structured as follows:

59 I discuss the reasons for this shift infra at Part IV. 60 See Andrew Ross Sorkin, Private Investment Firms to Pay $11.3 Billion for SunGard Data, THE N.Y. TIMES, Mar 28, 2005. See also Martin Sikora, LBO Funds Offer Incentives To Drive High-Priced Deals: Groups propose "reverse" breakup fees, dropping the financing out while angling for SunGard and Neiman Marcus buys, M&A: THE DEALMAKERS’ J.. Sept. 1, 2005 (describing the SunGard transaction as “a major shift from traditional leveraged dealmaking”). 278

8-Oct-08] Failure of Private Equity 15

Chart 1B61

Financing Bank(s) Private SunGard Equity Debt Financing Fund Structure (Commitment letter 100% w/bridge financing) Ownership interest Circa 2005 Limited Market- Out – Mirror Conditions

Parent Equity Infusion (w/agreement)

Financing No Recourse Condition Guarantee Added Of Reverse Termination Merger Target Fee Reverse Subsidiary Termination Fee Added

Comporting with the prior historical structure, the equity portion of the transaction was set forth in an equity commitment letter executed by the private equity firms’ funds.62 The debt portion of the transaction was agreed through a commitment letter by five investment banks and included a bridge financing facility.63 SunGard also negotiated the removal of the financing condition from the main agreement between SunGard and the private equity consortium. In addition to negotiating the deletion of a financing condition, SunGard was able to obtain a debt commitment letter which had conditions reciprocal to those in the acquisition agreement. In other words, if the conditions to the acquisition agreement were satisfied so would the debt commitment letter conditions. By better aligning the terms of the debt commitment letter and the main acquisition agreement, the financing for the transaction was more certain if the conditions in the main agreement were fulfilled. Finally, the SunGard debt commitment letter contained a limited “market out” and “lender out”.64 The result was a transaction structure

61 This chart was prepared from information contained in the SunGard Definitive Proxy Statement on Schedule 14A, at 49-54 (June 27, 2005), available at http://www.sec.gov/Archives/edgar/data/789388/000119312505131157/ddef14a.htm [hereinafter SUNGARD PROXY STATEMENT]. 62 Id. at 49-50. 63 Id. at 50-53. 64 In the words of two prominent practitioners, “a ‘Lender [out]’ occurs, for example, in the event that the lending sources are prohibited from funding due to legal prohibitions or lender insolvency.” See Sorkin & Swedenburg, supra note 50, at 3. 279

16 82 S. Cal. L. Rev. __ [8-Oct-08 more favorable to the acquiree since completion was contractually more certain.65 Importantly, though, by agreeing to a more certain debt commitment letter and providing bridge financing, the banks now took on the risk of a market deterioration in between the time of the signing of the transaction and the closing. If the value of the debt declined during this time period, the banks would suffer the loss. This would appear to be a rational decision in 2005 and the days of easy credit, but it would be a decision that would turn out to savagely haunt these financial institutions.

In exchange for agreeing to the removal of the financing condition in the main agreement, SunGard also agreed to a cap on the private equity consortium’s maximum liability for breach of the acquisition of $300 million and a bar on specific performance in the acquisition agreement.66 In other words, if the shell was unable to complete the buy-out because the financing arrangements failed or otherwise because it refused to do so, i.e., the agreement was intentionally breached, the private equity funds only liability was to pay a fee of $300 million to SunGard as compensation. The fee was called a reverse termination fee because it was patterned upon termination fees that acquirees typically agreed in acquisition agreements to pay acquirers if they subsequently accepted a higher offer from another bidder. The reverse termination fee in the SunGard transaction amounted to three percent of the transaction value and was the same amount as the termination fee.67 And since the shells were still that, empty corporations without substantial funds, the private equity funds issued a guarantee for this payment.68

This type of structure had previously been utilized in other transactions, but apparently not in private equity deals.69 The SunGard “structure” was the first private equity transaction of any significance to employ such architecture.70 After SunGard the structure quickly took hold in private equity transactions. The following chart sets forth my own calculations as to the percentage of private equity deals utilizing a reverse termination fee structure from 2004 through 2008:

65 See infra notes 178-179 and accompanying text. 66 See SUNGARD PROXY STATEMENT, supra note 61, at 91. 67 Id. 68 Id. at 54. 69 These included the buy-outs of Extended Stay America, Prime Hospitality, Boca Resorts, Wyndham International and La Quinta Corporation. 70 Email from Alan Fishbein, dated Sept. 10, 2008. 280

8-Oct-08] Failure of Private Equity 17

Chart 1C 71

Chart 1C shows a rapid shift in practice as the use of financing conditions in acquisition agreements dropped in inverse proportion to utilization of the reverse termination fee structure. The reverse termination fee became the norm in private equity acquisitions, but this structure sometimes varied depending upon the agreement of the parties. For example, in the 2005 private equity buy-out of Neiman Marcus, the private equity acquirers agreed to a two-tiered termination fee.72 A lower fee would be paid if the private equity fund breached the agreement and failed to complete the transaction due to a failure of financing. A higher fee phrased as a cap on the private equity consortium’s maximum liability would be paid if the private equity fund willfully breached the agreement and refused to complete the transaction when all of the conditions to completion, including financing, were satisfied. A third variation of this arrangement arose in other buy-outs, such as that of Penn National Gaming, where the acquiree could under the terms of the contract force the private equity shells to specifically perform and enforce the debt and equity commitment letters to complete the transaction.73 If for some reason the debt or equity financing

71 Author calculations based on data provide by Mergermetrics (Search data on file with author). Information is only for announced public private equity deals greater than $100 million in value and for which a transaction document was available. 72 The Neiman Marcus Group Inc. Definitive Proxy Statement on Schedule 14A, at 67-68 (July 18, 2005), available at http://www.sec.gov/Archives/edgar/data/819539/000119312505143823/ddefm14a.htm [hereinafter NEIMAN MARCUS PROXY STATEMENT]. 73 The Penn National Gaming, Inc. Definitive Proxy Statement on Schedule 14A, at 90-91 (Nov 7, 2007), available at http://www.sec.gov/Archives/edgar/data/921738/000119312507238601/0001193125-07-238601- index.htm 281

18 82 S. Cal. L. Rev. __ [8-Oct-08 became unavailable then termination of the agreement and receipt of the reverse termination fee was the acquiree’s only remedy.

In its first two variations, this arrangement provided a flat-out ability to the private equity firm to exit a transaction simply by paying the reverse termination fee. The third was supposed to be a more certain structure for acquirees because so long as the debt and equity commitment letters were enforceable, the private equity funds could not simply “walk” on a transaction. Rather, the acquiree could go to court to force the shell subsidiaries to enforce and draw on the debt and equity commitment letters to complete the acquisition. Finally, the Neiman Marcus transaction was notable for completely eliminating the “market-out” in debt commitment letters, a change to the structure that would persist in its variations.74

The reverse termination fee structure in its variations was the blueprint of private equity heading into Summer 2007. I will discuss the reasons for this shift in structure infra at Part IV, but before I do it is necessary to first talk about the events of Fall 2007 and onward.

II. THE FAILURE OF PRIVATE EQUITY

The structure of private equity appeared to have achieved a new equilibrium after a significant transformatory period from 2005-2007. Lawyers in private equity transactions had leveraged the increased willingness of lenders to provide more flexible terms to renegotiate the boiler-plate of private equity. However, the credit crisis ensuing in the summer of 2007 and subsequent events over the course of the following year would expose flaws in this structure. In this Part II, I discuss the events of the summer of 2007 and afterwards. These would be the mother of all shocks to the structure of private equity. The ensuing events would reveal not only the inherent problems with the private equity structure, but would open up a further window into the manner of negotiation of these complex contracts and their evolution.75

A. The Failure of Contract

The first public signs of a significant disruption in the takeover markets emerged in late July and early August of 2007. During that period acquirers in two significant public takeover transactions, the pending acquisitions of

74 See NEIMAN MARCUS PROXY STATEMENT, supra note 72, at [•]. 75 See Steven M. Davidoff, Private Equity’s Option to Buy, M&A LAW PROF, Aug. 16, 2007, available at http://lawprofessors.typepad.com/mergers/2007/08/private-equitys.html 282

8-Oct-08] Failure of Private Equity 19

Accredited Home Lenders and Radian, attempted to terminate their acquisition agreements. Each of these acquirers invoked a material adverse change (MAC) clause in the agreement to do so.76 A MAC clause provision is a clause in an acquisition agreement which permits an acquirer to refuse to complete the transaction if a material and adverse change as defined in the acquisition agreement occurs to an acquiree prior to the time of completion of the acquisition.77 The invocation of these clauses in these two instances was a sign that market and economic turbulence was beginning to affect takeover transactions. However, these two, first MAC claims were confined to the industries most relevantly affected by the summer subprime mortgage crisis – both of the affected acquiree companies Accredited Home Lenders and Radian were in the subprime lending business.78 Moreover, both of these transactions utilized a non-private equity acquisition structure. Rather, each was structured in the traditional strategic manner and did not have a financing out or a reverse termination fee provision, were backed fully by the assets of the acquirer, and provided for specific performance of the transaction.79

But as August progressed, stock market volatility increased and the credit markets became increasingly illiquid, public attention turned to the $[•] billion in pending private equity transactions.80 In light of the disruption in this market a number of public commentators and news sources began to report on the private equity reverse termination structure, questioning the willingness of private equity firms to complete these acquisitions. The first prominent news piece was published in the N.Y. Times on August 21, 2007 and was entitled “Can Private Equity Firms Get Out of Buyouts?” 81 The article, by Andrew Ross Sorkin, highlighted the reverse termination fee structure now common-place in private equity buy- outs, explored the willingness of private equity acquirers to terminate these transactions, and discussed the reputational constraints on their ability to do so.82

76 These four transactions were the proposed takeovers of Accredited Home Lenders, the home supply unit of Home Depot, Inc., SLM Corp. and Radian. 77 Wei Lingling, Accredited Home Sues Lone Star to Save Deal, THE WALL ST. J., Aug 14, 2007, at C; Paul Gores, MGIC sues merger partner for data Subprime crisis prompts mortgage insurer's lawsuit, MIL. J. SEN., Aug 22, 2007, at D. 78 See Steven M. Davidoff & Kristen Baiardi, Accredited Home Lenders v. Lone Star Funds: A MAC Case Study (Feb. 11, 2008). Wayne State University Law School Research Paper No. 08-16 Available at SSRN: http://ssrn.com/abstract=1092115. 79 See supra note 77. 80 Pali Capital Arbitrage Situations (Jul 31, 2007). 81 Andrew Ross Sorkin, Can Private Equity Firms Get Out of Buyouts?, THE N.Y. TIMES, Aug 21, 2007. See also Steven M. Davidoff, Private Equity’s Option to Buy, Aug 16, 2007, at http://lawprofessors.typepad.com/mergers/2007/08/private-equitys.html. 82 Id. 283

20 82 S. Cal. L. Rev. __ [8-Oct-08

During August and through mid-November, private equity firms in two pending public transactions with reverse termination fee structures did indeed attempt to terminate acquisitions agreed prior to the summer credit crisis. These involved the buy-outs of Acxiom and Harman.83 However, the acquirers did not invoke the reverse termination fee provisions negotiated in their transaction agreements. Rather, these private equity acquirers asserted real or ostensible MAC claims to terminate their obligations.84

They did so for at least three reasons. First, the deterioration in the markets and general economy provided a colorable basis to make this assertion. Second, a MAC claim provided reputational cover; instead of being labeled as “walking” on their contractual obligations, a MAC claim provided historically legitimate grounds for an acquirer to terminate the transaction; it is generally perceived as acceptable for a acquirer to invoke a MAC.85 Finally, a MAC claim provided negotiating leverage to the private equity firm. Under the terms of each of these agreements, if the private equity firm was successful in claiming a MAC it could terminate the agreement without any required payment to the acquiree.86 Moreover, the maximum liability of the private equity firms if their MAC claim failed was capped at the reverse termination fee.87 The assertion of a MAC in combination with a reverse termination fee provision thus provided the private equity firms with negotiating leverage by setting their maximum liability in any settlement or litigation.88

Both of these MAC claims were ultimately settled through an agreement among the parties which terminated the acquisition agreement.89 The legitimacy of these MAC claims and the dynamic created by the interaction of these claims with the reverse termination fee was revealed by the amounts the private equity firm ultimately paid to the acquirees to terminate

83 See generally Paul S. Bird, Deals Redefined, THE DAILY DEAL, Dec. 19, 2007. 84 See id. See also Michael De la Merced, When a Deal is no longer a Deal, THE N.Y. TIMES, Sept 24, 2007. 85 See Andrew Ross Sorkin, After the Party, THE N.Y. TIMES, Oct 3, 2007. 86 See Agreement and Plan of Merger by and among Axio Holdings LLC, Axio Acquisition Corp. and Acxiom Corporation, Dated as of May 16, 20077 available at http://www.sec.gov/Archives/edgar/data/733269/000073326907000018/ex2-1mergeragmt.htm (Section 8.1) [hereinafter ACXIOM AGREEMENT]; Agreement and Plan of Merger among KHI Parent Inc., KHI Merger Sub Inc., and Harman International Industries, Incorporate, Dated as of April 26, 2007 available at http://www.sec.gov/Archives/edgar/data/800459/000095013407009341/d45945exv2w1.htm (Section 7.01) [hereinafter HARMAN AGREEMENT]. See also Kelly Holman, Big MAC Attack? Material adverse change' clause getting more attention, INV’T DEALERS’ DIGEST, Oct 1, 2007. 87 See ACXIOM AGREEMENT, supra note 86, at Section 9.8; HARMAN AGREEMENT, supra note 86, at Section 7.02. Compare this with the MAC claim in Accredited Home Lenders at the same time. In that deal the parties renegotiated the transaction 88 Interview with Acxiom General Counsel. Jerry C. Jones, Oct 10, 2007. 89 See Kelly Holman, Breaking Up is (Less) Hard to Do: PIPE-like deal resolves Harman dispute, INV. DEALERS' DIG., Oct 29, 2007. 284

8-Oct-08] Failure of Private Equity 21 the transaction. In each case the payment was near to the reverse termination fee amount.90 Thus, in this early Fall period private equity firms could be seen as attempting to avoid reputational tarnish by asserting MAC claims to avoid being viewed as invoking the reverse termination fee provisions. The validity of these MAC claims was belied by the amounts privately negotiated and paid by the private equity firms; the settlement approximated the reverse termination fee. The result was beneficial to the private equity firms – it may have protected their reputation -- but their actions left acquirees publicly damaged by these claims of an adverse event to their business. In most of these cases this also left the acquirees’ stock prices trading below their price prior to the announcement of the acquisition agreement.91

B. The Failure of Norms

In Federalist Paper Number 15, Alexander Hamilton observed that reputation is a “less active influence” constraining behavior when a nefarious deed is done by many.92 Hamilton’s observation aptly applies to the events surrounding the Fall 2007 wave of private equity acquisition terminations. Initially, no single private equity firm was willing to stain its reputation and harm its competitive position in the buy-out market by invoking a reverse termination fee provision. Instead, these firms asserted MAC claims to publicly justify termination and avoid being labeled as “walking” on their transactions and an untrustworthy future acquirer. However, as the Fall progressed the reputational forces on private equity firms to complete buy-outs became diluted as the credit markets remained illiquid and the number of terminated private equity deals increased. The reputational impact was diluted.93

This prominently evidenced itself on November 14, 2007 when the private equity fund controlled by Cerberus attempted to terminate its

90 The Acxiom agreement provided for a two-tiered reverse termination fee. However, the MAC clause invocation permitted the private equity firm to treat it as a failure of financing to force a settlement in an amount equal to 97.38% of the lower reverse termination fee. See Interview, supra note 88; Acxiom Press Release, October 10, 2007 available at http://www.acxiom.com/72451/Acxiom_Release. In Harman the $200 million reverse termination fee was invested into the company as part of a $400 million investment in connection with the termination of the agreement to be acquired. See Harman Press Release, KKR And GS Capital Partners To Invest In Harman International, Oct 22, 2007, available at http://www.harman.com/press/financial_press.aspx?st=. 91 See Elizabeth Nowicki, Private Equity Deals of 2007: Lessons To Learn, at 5 (2008) (unpublished draft on file with author) (listing the post-termination declines in stock prices of private equity acquirees in failed transactions). 92 Alexander Hamilton, Federalist Paper No. 15, The Insufficiency of the Present Confederation to Preserve the Union For the Independent Journal. 93 See Karen Donovan, Private Equity: Breaking Up Is Not That Hard to Do, Portfolio.com, Dec. 27, 2007, available at http://www.portfolio.com/views/blogs/daily-brief/2007/12/27/private-equity-breaking-up-is- not-that-hard-to-do (“Private-equity firms, meanwhile, seem to be saying: Reputation? What reputation? Here's a $100 million and watch me walk away.”) 285

22 82 S. Cal. L. Rev. __ [8-Oct-08 agreement to acquire United Rentals.94 Cerberus did not assert a MAC to justify its action. Rather, the shell subsidiaries owned by Cerberus and who were the parties to this agreement simply invoked the reverse termination provision in the acquisition agreement.95 Cerberus argued that this provision permitted it to terminate its obligations for any reason upon payment of a $100 million reverse termination fee.96 Cerberus had decided that any reputational impact was overcome by the declining economic return of the transaction. In assessing the reputational damage Cerberus was no doubt influenced by prior private equity terminations and their dilutive effect on any such reputational loss.97

United Rentals sued the Cerberus shell subsidiaries in Delaware Chancery Court challenging their attempt to terminate the agreement.98 United Rentals argued that the contract provided for United Rentals to require specific performance of the shell subsidiaries obligations.99 In other words, the parties’ dispute centered upon the type of reverse termination fee structure they had negotiated, the pure reverse termination fee or specific performance structure. United Rentals argued that this contract provided for specific performance of the shell subsidiary entities financing commitments. Only if the financing then failed could the entities terminate the agreement.100 The Cerberus shell entities argued that the same language of the contract barred specific performance and that their only liability was for $100 million.101 The suit was complicated by the fact that the acquisition agreement was governed by Delaware law and had a Delaware choice of forum clause while the guarantee provided by the Cerberus fund of the shell entities’ obligations was governed by New York law and had a New York choice of forum clause.102 The Cerberus fund leveraged this disjunction to bring a separate suit in New York for declaratory relief that the separate

94 See United Rentals Current Report on Form 8-K, dated Nov. 14, 2007, available at http://www.sec.gov/Archives/edgar/data/1067701/000101905607001167/0001019056-07-001167-index.htm 95 Id. 96 See Schedule 13D/A filed by RAM Holdings, Inc., dated Nov. 14, 2007, at 2, available at http://www.sec.gov/Archives/edgar/data/1067701/000090571807000317/united13dam1.txt 97 See Donovan, supra note 93; Andrew Ross Sorkin, If Buyout Firms Are So Smart, Why Are They So Wrong?, THE N.Y. TIMES, Nov. 18, 2007. 98 See United Rentals Inc. v. RAM Holdings Inc. and RAM Acquisition Corp., dated Nov. 19, 2007, available at http://lawprofessors.typepad.com/mergers/files/URI_complaint.pdf. See also Matthew Karnitschnig and Lingling Wei, Economy Conspires to Dog Cerberus United Rentals Sues Over Collapsed Deal; Chrysler Loans Stall, THE WALL ST. J., Nov. 20, 2007, at C1. 99 See United Rentals, Inc. v. RAM Holdings, Inc., et al., 937 A.2d 810, 830-31 (Del. Ch. 2007). 100 Id. 101 Id. at 832-833. 102 See Steven M. Davidoff, Cerberus Sues in New York, M&A Law Prof (Nov. 23 2007), available at http://lawprofessors.typepad.com/mergers/2007/11/cerberus-sues-i.html. The reason for this is that bank finance documents have traditionally been governed by New York law while in this time period due to an adverse case decided under New York law, acquisition agreements were governed by Delaware law. The choice of forums followed from the choice of law. I discuss why lawyers could agree to such a disjunction infra at notes 250-254. 286

8-Oct-08] Failure of Private Equity 23

terms of its guarantee limited its liability to $100 million.103 The New York suit was never substantively litigated, though, as Chancellor Chandler, the judge in the Chancery Court, beforehand found that the contract language was ambiguous.104 This case was ultimately a contract dispute, and Chancellor Chandler applied standard contract interpretation principles to hold in favor of Cerberus’s reading of the agreement.105 When United Rentals announced that it would not appeal this decision, Cerberus promptly terminated the acquisition agreement and paid United Rentals $100 million.106

The Cerberus/United Rentals dispute and Cerberus’s subsequent termination of its agreement resulted in a further deterioration of the reputational force preventing exercise of a reverse termination fee provision. In the period from December through February, 2008 three additional private equity transactions would be effectively terminated, the pending acquisitions of PHH, Reddy Ice and Myers Industries.107 In each case, no MAC claim was publicly asserted, but instead the acquirers merely exercised the reverse termination fee provision in its agreement to exit the transaction.108 In each instance, the agreement clearly prohibited specific performance and permitted this action.109 Thus by early 2008 the fundamental understandings of the parties in private equity agreements appeared to have fallen by the wayside and the inherent optionality in this type of a reverse termination fee structure realized. A reverse termination fee provision appeared to become exercisable without significant reputational impact or other external normative constraints.

103 See Cerberus Partners, L.P., et al. v. United Rentals, Inc., No. 7603876 (Nov. 21, 2007). 104 United Rentals, 937 A.2d at 834. 105 Chancellor Chandler found the language of the contract to be ambiguous and therefore looked to the extrinsic evidence to find an objective, reasonable meaning of the contract language. The court then applied the forthright negotiator rule to find that United Rentals knew or should have known Cerberus’s understanding of the contract and did not correct it. Id. at 834-843. 106 See United Rentals, Inc. Current Report on Form 8-K, dated December 26, 2007, available at http://www.sec.gov/Archives/edgar/data/1047166/000101905607001342/ur_8k.htm. 107 See PHH Corporation Current Report on Form 8-K, dated Jan. 7, 2008, available at http://www.sec.gov/Archives/edgar/data/77776/000095012308000112/y46075e8vk.htm; Reddy Ice Current Report on Form 8-K, dated Feb. 1, 2008, available at http://www.sec.gov/Archives/edgar/data/1268984/000110465908006289/a08-4392_18k.htm; Meyers Industries, Current Report on Form 8-K, dated Dec. 10, 2007, available at http://www.sec.gov/Archives/edgar/data/69488/000006948807000074/form8k121007.htm. 108 Id. 109 See Agreement and Plan of Merger by and among Myeh Corporation, Myeh Acquisition Corporation and Myers Industries, Inc., dated Apr. 24, 2007 available at http://www.sec.gov/Archives/edgar/data/69488/000006948807000037/exhibit101.htm (Section 8.4(f)); Agreement and Plan of Merger by and among PHH Corporation, General Electric Capital Corporation, and Jade Merger Sub, Inc., dated Mar 15, 2007 available at http://www.sec.gov/Archives/edgar/data/77776/000095012307003874/y32034exv2w1.htm (Section 9.10); Agreement and Plan of Merger by and among Frozen LLC, Hockey Parent Inc., Hockey Mergersub, Inc. and Reddy Ice Holdings, Inc., dated Jul 2, 2007, available at http://www.sec.gov/Archives/edgar/data/1268984/000110465907051661/a07-17958_1ex2d1.htm (Section 9.7(b)). 287

24 82 S. Cal. L. Rev. __ [8-Oct-08

C. The Failure of Specific Performance

The economics and parameters of the pure reverse termination fee structure were largely redefined by the Fall wave of collapsed private equity acquisitions. By 2008 most of these deals had either been terminated or consummated in accordance with their terms. However, into the new year a number of significantly larger multi-billion dollar private equity transactions remained pending.110 The majority of these were structured utilizing a specific performance reverse termination fee structure rather than a pure reverse termination fee.111 The closing of these transactions was delayed into the winter of 2008 due to regulatory or financing issues.112 At the time many speculated that these deals remained outstanding in part due to their less “optional” structure: the provision of specific performance prevented the private equity firms from simply terminating the agreement unless financing became unavailable.113 Given that the acquirers couldn’t simply terminate their obligations, they instead waited, delaying the deal and hoping the credit and stock markets improved sufficiently to make the economics of their transactions again viable.

In the new year 2008 as the credit crisis continued and the economic cycle trended further downward, these transactions continued to be stressed by extrinsic shocks.114 The result was another wave of litigation, this time implicating the viability of this second form of private equity structure. The first of these disputes occurred at the end of January, 2008 and arose out of the pending sale of Alliance Data Services to funds affiliated with The Blackstone Group. At that time, it was disclosed that the Office of the Comptroller of the Currency was refusing to grant a required regulatory approval for ADS to be acquired by Blackstone.115 The OCC justified its refusal on the grounds that the post-acquisition leverage of ADS would leave ADS insufficiently capitalized to support its national bank subsidiary.116 The OCC did however express a willingness to reverse its

110 The backlog of pending U.S. private equity transactions as of January 9, 2008 was approximately $108 billion. Pali Arbitrage Spreadsheet (excludes transaction less than $100 million). 111 The five biggest pending deals were Alliance Data Services, BCE, Clear Channel, Huntsman and Penn National Gaming. All but Clear Channel contained a specific performance form of the reverse termination fee structure. 112 See generally Bill Barnhart, Acquisitions, mergers follow natural course, CHI. TRIB., Jan. 30, 2008. 113 See Steven M. Davidoff, Who’s Next for the Deal Dead Pool?, N.Y. TIMES DEALBOOK, Jan 10, 2008, available at http://dealbook.blogs.nytimes.com/2008/01/10/whos-next-for-the-deal-dead-pool/. 114 See Heidi N. Moore, The Buyout Financing Backlog: Are We There Yet?, THE WALL ST. J. DEAL J., Jan. 30, 2008, available at http://blogs.wsj.com/deals/2008/01/30/the-buyout-financing-backlog-are-we-there-yet/ (reporting that the backlog of private equity-related debt remained at over $250 billion as of January 2008). 115 See Michael J. de la Merced, Deal to Buy Credit Card Processor Is in Peril, THE N.Y. TIMES, Jan. 29. 2008. 116 See Complaint for Alliance Data Systems Corporation v. Aladdin Solutions, Inc. et al., Civil Action 288

8-Oct-08] Failure of Private Equity 25 position if the acquiring Blackstone fund itself provided a back-stop, uncapped guarantee of ADS’s bank liabilities effective upon completion of the sale.117

ADS sued in Delaware Chancery Court to compel the Blackstone fund to provide this guarantee.118 In accordance with the boiler-plate of the specific performance reverse termination fee structure used at that time, ADS had negotiated an acquisition contract which provided that ADS could sue to force performance of the Blackstone shell subsidiaries’ obligations under the agreement.119 This arguably included the subsidiaries’ contractual obligation to use “reasonable best efforts” to obtain any necessary regulatory approvals, including OCC clearance, for the transaction.120 ADS argued in court that the requirement to use “reasonable best efforts” by the shell subsidiaries required them to sue the Blackstone fund itself, their parent, to compel it to issue the OCC requested guarantee.121 Blackstone countered that ADS had only entered into the acquisition agreement with thinly capitalized shell subsidiaries, a fact which ADS was fully aware of at the time it entered into the agreement.122 The Blackstone fund’s only obligation was under its equity commitment letter issued to these subsidiaries and its own guarantee of the reverse termination fee.123 Therefore, the shell entities could not force the Blackstone fund to provide the OCC guarantee, and since these entities could not provide the guarantee required by the OCC the transaction could not be completed.124

Blackstone’s response highlighted a fundamental limitation of the specific performance form of private equity structure. The private equity shell subsidiaries are corporate limited liability entities whose only real assets are their financing commitments and agreement to acquire the acquiree. If regulators or other events require the shell subsidiaries to act beyond these assets, specific performance becomes meaningless since no assets are available. The agreement thus effectively becomes unenforceable

No. 3507-VCS, at ¶¶8-10, Jan. 29, 2008, available at http://www.sec.gov/Archives/edgar/data/1101215/000129993308000505/exhibit2.htm [hereinafter ALLIANCE DATA COMPLAINT]. 117 Id. 118 Id. See also Michael J. de la Merced, Credit Card Processor Sues Blackstone to Finish Buyout, THE N.Y. TIMES, Jan. 31, 2008. 119 See Agreement and Plan of Merger by and among Aladdin Holdco, Inc., Aladdin Merger Sub, Inc. and Alliance Data Systems Corporation, Dated as of May 17, 2007, available at http://www.sec.gov/Archives/edgar/data/1101215/000095013407011838/d46889exv2w1.htm (Section 9.8.2). 120 See Alliance Data Complaint, supra note 116, at ¶ 11-13. 121 See Hearing Transcript, Alliance Data Systems Corporation v. Aladdin Solutions, Inc. et al., Civil Action No. 3507-VCS, at 13-15, Feb. 4, 2008, available at http://graphics8.nytimes.com/images/blogs/dealbook/ADSxscript.pdf [hereinafter ADS HEARING]. 122 Id. at 31-35. 123 Id. 124 Id. 289

26 82 S. Cal. L. Rev. __ [8-Oct-08 unless the private equity fund voluntarily agreed to support any such arrangements.

ADS attempted to sidestep this dilemma by arguing that the “reasonable best efforts” clause in its acquisition agreement contemplated more, a fact that the parties were aware of at the time of the agreement’s negotiation.125 The shell subsidiaries could be required under this clause to sue their parent Blackstone fund for any additional sums or contractual requirements required to satisfy regulatory demands. However, the meaning of “reasonable best efforts” under Delaware law had yet to be addressed substantively in any court and was therefore uncertain.126 And Vice Chancellor Strine, the judge assigned to adjudicate ADS’s complaint, openly questioned ADS’s argument in a hearing; Vice Chancellor Strine correctly noted that the Blackstone fund was not contractually bound to provide the OCC demanded guarantee in this structure and so the grounds for any on-suit by the Blackstone subsidiaries against their parent would likely be slim.127 In the wake of Strine’s comments and his apparent favorable view of Blackstone’s arguments, ADS withdrew its complaint. At the time ADS cited Blackstone’s public statements that it was still committed to completing the transaction as the reason for this withdrawal.128

ADS’s statement upon withdrawing its suit illustrates another role of litigation in these disputes. In one strand of contract theory, forces push against litigation and towards private settlement due to the prohibitive costs of lawsuits.129 But in the private equity context, the relative costs of litigation are low compared to the billions in value at stake.130 Litigation thus served three purposes in the post-August private equity disputes. First, litigation provided a platform for each of the parties to further elaborate their private understandings of the contract and attempt to enforce them. Second, litigation served as a public forum to enforce reputational norms.131

125 Id. at 13-15. 126 A Westlaw search of the phrase “reasonable best efforts” in the Delaware case database revealed only eight cases even mentioning the term. None of them offered an interpretation of the meaning of the term. 127 ADS HEARING, supra note 121, at 39-42. 128 See Alliance Data Systems Current Report on Form 8-K, dated Feb. 8, 2008, available at http://www.sec.gov/Archives/edgar/data/1101215/000129993308000705/htm_25430.htm. Blackstone did not follow through on its public commitment. The agreement was subsequently terminated and the parties are now litigating in Delaware Chancery Court the issue of whether Blackstone is required to pay the $170 million reverse termination fee. See Phil Milford, Alliance Sues Blackstone for $170 Million Fee in Dropped Buyout, BLOOMBERG NEWS, May 31, 2008, available at http://www.bloomberg.com/apps/news?pid=20601127&refer=law&sid=a54ZIHZ3jl2U. 129 See Hill, supra note 11. 130 Given the centrality of these provisions and the low costs of drafting them, one common explanation – that the likelihood of litigation was low, was arguably not a factor in the private equity contract. But see Hill, infra note 207, at 54. 131 This is similar to the channeling theory of contracts – that parties use contracts to speak publicly about 290

8-Oct-08] Failure of Private Equity 27

For example, even though Blackstone had reverted to a formalistic view of the contract ADS was still attempting to invoke extra-contractual norms to complete the transaction. Despite ADS’s apparently weak legal case, the company asserted in court that their litigation was about enforcing “a commitment from [Blackstone] in writing to close this [transaction] as expeditiously as possible [and] to enforce the agreement.”132 This is also illustrated by a quote from the statement released by Huntsman upon being sued by Hexion, a company wholly owned by the private equity fund Apollo, to terminate their acquisition agreement in another private equity dispute:

Apollo’s recent action in filing this suit represents one of the most unethical contract breaches I have observed in fifty years of business. Leon Black and Josh Harris [partners in Apollo] should be disgraced. Our company will fight Apollo vigorously on all fronts.133

Litigation also provided a forum for the parties to obtain a quick assessment of the risks of an adverse judgment and pushed them towards resolution of their disputes. This was unique to private equity disputes and due to the overwhelmingly selection of Delaware in acquisition agreements as the forum for these disputes.134 The Delaware courts have a reputation for speed and efficient adjudication and they obliged in these disputes, acting quickly to adjudicate them. Here, the Delaware judges also pushed the parties towards resolution of their disputes through their pre-trial opinions and public statements.135 Illustratively, ADS’s withdrawal of their suit was also likely influenced by Vice Chancellor Strine’s negative comments on their case.

The ADS litigation ultimately exposed the limits of this structure in respect to its contractual terms. A second dispute involving the sale of Clear Channel’s TV station business to Providence Equity Group would highlight the more direct difficulties of forcing the shell subsidiaries to enforce and draw upon their own financing commitments. The Clear Channel TV station dispute unfolded in litigation in two jurisdictions. Wachovia Bank sued the Providence Equity shell subsidiaries in a North Carolina court to terminate its obligations under its debt commitment letter to finance the subsidiaries’ their relationship. See Lon L. Fuller, Consideration and Form, 41 COLUM. L. REV. 799, 801-3 (1941). However, while this occurred in the private equity context, it largely failed to be effective. 132 See ADS HEARING, supra note 121, at 66. 133 See Press Release of Huntsman, July 19, 2008, available at http://huntsman.com/eng/News/News/Huntsman_Rejects_Apollo_Attempt_to_Back_Out_of_Merger_Agreement/ index.cfm?PageID=7379&News_ID=5800&style=328. 134 See infra note 252. 135 See Steven M Davidoff, Battle-Testing Delaware, THE N.Y. TIMES, 291

28 82 S. Cal. L. Rev. __ [8-Oct-08 acquisition of the Clear Channel TV business.136 In addition, uncertain as to Providence Equity’s commitment to the transaction, Clear Channel sued the Providence Equity shell subsidiaries in Delaware Chancery Court to force them to litigate against Wachovia to enforce their debt commitment letter and equity commitment letter.137 Both litigations were settled shortly after their filing in an agreement that included a reduction in the purchase price.138 But a notable hearing before Vice Chancellor Strine highlighted the difficulty of forcing these shell subsidiaries to act when their parent refused to complete the transaction. Vice Chancellor Strine mused that in such circumstances a remedy of specific performance could set free the shell subsidiaries under the guidance of a special master.139 The shell subsidiaries obligation to use reasonable best efforts to obtain financing would thus be interpreted to include a search to parties for financing and funds other than to the recalcitrant private equity firms.140

. The Clear Channel TV Station case was settled before a ruling could be issued. This left open the scope and means of any specific performance remedy against shell subsidiaries in circumstances where the private equity fund parent refused to provide additional funds. The legal availability of specific performance in a cash transaction was still an uncertainty in many states, including Delaware.141 And the dual litigation in the Clear Channel TV Station dispute due to the disharmony in the forum selection clauses in the financing documents and acquisition agreement raised the real possibility that the structure could completely collapse.142 In other words, not only could the private equity firm but the financing banks could breach their financing commitment letters. This would create a situation where a acquiree would be forced to sue the shell subsidiaries and through some type of judicially ordered mechanism arrange a suit on behalf of the subsidiaries against the banks and/or private equity firms to obtain necessary financing. And the suits would have to be in different

136 See Andrew Ross Sorkin & Michael J. de la Merced, Bank’s Suit May Hurt Deal for Clear Channel Unit, THE N.Y. TIMES, Feb. 25, 2008. 137 Id. 138 See Andrew Ross Sorkin & Michael J. de la Merced, Lawsuit Is Settled Over Sale of Clear Channel’s TV Unit , THE N.Y. TIMES, Mar. 15, 2008. 139 Hearing in Clear Channel Broadcasting, Inc., et al. v. Newport Television LLC, No. 3550-VCS, at 80- 81, dated Feb. 26, 2008, available at http://lawprofessors.typepad.com/mergers/files/bd022682.ecl.pdf. 140 Id. 141 Specific performance is a remedy granted only when monetary damages are inadequate. Vice Chancellor Strine had granted specific performance of a cash acquisition in the seminal case of IBP v. Tyson. However, practitioners still speculated that this case might be unusual and therefore specific performance still inappropriate under Delaware law since the damages on a cash transaction were ascertainable in monetary terms. See Kevin Miller, URI's Request for Specific Performance: The Elephant in the Room, at http://www.deallawyers.com/blog/archives/000838.html, Dec. 4, 2007. 142 The reason for this dichotomy is that bank financing arrangements have historically been governed by New York law; and since 2005 private equity agreements largely by Delaware law. I discuss the failure to harmonize these clauses infra at 250-253 notes and accompanying text. 292

8-Oct-08] Failure of Private Equity 29 jurisdictions due to the differing forum selection clauses. While an acquiree could theoretically perform such acrobatics, the structure appeared to be collapsing under its own weight at this point.

D. The Failure of Financing

In light of this market disruption, the financing banks stood to lose substantially on each of these transactions if they completed. So, it was not surprising that evidence began to emerge in the Fall that these banks were actively agitating to terminate these transactions. In the Fall of 2007, lenders in the HD Supply and Reddy Ice private equity acquisitions had interpreted their debt commitment letters to attempt to escape financing renegotiated transactions.143 In each of these deals, the banks had asserted that the renegotiation of the terms of the transaction constituted an “adverse change” under their debt financing letter entitling it to terminate that letter.144 The Reddy Ice transaction ultimately was terminated through payment of the reverse termination fee by the private equity firm while the banks’ position forced a renegotiation of the HD Supply transaction.145 In addition, in the Acxiom transaction two of the financing banks had agreed to reimburse the private equity firms for part of the reverse termination fee. Presumably they did this so as to incentivize the private equity firms to terminate the transaction.146 In all these instances, these disputes and arrangements remained private and did not result in any litigation or public dispute.147 However, the stress in the banking/private equity relationship broke to the surface in the litigation surrounding the Clear Channel TV station transaction. In that transaction, Wachovia sued and asserted that a renegotiation of the purchase price constituted an “adverse change” under its debt financing letter.148 The litigation was settled by Wachovia before any decision could be rendered, but in the settlement Wachovia achieved a reduction of its financing commitment.149

The most significant failure in the private equity/lender relationship occurred in the litigation over the $26 billion purchase of Clear Channel by

143 See Henny Sender, et al., Home Depot Hit As Credit Crunch Squeezes Deals, THE WALL ST. J., Aug. 27 2007; David Carey & Vipal Monga, Morgan Stanley hints at pull from Reddy, THEDEAL.COM, Sept. 13, 2007. 144 Id. 145 See Reddy Ice Current Report on Form 8-K, dated Feb 1, 2007, available at http://www.sec.gov/Archives/edgar/data/1268984/000110465908006289/a08-4392_18k.htm; Sender, supra note 143. 146 See Andrew Ross Sorkin, Acxiom Shows Breaking Up Is Costly, THE N.Y. TIMES, Oct. 10, 2007. 147 See Vipal Monga, When friends fall out, THEDEAL NEWSWEEKLY, Sept. 21, 2007. 148 See Andrew Ross Sorkin & Michael J. de la Merced, Bank Sues Acquirer of Clear Channel Unit, Jeopardizing Deal, THE N.Y. TIMES, Feb. 25, 2008, C2. 149 See Andrew Ross Sorkin & Michael J. de la Merced, Lawsuit Is Settled Over Sale Of Clear Channel's TV Unit, THE N.Y. TIMES, March 15, 2008, at C3. 293

30 82 S. Cal. L. Rev. __ [8-Oct-08

Bain Capital and Thomas H. Lee. On March 26, 2008, Bain and Thomas H. Lee sued their financing banks in New York Supreme Court.150 In their complaint, the private equity firms alleged that the banks had breached their commitment letters by demanding unreasonable terms that were onerous and unusual in the final negotiated credit documentation in an attempt to terminate their obligations under the letter, something the banks were incentivized to do since they would incur an estimated $2.6 billion collective loss if they were forced to fund.151 Market conditions had changed that substantially from the time the banks had initially agreed the financing terms in their debt commitment letter.

The private equity firms asserted that this language violated the requirement in the debt commitment letter that the final negotiated credit agreements “contain the terms and conditions set forth in this Commitment Letter and shall be customary for affiliates of the Sponsors.”152 This “sponsor precedent” clause was considered quite friendly to the private equity firms since it narrowed the scope of precedent to be referenced to that in which the sponsor, another name for the private equity firm, had previously agreed to and so ensured that the private equity firms would obtain a favorable deal.

The banks countered that the 71 page debt commitment letter was unenforceable because it contained too many open terms – only with final documentation would the contract be sufficiently complete to be enforceable.153 In addition, the banks argued that a specific performance remedy was unavailable to the private equity firms – they could only collect money damages.154 In this case the private equity firms would be subject to a Catch-22 since the banks also argued that money damages would be unavailable under the terms of the debt commitment letter which they asserted barred monetary damages and the acquisition agreement which in any event limited the private equity firms’ monetary damages to $500 million.155 Finally, the banks argued that the sponsor precedent language cited by the private equity firms was essentially meaningless since, due to

150 The banks were Citigroup, Deutsche Bank AG, Credit Suisse Group, Morgan Stanley, Royal Bank of Scotland Group and Wachovia Corp. See Complaint in BT Triple Crown Merger Co., Inc., et al. v. Citigroup Global Markets Inc., dated Mar 25, 2008, available at http://graphics8.nytimes.com/images/blogs/dealbook/CCU_NY_lawsuit.pdf. 151 Id. at ¶¶ 6-8. 152 Id. at ¶ 16. 153 See Memorandum of Law in Support of Defendants’ Motion for Summary Judgment, BT Triple Crown Merger Co., Inc., et al. v. Citigroup Global Markets Inc., at 21-25, dated Apr 10, 2008, available at http://graphics8.nytimes.com/images/blogs/dealbook/CCUdismissmotion.pdf 154 Id. at 16-21. 155 Id. at 10-16. 294

8-Oct-08] Failure of Private Equity 31 the state of the market, no transaction was customary or similar.156

This dispute settled with a renegotiated price, an increase in the equity commitment by the private equity firms and the interest rate they were to pay on their bank debt and a decrease in the amount of debt financing provided by the banks.157 In this tissue of legal points was a fundamental truth – the banks in the Clear Channel case argued that the debt commitment letters were an option that they could refuse to fund on with at most a penalty equivalent to private equity’s reverse termination fee.158 The wave of litigation and dispute which had enveloped private equity now had placed into doubt the entirety of the private equity structure: the reputational forces pushing private equity firms to complete acquisitions, the mechanics of enforcing the specific performance form of the private equity structure, and the enforceability of debt commitment letters.

III. NEGOTIATING THE STRUCTURE OF PRIVATE EQUITY

The collapse of the private equity structure brought recrimination and criticism among and upon the participants in private equity buy-outs – financial institutions, private equity firms, acquirees and lawyers in particular. The criticisms were broad-based and were directed at the following alleged lapses:159

Failure of Certainty. Acquirees were condemned for agreeing to optional takeover structures. Critics claimed that acquiree boards were wrong to agree to reverse termination fee provisions thereby forgoing deal completion certainty equivalent to that in strategic transactions.

Failure of MAC Clauses. In the first wave of post-August 2007 private equity terminations, MAC clauses functioned in a manner different than acquirees and their attorneys likely intended. The existence of these clauses permitted acquirers to invoke them to provide reputational cover for otherwise exercising reverse termination fee provisions. Moreover, the repeated invocation of MACs and the success of acquirers in terminating their transactions lessened reputational constraints allowing for the exercise of the reverse termination fee provision without an accompanying MAC claim. The net effect was to deprive acquirees of the full reputational

156 See Motion on Expert Testimony, BT Triple Crown Merger Co., Inc., et al. v. Citigroup Global Markets Inc., at [●], dated [●]. 157 See Clear Channel Current Report on Form 8-K, dated May 15, 2008, available at http://www.sec.gov/Archives/edgar/data/739708/000095013408009540/d56929e8vk.htm. 158 See Steven M. Davidoff, A Clear Channel Scorecard, N.Y. TIMES DEALBOOK, Apr 16, 2008, available at http://dealbook.blogs.nytimes.com/2008/04/16/clear-channel-scorecard/#more-22511. 159 See Steven H. Goldberg, Deals gone bad, THEDEAL.COM, Aug. 19, 2008. 295

32 82 S. Cal. L. Rev. __ [8-Oct-08 closing force they had likely counted upon when negotiating their agreements.160

Failure of Specific Performance. The attorneys negotiating the specific performance mechanism of the reverse termination fee structure did so in the belief that it was an enhanced form of the reverse termination fee structure.161 But in negotiating this structure, these attorneys failed to fully account for the problems with enforcing this arrangement through shell subsidiaries, the lack of judicial precedent governing enforcement of this mechanism, the difficulty of forcing shell subsidiaries to enforce debt and equity commitment letters with differing choice of law and forum clauses and the uncertainty as to whether specific performance could even be awarded under Delaware law.162

Failure of “Best Efforts”. A similar failure appeared to occur with respect to “best efforts” clauses negotiated in the pre-August 2007 private equity structure. Attorneys often negotiated that the shell subsidiary would use its reasonable best efforts to complete this transaction, particularly to obtain regulatory approvals or replacement financing.163 But the meaning of reasonable best efforts is still uncertain under Delaware law.164 Moreover, the private equity parties subject to this clause were shell subsidiaries with little or no assets.165 The implementation of this clause was thus dependent upon the voluntary cooperation of the private equity funds themselves. If regulators or the private equity firm refused to cooperate, it would result in substantial difficulties and legal uncertainties in forcing the private equity shells to close. This is exactly what happened in the ADS litigation.166

Failure of Drafting. Finally, the private equity agreements themselves contained ambiguities and conflicting provisions. In some cases like the URI-Cerberus litigation it appeared that these defects were attorney drafting mistakes.

Shareholders and commentators attributed these lapses to a number of

160 Moreover, the parties often negotiated two-tiered reverse termination fees, a higher fee if the acquirer simply breached its obligations but a lower one if financing was unavailable. In at least one significant post- August transaction, the failed buy-out of Acxiom, the private equity acquirer invoked a MAC provisions and the similar prospect of lenders doing the same to force a settlement at the lower reverse termination fee. See Interview, supra note 88 . 161 See supra notes 72-74 and accompanying text. See also David Leinwand and Victor Goldfeld, Stresses on the New LBO Deal Architecture: United Rentals Goes to Court (Cleary Gottleib Newsletter Jan. 2008). 162 See also Kevin Miller, The ConEd Decision - One Year Later: Significant Implications for Public Company Mergers Appear Largely Ignored, 10 THE M&A LAWYER 9, Oct. 2006. 163 See supra notes 119-120 and accompanying text. 164 See supra notes 126-128 and accompanying text. 165 See supra notes 122-124 and accompanying text. 166 See supra Part III.C. 296

8-Oct-08] Failure of Private Equity 33 factors, including lax oversight, a failure to recognize and appreciate the risks of the private equity structure, and management’s oft conflicted role in these buy-outs which led them to overlook the problems with this structure.167 The lawyers for these acquirees were also criticized for negotiating these optional structures and accused of failing to properly advise acquirees of the inherent risk.168 But this criticism belies a much more nuanced picture in the negotiation of these structures. Upon examination, acquirees and their lawyers did appear to make a number of miscalculations but their failures were premised upon other fault lines.

A. Optionality and the Private Equity Structure

a. Structuring the reverse termination fee

Criticism of the private equity structure was principally directed at the optionality and resulting uncertainty it created. In its purest form, the reverse termination fee structure created an option.169 The private equity firm had the discretion to exercise this option, and if the firm did so, it could terminate the transaction and pay the reverse termination fee. A private equity acquirer could thus assess the benefits of the transaction before completion and decide whether it was more economical to complete the transaction or otherwise pay the reverse termination fee and terminate the acquisition agreement.

This option was not calculated according to any option pricing method. Nor did it appear to be calculated by reference to the damage incurred by a acquiree in the event it was exercised by the private equity firm in order to terminate the transaction. The amount ultimately paid also did not deter acquirers from exercising it in many instances. Rather, the amount of the reverse termination fee was set normatively by reference to the termination fee typically paid by acquirees, approximately three percent of the transaction value.170 Setting the fee at 3% for acquirer and acquiree made for a symmetrical penalty.171

167 See Davidoff, supra note 9. 168 See, e.g., Nowicki, supra note 91, at 3. 169 See Davidoff, supra note 75. 170 From 2005-2007, the average size of a reverse termination fee was [•] percent of the transaction value. MergerMetrics Database. See PETER V. LETSOU, CASES AND MATERIALS ON CORPORATE MERGERS AND ACQUISITIONS 619 (2006) (“[Break-up] fees are almost always set at approximately 3 percent of the transaction value.” ) . It largely remained at this amount from transaction to transaction without variance. In my dataset of [174] private equity transactions, I calculated that during the period from 2004-2007 the reverse termination fee ranged from []-[]% of the transaction value with a mean percentage value of [] and a median of []. Author calculations based on data from the Merger Metrics Database. 171 When asked how the reverse termination was set, attorneys confirmed that setting the reverse termination fee at 3% was an adoption of the norm for acquiree termination fees. Interview D. 297

34 82 S. Cal. L. Rev. __ [8-Oct-08

The fact that each of these penalties existed for different reasons and worked differently shows the strength of the norm in operation. The reverse termination fee provided a liquidated damages remedy equivalent to the termination fee paid by acquirees.172 This latter fee was capped by Delaware case-law and was designed to deter competing bids and compensate bidders for the costs associated with making a trumped offer.173 But the same principles did not apply in the reverse termination fee context. The fee in a number of prominent instances did not deter exercise of the option and in hindsight the amount appeared to under-compensate acquirees for the losses incurred by the acquiree company and its shareholders.174 Evidence of this came from the post-termination share trading prices of acquirees against whom these provisions were invoked. In the months after the exercise of this provision, the share prices of these companies traded significantly below the pre-offer price.175

Despite the seeming miscalculation of the reverse termination fee, market participants interviewed for this study almost all asserted that the optionality of the reverse termination fee structure was well known prior to August 2007 among lawyers and transaction participants.176 The testimony in the United Rentals/Cerberus dispute supports these assertions. In that dispute, the parties disagreed over the language of their termination – whether it was a specific performance reverse termination fee structure or a pure reverse termination fee structure. The description in the judicial opinion of the parties bargaining, though, reveals that they did appreciate the choice:

Throughout the course of negotiation of the Acquisition agreement, [United Rentals (“URI”)] contends that it communicated to [Cerberus Parent Shell Subsidiary’s (“RAM”)] principal attorney contract negotiator, Peter Ehrenberg of Lowenstein Sandler PC, that URI wanted to restrict RAM’s ability to breach the Acquisition agreement and unilaterally refuse to close the transaction. URI further maintains that URI’s counsel, Eric Swedenburg of Simpson Thacher & Bartlett LP, made clear to Ehrenberg that it was very important to URI that there be

172 See Brazen v. Bell Atlantic Co., 695 A.2d 43 (1997) (applying a liquidated damages analysis to a termination fee in a strategic merger to determine the validity of a termination fee provision). 173 For a discussion of termination fees in the context of change of control transaction. See generally In re Toys "R” Us, Inc. Shareholder Litigation, 877 A.2d 975 (Del. Ch. 2005). 174 See Vipal Monga, Turning the tide, THE DEAL, Aug. 29, 2008 (“James Woolery, a partner in Cravath, Swaine & Moore LLP's M&A practice [asserts] that reverse termination fees, typically set at around 3%, often understate[d] the risk of deal failure.”) 175 See, e.g., Nowicki, supra note 91, at 7. 176 Interviews with D, E. 298

8-Oct-08] Failure of Private Equity 35

“deal certainty” so that RAM could not simply refuse to close if debt financing was available. On the other side of the negotiation table, the RAM entities argue that Ehrenberg consistently communicated that Cerberus had a $100 million walkway right and that URI knowingly relinquished its right to specific performance under the Acquisition agreement.177

The reverse termination fee structure also provided more closing certainty than the structure it supplanted. In the pre-2005 structure, the structure was wholly optional – the acquiree entered into an agreement with thinly capitalized shell subsidiaries and the agreement itself contained a financing condition.178 If the subsidiaries refused to perform or otherwise financing failed the acquiree was left with no compensation or recourse to the private equity firms except through a veil piercing or other creative litigation argument.179 The reverse termination fee structure reduced optionality in the structure by imposing a penalty upon private equity firms for refusing to complete transactions, and in its specific performance form purported to ensure that the acquisition would occur if all of the conditions to completion were fulfilled and financing was available.

b. Explaining the Reverse Termination Fee

Attorneys at the major private equity law firms on some level thus appeared to be aware of the optionality embedded in these structures. They also appeared to negotiate among these structures to select more or less optional structures. However, the question remains why acquirees advised by their lawyers assumed this transaction risk? Furthermore, why did lawyers set the size of the reverse termination fee through an inapposite norm rather than varying it depending upon the circumstances of the transaction? Within these questions is another one concerning whether acquirees and their attorneys fully comprehended or were advised of the risks inherent in these structures, and if so why they still agreed to these contracts.

Attorneys interviewed for this Article generally stated that they were aware of the optionality inherent in this structure, and asserted that they had

177 United Rentals, Inc. v. RAM Holdings, Inc., et al., 937 A.2d 810, 818-819 (Del. Ch. 2007). 178 See Sorkin & Swedenburg, supra note 50, at 3. 179 The arguments in the ADS litigation – that a parent had some sort of responsibility to a subsidiary represent this type of creative lawyering. Unfortunately, for acquirees though, a veil piercing argument was often unavailable since the private equity guarantees typically contained strong no-recourse language. See, e.g., Steven M. Davidoff, The Private Equity Lawsuit of the Year, N.Y. TIMES DEALBOOK, Feb. 27, 2007, available at http://dealbook.blogs.nytimes.com/2008/02/27/the-private-equity-lawsuit-of-the-year/ (discussing the language of the no recourse guarantee in the acquisition of 56 television stations by Providence Equity from Clear Channel). 299

36 82 S. Cal. L. Rev. __ [8-Oct-08 informed their boards of this optionality.180 In both instances though participants repeatedly asserted that the negotiated contractual terms were not the only basis for agreeing to this transaction structure. Rather attorneys for acquirees in these scenarios relied upon the interaction of contractual terms and extra-contractual forces and understandings to complete the private equity contract. The parties bargained to create incentives within the contract to force a closing primarily the reverse termination fee. But attorneys for acquirees also relied upon reputational norms and conventions as well as other external forces to fill perceived “gaps” in the contract.181 Again, the judicial description of the bargaining between URI and Cerberus provide support for these assertions:

Swedenburg explained that URI would require a reverse break-up fee of sufficient size to ensure that it would be “scary” and “painful” for the RAM Entities to walk away from the transaction. Swedenburg noted that URI was not content merely to rely upon the reputational fallout that would ensue if the RAM entities and their affiliates failed to close.182

The private equity contract created a monetary penalty that was deemed to be completed in part by the reputational norm pressuring the private equity firm to complete the transaction.183 In this multiplayer game, private equity was a repeat player. As such, it was assumed that the reputational incentive to close would keep them from exercising the reverse termination fee option and being perceived as reneging on their deals.184 The penalty for failure to follow this norm would be a higher price paid in future transactions to compensate acquirees for this failure and increased risk as well as any other public approbation for this action. This penalty together with the required reverse termination fee was presumed to be adequate to prevent exercise of this option.

Moreover, the acquisition contract and its negotiation served as a bonding mechanism enhancing these norms and constraints. The negotiation process not only established the legal parameters of their agreement but in the discourse of the parties also established a relationship to sustain their

180 Interviews with F, G. 181 See David Charny, Nonlegal Sanctions in Commercial Relationships, 104 HARV. L. REV. 373, 392-94 (1990) (discussing types of nonlegal sanctions as penalties in commercial transactions) 182 United Rentals, 937 A.2d at 825. 183 See Ronald H. Coase, The Nature of The Firm: Influence, 4 J. L. ECON. & ORG. 33, 44 (1988) (arguing that opportunistic behavior is typically checked by reputational constraints). But see Oliver E. Williamson, Transaction-Cost Economics: The Governance of Contractual Relations, 22 J.L. & ECON. 233, 241 (1979). 184 Cf. Rubén Kraiem, Leaving Money on the Table: Contract Practice in a Low-Trust Environment, 42 COLUM. J. TRANSNAT'L L. 715, 736 n.31 (2004). 300

8-Oct-08] Failure of Private Equity 37 transaction. Here, the conversations outside the four corners of the contract solidified the acquiree’s willingness to rely on these extra-contractual factors to complete its acquisition contract. Again, the description in the URI judicial opinion of the parties negotiation provides support for this observation:

On July 21, 2007, in a conversation between Mayer, Kochman, and McNeal, Mayer indicated that he thought RAM was purchasing an “option,” Kochman strongly disagreed with the contention. Kochman testified about that conversation:

A. He said, you know, “Gee, that’s a lot of money. You know, I view this as an option. And my LPs would be very unhappy if I, you know, burnt that 100 million plus dollars.” And I was taken aback by that.

Q. And what did you say to him?

A. I said, “You know, that's crazy. That’s a nonstarter. This is not an option. That’s something I would never take back to the board.” And I laid into him fairly good and said that this is a board that has concerns about your ability to consummate transactions. They see what’s going on with Chrysler. They don’t view you in the same breaths as KKR or Blackstone. And, you know, it’s a complete nonstarter.

Q. Did he respond to that?

A. He backed away. He said, “Time out. You know, I’m 100 percent committed to this transaction. I’m going to take you—I’m going to tell you right now that the debt financing and the commitment letters we have in hand are designed exactly for difficult markets. We’ll get this deal done. I’m going to take you under the tent.”185

A third force, the economic incentives of private equity firms, was also cited by attorneys as completing the private equity structure and justifying

185 United Rentals, 937 A.2d at 826-27. Another example is from a Wall Street Journal article reporting on the agreement negotiations between Huntsman and Hexion. According to the article, Jon Huntsman, the Chairman of Huntsman’s board of directors “states he was suspicious of Apollo's willingness to close the deal at that price. ‘It was important to me that I have Black and Harris shake hands with them at our Deer Valley home,’ he [said]. ‘I wanted them to look [the senators] in the eye and tell them it was a done deal.’ Mr. Huntsman says that Mr. Black assured the group he had a ‘100% commitment to close the deal,’ and that Mr. Harris assured him repeatedly the deal was ‘solid.’” See Susan Pullman & Peter Lattman, As Buyout Bust Turns Bitter, A Major Deal Lands in Court, THE WALL ST. J., Sep. 9, 2008. 301

38 82 S. Cal. L. Rev. __ [8-Oct-08 its contractual optionality.186 Private equity is in the business of acquiring companies. When they enter into an agreement to do so, this force and the momentum it creates pushes the private equity firm to complete the acquisition. In other words, private equity firms, initially enter into the transaction because of a desire to acquire the company. The transactional costs incurred to first enter into this agreement and the commitment it represents also pressure private equity firms to complete their transactions.187

The particular reputational and economic forces upon private equity provide a strong explanation for why a different, more certain structure has historically been utilized for strategic transactions – acquisitions made by a functioning company rather than a private equity firm. Strategic transactions lack the optional nature of private equity acquisitions. In strategic transactions the structural norm is to eschew financing conditions and reverse termination fee structures.188 Instead acquiree companies obligate the acquirer to specifically perform the acquisition in case of a breach of the acquisition agreement.189 Moreover, unlike the private equity context, this agreement is secured by the assets of the acquirer.190

The traditional reason offered for this dichotomy in structure is a financing rationale. In a strategic transaction an acquirer has assets to secure its obligations and is not dependent upon the vicissitudes of the financing market to complete its transaction. But, many strategic acquirers employ substantial leverage to effect acquisitions while private equity funds do have assets – the contractual agreements of their limited partners to fund. Accordingly, a stronger explanation is that a strategic acquirer is not in the business that private equity firms are – acquiring companies. If a strategic acquirer reneges on its acquisition agreement the reputational loss is likely to be less since it is not consistently in the acquisition market, and stronger contractual protections are justified.191

Thus, acquiree attorneys relied upon these forces to negotiate the

186 Interviews with K, L. 187 Cf, John C. Coates IV & Guhan Subramanian, A Buy-Side Model of M&A Lockups: Theory and Evidence , 53 STAN. L. REV. 307 (2000) (positing that bidders may remain committed to pre-agreed acquisitions due to transactional switching costs since “[o]nce a given strategy has been devised and communicated, remaining committed to that strategy will be ‘free,’ and optimal, as long as the net gain from switching to a new strategy is less than the cost of communicating the new strategy.”) 188 See Steven M. Davidoff, Wrigley and the Future of M&A, N.Y. TIMES DEALBOOK, May 1, 2008, available at http://dealbook.blogs.nytimes.com/2008/05/01/wrigley-and-the-future-of-ma/. 189 See Yair Jason Listokin, The Empirical Case for Specific Performance: Evidence from the Tyson-IBP Litigation (Mar 1, 2005), available at SSRN: http://ssrn.com/abstract=679874 (last visited Feb 10, 2008). 190 See WILLIAM J. CARNEY, MERGERS AND ACQUISITIONS: CASES AND MATERIALS (2nd Ed. 2007). 191 In the wake of the credit crunch, reverse termination fee clauses have crept into strategic deals with financing risk. It remains to be seen whether these features will persist when the markets eventually stabilize. 302

8-Oct-08] Failure of Private Equity 39

structure of private equity. Here, attorneys for acquirees asserted that this structure was the best deal available, that private equity firms were unwilling to agree to more certain, different terms.192 Boards advised by their attorneys made the conscious decision to accept these transaction premiums in exchange for accepting this deal uncertainty.193 Moreover, these attorneys also highlighted competitive bidding situations wherein private equity firms did indeed agree to strategic type completion structures.194 They cited the existence of these provisions in this context to justify their agreement as merely a reflection of the competitive environment, resulting bargaining strength of the parties and consequent agreement achieved.

The notion that parties sometimes leave contractual terms incomplete or vague in order to reach a contract and otherwise avoid a dispute is well discussed in the contracts literature.195 In that scenario, parties prefer to leave any further disputes to the courts, and any incentives to ex post facto negotiation in order to avoid such litigation, rather than fail to reach an agreement altogether.196 In the private equity model, it appears that the arguments of attorneys that these contractual terms were the best available are a species of this theory. Acquirees advised by their attorneys concluded that the contractual terms themselves were enhanced by the normative and economic incentives and constraints existent. In such circumstance they were willing to agree to a reverse termination fee structure and negotiate away contractual protections due to reliance on these factors. The bargain struck in contract was thus perceived by acquirees as an acceptable one due to reliance upon extrinsic forces and the consequent perceived unavailability of a better bargain.

192 See Monga, supra note 174 (“Instead, at the time the deal was struck, the directors saw the fee as a basic price of doing business, given that none of the potential acquirers of the company would sign without the conditional put. In addition, the board felt the chances the transaction would fall apart were slim.”) 193 Interview with N. 194 See, e.g., Davidoff & Baiardi, supra note 78 (detailing the competitive bidding by Lone Star Funds for Accredited Home Lenders and the specific performance contract which arose). Interestingly, at no point did anyone state that this certainty or uncertainty affected deal premiums. There is evidence that post August 2007 there has been some attempts to relate these two at least in public descriptions of the transaction negotiation. See, e.g., Schedule 14D-9 filed by Golden Telecom, at 23-25, dated Jan. 18, 2008, available at http://www.sec.gov/Archives/edgar/data/1089874/000119312508008895/dsc14d9.htm. 195 See Bernheim & Whinston, supra note 11. It is also a publicly commented upon phenomenon. See Ben Hallman, Our Dealmakers of the Year Section Spotlights Achievement in Corporate Work but The Lawyering in the United Rentals Buyout, on the Other Hand, Didn’t Lead to Glory – Just to Court, 30 AM. LAWYER 4, Apr. 2008 (“Lawyers familiar with the deal say they believe the United Rentals case offers a glimpse into a little- noticed but common practice: Deal lawyers often agree to contracts with ambiguous language for the sake of compromise.”). This practice is one similar to parties otherwise agreeing to leave a contractual term gap in a contract. See Lisa D. Bernstein, Social Norms and Default Rules Analysis, 3 S. CAL. INTERDISC. L. J. 59, 66 n. 34 (1993-94); George Geis, An Embedded Options Theory of Indefinite Contracts, 90 MINN. L. REV. 1664 (2006). 196 Additionally, parties calculate the likelihood of those contingencies is sufficiently remote that it’s not worth the added negotiating cost, or the increased risk of failure to complete a deal. 303

40 82 S. Cal. L. Rev. __ [8-Oct-08

B. Other Facets of the Private Equity Structure

In this light, acquirees’ and their agent attorneys’ agreement to a reverse termination fee structure was likely justified to themselves. It was seen as the only available bargain agreeable to private equity firms and otherwise completed by externalized forces. Even so, attorneys for acquirees did not then fix the other known embedded flaws in the private equity structure. Instead, lawyers again appeared to rely on norms and external understandings and constraints to complete the structure blessing the final deal as the best available bargain.

In the negotiation of the reverse termination fee the contract served as a bonding mechanism among the parties and its negotiation as a vehicle to create a relationship and to signal its strength.197 This practice carried through to other parts of the private equity contract negotiation. So for example, in selecting the efforts to be required of the shell subsidiaries to obtain regulatory approvals or alternative financing, lawyers typically negotiated among “reasonable best efforts” and “best efforts” with the latter considered a stronger contractual obligation.198 In reality the choice of “best efforts” over “reasonable best efforts” did not create a significantly incremental requirement on the acquirer because in either case the acquirer was a shell subsidiary with limited assets. The acquirers efforts were therefore per se limited. Moreover, lawyers often had a mistaken understanding of what best efforts or even reasonable best efforts meant and the difference between the two.199 This misunderstanding was often enhanced by the lack of case-law on the meaning of “reasonable best efforts”.200

Despite the confusion about the meaning of “best efforts” and its variants, the parties could signal their commitment to the transaction by selecting the higher or lower standard. It did not matter for contractual purposes, but it permitted the parties to gauge their commitment and allowed for future bargaining after the execution of the contract over the meaning of the parties commitment. Moreover, the negotiation over the selection among “efforts” would orally establish the level of conduct

197 See Mark C. Suchman, The Contract as Social Artifact, 37 LAW & SOC’Y REV. 91, 111 (2003). (“contract rituals provide symbolic reassurance that the parties are entering into a predictable, controllable, and mutual relationship within a social order composed of voluntary arm’s-length exchanges between equally endowed strangers”). 198 See Kenneth A. Adams, Understanding“Best Efforts” And Its Variants (Including Drafting Recommendations), THE PRACTICAL LAWYER, 12-13, Aug 2004 (asserting that lawyers’ conventional understandings of best efforts and its variants is at odds with the case law) 199 Id. 200 See supra note 126 and accompanying text. 304

8-Oct-08] Failure of Private Equity 41 expected of the parties under the agreement prior to the contract’s execution. I believe this is why the lawyers in the ADS transaction inexplicably negotiated a contract which required the Blackstone shell subsidiary to use reasonable efforts to complete the transaction when the real party whose action would be required, the Blackstone fund itself, was not so obligated. ADS and its lawyers through discussion and negotiation of the regulatory efforts covenant likely satisfied themselves over what Blackstone was and was not willing to do. The need to document these actions in the contract was thus seen as an appropriate risk to forgo. Instead, the dialogue established that Blackstone could be trusted to engage in the necessary conduct to complete the transaction. Unfortunately for ADS, the assumption turned out wrong.201

Similarly, the choice of a specific performance model over a pure reverse termination fee was a means for the parties to signal higher commitment to the transaction. This was true even though there were doubts about the effectiveness of the specific performance form. In other words, lawyers had their extra-legal interpretation of the contract which served to create an understanding of the intentions of the parties at times at odds with the contract itself. The negotiation in the URI/Cerberus acquisition amply illustrates this. Eric Swedenburg, attorney at Simpson Thacher for URI, started the negotiations by insisting that this would not be an optional deal, but rather one modeled on the specific performance form.202 The failure of URI to obtain this form of deal structure was likely a signal of URI’s commitment, one which may have been ignored by URI due to the outside conversations between Cerberus and URI in which Cerberus committed itself to completing the transaction.

This extra-legal discourse, the signals it sent and the understandings it led to also explains the existence of drafting error, contract gaps and the agreement of acquirers and acquirees to flawed and potentially flawed contract terms. Attorneys for acquirees relied on extra-legal contractual understandings and forces to ensure that the contract completed. In this circumstance, they did not push to flush out all of the errors in the contract or otherwise to make a more complete contract. Instead they relied upon these forces to forgo fixing the contract.203 In other words, the attorneys felt

201 Another example came in the Penn National Gaming transaction. There the lawyers did not contractually bind the acquiring funds to cooperate with needed regulatory approvals despite the clear knowledge that such cooperation would be required. 202 See United Rentals, Inc. v. RAM Holdings, Inc., et al., 937 A.2d 810, [●] (Del. Ch. 2007). 203 The time constraints of these transactions also often permitted attorneys to justify these “mistakes”. See, e.g., The Partner Survey: The View From the Top, AM. LAW., Jun 1999, at 79, 82 (surveying partners at the top 100 grossing law firms and finding that 52% who responded and worked greater than a sixty hour work week “worked so fast they made mistakes”). 305

42 82 S. Cal. L. Rev. __ [8-Oct-08

they could safely ignore these failings since other extra-legal factors provided them confidence that they would not matter.204 Moreover, as the United Rentals case showed, the extra-contractual dialogue between attorneys and parties reinforced their own interpretation of these clauses even in the case of erroneous or ambiguous drafting -- interpretations they likely felt that their own created relationships would work to enforce.205 The contract would sometimes be ignored for the private understandings of the parties.

It is here where the importance of the contract comes into play and explains why parties did not simply negotiate a letter of intent or a less formal agreement. First, the contract provided a blueprint for the parties to follow.206 Second, the contract negotiation itself created a mechanism through which the parties could bond and gain an understanding of each others’ commitment and intentions. Third, the contract created its own mechanics pushing towards a closing by establishing a formal relationship.207 But the contract terms were also negotiated to be enforced and altered depending upon the relational bonding of the parties. Even in the end-game the parties attempted to enforce the private understandings of the parties and reputational norms.208 This was ably illustrated in both the ADS and Huntsman litigations.

Finally, the flaws and uncertainty, deliberate and otherwise, in the private equity structure provided a means for further negotiation when the relationship did indeed begin to collapse. They ensured that the parties would likely continue their dialogue before and after any litigation was

204 The lawyer in this instance was the primary translator of this granular language into broader concepts for the client to approve and understand since they were the primary negotiator for their principal. But he or she in certain circumstances also appeared to be acting autonomously making granular choices with little oversight or approval by the client itself. See also Coates, IV, supra note 19, at 1303 (2001) (examining the adoption of structural defenses by companies at the IPO stage and finding that “Corporate lawyers, at least at the IPO stage, appear to be working relatively free of market, ethical, or other constraints, and many appear to be making choices, and mistakes, without determining whether such choices are in the long-term interests of their clients (that is, pre-IPO owner-managers).”); Donald C. Langevoort & Robert K. Rasmussen, Skewing the Results: The Role of Lawyers in Transmitting Legal Rules, 5 S. CAL. INTERDISC. L.J. 375 (1997) (positing a theoretical construct that lawyers may tend to overstate legal advice). See generally George M. Cohen, When Law and Economics Met Professional Responsibility, 67 Fordham L. Rev. 273 (1998). 205 As Professor Claire Hill has observed, the institutional structure of law firms often produces lawyers who are incentivized to produce “good enough” drafting safe in the assumption that the specter of litigation and other transaction costs will prevent any errors from discovery or exploitation. See Hill, supra note 19, at [•]. 206 See Suchman, supra note 199, at 110. Cite also Bernstein. 207 See Claire A. Hill, A Comment on Language and Norms in Complex Business Contracting, 77 CHI. KENT LAW REV. 29, 47-48 (2001) (noting that despite common “technical” failures to meet closing conditions in contracts, a party entitled to assert a termination right in such circumstance will instead close often after a renegotiation). 208 See also Andrew Willis, BCE deal on track for August closing: Teachers in sound negotiating position, battered banks honour commitment to buyout, GLOBE & MAIL, Jun 23, 2008 (reporting on negotiations over the BCE buy-out and reporting that ‘[f]or all the noise, it's worth noting that Citigroup and Deutsche Bank [the lead lenders] have never failed to honour their commitment to a buyout,’ said a source working the lenders.”) 306

8-Oct-08] Failure of Private Equity 43 commenced.209 This would force them to work towards completion of the transaction and resolution of their disputes. Moreover, the unique nature of litigation in Delaware – which provided a forum for speedy and public dispute resolution – worked to reinforce the reputational norms towards resolution of disputes and air the parties’ private understanding of the private equity structure and contract.210

In sum, the reverse termination fee was not the only result of extra- legal considerations. Attorneys negotiated in reliance upon extra-legal norms, signals and understandings to complete the private equity structure and establish a relationship to engender completion of the transaction even upon litigation. An example again comes from the failed United Rentals transaction. In that transaction, Simpson Thacher, the attorneys for the acquiree likely negotiated an ambiguous document, engaged in relational bonding in negotiating the contract and pushed for a higher reverse termination fee to offset the failure to negotiate a more certain closing structure with the four corners of the contract itself.211 The contract was not just negotiated for its bonding, signaling, and terminology – but as a backstop in the final end-game of litigation.212 In such a circumstance, the created relationship and understandings could engender further bargaining before litigation and a platform to enforce reputational norms and any other understanding during such litigation.

IV. THE PATH DEPENDENCY OF THE PRIVATE EQUITY STRUCTURE

The preceding Part discussed how acquirees and their attorneys faced with the structure of private equity negotiated and justified it. Nonetheless, the question remains how the private equity structure came to be and why these forces worked so strongly to prevent change. The private equity industry is a highly sophisticated one with the capability to renegotiate the structure at any time. The primary impediments to fixing these problems – many of which were known -- were two-fold: time and obtaining the agreement of the private equity parties. The transaction costs to fix these flaws were low, and lawyers for acquiree companies were incentivized to bargain for certainty through success fees in addition to their regular fees. For example, counsel representing United Rentals, Simpson Thacher, was

209 See Hill, supra note 11, at [●]. 210 See supra notes and accompanying text. 211 See supra at Part III.A.2. 212 In fact, this is what likely led Cerberus to make an offer to settle the dispute on the eve of trial, an offer which United Rentals rejected. See Cerberus Offers to Redo United Rentals Deal, THE N.Y. TIMES, Nov 16, 2007. 307

44 82 S. Cal. L. Rev. __ [8-Oct-08

to be paid a $5 million bonus if the transaction completed.213 This was on top of their multi-million dollar legal fee.214 So, the question remains why attorneys were so reliant on these extra-legal considerations and, to the extent they were aware of problems, why they didn’t fix these flaws and ambiguities? Instead they relied, or perhaps over-relied, on extra-legal forces and constraints.

The answer comes from another force: path dependency. The private equity structure, the boiler-plate of private equity, appears to be one largely set by path dependencies.215 But, the structure’s stickiness is not primarily a result of network effects and excess transactional switching costs – the most commonly cited reasons for path dependency in boiler plate.216 Rather, it is an agency problem. Private equity lawyers are not incentivized to rethink and renegotiate the boilerplate of private equity, but rather rely upon pre- existing, “good enough” structures. Dependencies in these complex contracts are therefore strong and the structure is particularly resistant to externalized shock.217 Accordingly, change when it does happen trends towards the incremental and is likely to be piecemeal upon the existing structure creating further ambiguity and flaw.218

A. The Nature of Path Dependency in the Private Equity Structure

In the past twenty years there have been only two significant shifts in the structure of private equity. In the 1990s, equity commitment letters were added and the terms of debt commitment letters enhanced.219 In 2005, the

213 See United Rentals Eric Swedenburg Depositions Transcript, at [•], 214 Id. 215 I use the term boilerplate here broadly to include structure as well as contract terms. Cf. Todd D. Rakoff, Social Structure, Legal Structure, and Default Rules: A Comment, 3 S. CAL. INTERDISC. L. J. 19, 25 (1993) (“[i]n different social settings different norms are clustered around the practice of contracting itself . . . different norms about how many terms, and which terms, parties should specify.”) 216 See Marcel Kahan & Michael Klausner, Standardization and Innovation in Corporate Contracting (Or "The Economics of Boilerplate"), 83 VA. L. REV. 713, 729-36 (1997). The bulk of scholarship in the complex contracting area has been in the context of sovereign wealth bond and the movement to adopt collective action clauses in the indentures for this debt. See Robert Ahdieh, Between Mandate and Market: Contract Transition in the Shadow of International Order, 53 EMORY L. J. 691 (2004); Robert B. Ahdieh, The Role of Groups in Norm Transformation: A Dramatic Sketch, in Three Parts, 6 CHI. J. INT’L L. 231 (2005); Stephen J. Choi & G. Mitu Gulati, Innovation in Boilerplate Contracts: An Empirical Examination of Sovereign Bonds, 53 EMORY L.J. 929 (2004); Anna Gelpern & Mitu Gulati, Public Symbol in Private Contract: A Case Study, 84 WASH. U. L. REV. 1627 (2006). See also. For discussion of boilerplate generally, see Omri Ben-Shahar & James J. White, Boilerplate and Economic Power in Auto Manufacturing Contracts, 104 MICH. L. REV. 953 (2006); William W. Bratton, Jr., The Economics and Jurisprudence of Convertible Bonds, 1984 WIS. L. REV. 667;); Claire A. Hill, Why Contracts Are Written in "Legalese", 77 CHI.-KENT L. REV. 59 (2001); Marcel Kahan & Michael Klausner, Antitakeover Provisions in Bonds: Bondholder Protection or Management Entrenchment?, 40 UCLA L. REV. 931 (1993). 217 The most prominent example is of this in another context is the collective effort involved in shifting sovereign bond terms from unanimous to collective action clauses. See Choi & Gulati, supra note 216; Galpern & Gulati, supra note 216. 218 See Choi & Gulati, supra note 216, at 993-94. 219 See supra notes 56-58 and accompanying text. 308

8-Oct-08] Failure of Private Equity 45

SunGard transaction heralded a more substantive change in the structure by spurring adoption of reverse termination fee provisions and elimination of financing conditions.220 In both instances, though, the changes were incremental; the transactional equivalent of adding a new room onto a house rather than demolishing the home to rebuild. Rather than a fundamentally new configuration, these revisions rested upon the prior structure and still retained its core – externalized financing through limited liability shell vehicles.

At its simplest level, classical law and economics theory would characterize this result as efficient bargaining among the parties. In private equity transactions, the attorneys are sophisticated and well-compensated, transactional costs in negotiating new structures are low and limited largely by the need to complete these transactions in a short time-frame. The private equity structure is thus an efficiently bargained for result and any change in structure explained by shifting equilibrium among the parties; a new, efficient bargain achieved due to events which force the parties to achieve a new bargaining position.221 The private equity structure has changed so little due to the lack of such occurrences.222

Path dependency theory recognizes that efficient bargaining may be hindered by the initial structure set by the parties. Thereafter, first agreed terms are “locked-in” and become boilerplate and may persist due to network effects, informational deficits, and signaling effects which create excessive transactional switching costs.223 In these later contractual iterations, parties do not renegotiate these terms due to the costs associated with such a change. Thus, later generations of boilerplate may not reflect the bargain parties would necessarily negotiate absent these additional costs; optimal terms thus become sub-optimal or unintentionally suboptimal terms remain sub-optimal.224 Apocryphal examples abound in other

220 See supra notes 60-68 and accompanying text. 221 This comports with Professor Gilson’s theory of the transactional attorney as transaction cost engineering structuring a more efficient deal. See Gilson, supra note 17, at 253-56. See also Symposium: Business Lawyering and Value Creation for Clients, 74 OREGON L. REV. 1 (1995). 222 Most efficiency theorists still recognize that external transaction costs can hinder an efficient bargain. See, e.g., Ian Ayres & Robert Gertner, Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules, 99 YALE L.J. 87, 91 (1989); David Charny, Hypothetical Bargains: The Normative Structure of Contract Interpretation, 89 MICH. L. REV. 1815 (1991). 223 See Marcel Kahan & Michael Klausner, Path Dependence in Corporate Contracting: Increasing Returns, Herd Behavior and Cognitive Biases, 74 WASH. U. L.Q. 347, 353-55 (1996). For a discussion of what exactly path dependency is and its detractors, see Paul A. David, Path dependence, its critics and the quest for ‘historical economics’, in EVOLUTION AND PATH DEPENDENCE IN ECONOMIC IDEAS: PAST AND PRESENT, edited by P. Garrouste and S. Ioannides. For a more skeptical view See STAN LIEBOWITZ, RE-THINKING THE NETWORK ECONOMY: THE TRUE FORCES THAT DRIVE THE DIGITAL MARKETPLACE (2002). 224 Compare Frederick W. Lambert, Path Dependent Inefficiency in the Corporate Contract: The Uncertain Case with Less Certain Implications, 23 DEL. J. CORP. L. 1077 (1988) (arguing that “the case has not been made for the general application of path dependence to the contracting process, but unique attributes of 309

46 82 S. Cal. L. Rev. __ [8-Oct-08 contexts to support this theory including the continued use of QWERTY keyboards and triumph of Microsoft’s Windows systems software.225

In the private equity market participants describe the private equity structure and its evolution as simply the market norm.226 This is common terminology negotiators typically utilize in setting the terms of complex contracts. That is, in negotiations attorneys will reference the market – the predominant or preexisting structure or boilerplate utilized by other market participants -- to establish and validate usage of their own terms for the transaction being negotiated.227 From interviews this appears true of the private equity market also.228 The pre-August 20007 private equity structure was premised upon and negotiated from the structure utilized by others at the time.229 In these circumstances innovation is heuristically anchored to the market.230 Path dependency is created by market participants’ paying heed to the market norm and desire not to stray too far.231 In this light, the static nature of the private equity structure thus appears to be explained more by this path dependency than a transactional efficiency story.

This still does not explain the parties’ resort to a market norm or whether the norm itself might be efficient. More particularly, given the low legal costs and the sophisticated nature of negotiators in the private equity market, switching costs should not be strong forces in this paradigm. Why should the market hold such sway unless it is indeed the efficient equilibrium?

B. The Stickiness of the Private Equity Structure

Answers to this question again come by looking outside the four corners of the documented private equity structure. Prior to August 2007, market certain contracts, such as the corporate bond indenture, may exist in a unique environment conducive to path- dependent suboptimality”). 225 See generally Stan Liebowitz & Stephen E. Margolis, Policy and Path Dependence: From QWERTY to Windows 95, 18 REGULATION NO. 3, at 33 (1995). In other words, due to path dependency a superior product may not necessarily be the successful one in the market as a result of action by others. These people adopt a different less utility maximizing good thereby creating self-reinforcing feedback loops which engender further adoption of the less-superior good. Brian W. Arthur, Competing Technologies, Increasing Returns and Lock-in by Historical Events, 99(1) ECON. J. 116 (1989). 226 The assumption is that terms as agreed are efficient on their face. See, e.g., King & Smith, supra note 29, at 5 citing Robert Daines & Michael Klausner, Do IPO Charters Maximize Firm Value? Antitakeover Protections in IPOs, 17 J. L. ECON. & ORG. 83, 85 (2001) 227 See Russell Korbkin, Inertia and Preference in Contract Negotiation: The Psychological Power of Default Rules and Form Terms, 51 VAND. L. REV. 1583 (1998). 228 See, e.g., Interview A; Interview B. 229 See supra Part I.B. 230 See Korbkin, supra note 227. 231 See Omri Ben-Shahar & John Pottow, On the Stickiness of Default Rules, 33 FLA. ST. U. L. REV. 651, 687 (2006). Professors Kahan and Klausner attribute this in part to herd behavior. See Kahan & Klauser, supra note 223, at 350-53. 310

8-Oct-08] Failure of Private Equity 47 participants relied upon external norms and forces created within the contract itself to complete the private equity structure. Private equity firms were incentivized to complete transactions by reputational constraints; if the firms terminated acquisition transactions they would be seen as unwilling to keep their bargains. In the multi-player, repeat private equity market, after such an event future acquirees would become less willing to transact with them.232 Moreover, the relationship established by the private equity contract and the bonding, signaling and understandings reached during contract negotiation provided further assurances of transaction completion. In other words, forces outside the contract functioned to provide acquirees with assurances that acquisitions would complete.

Where the contract is only part of the understood bargain, attorneys may be incentivized to choose “good enough” options.233 Attorneys stick to the market “norm” rather than engage in further transactional engineering to attempt to further optimize the structure. They do so first because more creative solutions can expose them to criticism. The use of the market structure is less likely to be questioned – its use in other transactions validates it. But a different structure will be noticed in the market and commented upon perhaps negatively. This is a hidden switching transaction cost. Moreover, in negotiating a new structure mistakes or unanticipated events may expose the structure to failure and the innovating attorneys to criticism. Second, attorneys are disincentivized to take this risk since, extra- legal forces and constraints buttress the structure, making it an acceptable one. The result is that attorneys are significantly incentivized to keep to the “market” norm and adhere to boilerplate contractual structures in negotiating complex contracts.234

These forces mean that innovation is typically based on the prior structure and incremental at best. Attorneys look to add value to transactions, value which can be shown to their clients. But in doing so these lawyers rely upon market norms for guidance and are hamstrung by prior precedent. They seek to add value, but not too much value as

232 The private equity community is a small one, but the community ties in the private equity industry appear to be lesser than in the type of close-knit community scholars typically cite as a predicate for establishing relational contracting or extra legal dispute resolution mechanisms. See, e.g., ROBERT C. ELLICKSON, ORDER WITHOUT LAW: HOW NEIGHBORS SETTLE DISPUTES (1991) (examining extra legal dispute resolution mechanisms in the cattle-ranching community in Shasta County California). See also OLIVER E. WILLIAMSON, THE ECONOMIC INSTITUTIONS OF CAPITALISM: FIRMS, MARKETS, RELATIONAL CONTRACTING (1985); Holly Raider, Repeated Exchanges and Evidence of Trust in the Substance of Contracts (1999). 233 Cf. Hill, supra note 19, at 70-76 (asserting economic, psychological and other dynamics collectively function to engender imperfect contracts). 234 See Russell B. Korobkin, Inertia and Preference in Contract Negotiation: The Psychological Power of Default Rules and Form Terms, 51 VAND. L. REV. 1583 (1998); Russell B. Korobkin, The Status Quo Bias and Contract Default Rules, 83 CORNELL L. REV. 608 (1998) 311

48 82 S. Cal. L. Rev. __ [8-Oct-08 incentives to truly innovate in relational contracts are reduced since extra- contractual forces will serve to complete the contract; forces which are perceived as strong. Attorney-initiated alterations to boilerplate in complex contracts thus remain within the embrace of the existent market structure. It becomes largely a hard-fought negotiation over the details of terms and wording but not the structure itself. These smaller disputes permit attorneys to show that they are negotiating hard for their clients within the parameters of their ability and desire to do so.235

Substantive change when it comes is thus likely to be a result of external forces arising outside the traditional legal community.236 These are akin to the “interpretative shocks” which some have also theorized cause boilerplate to move substantively.237 Even then the forces mitigating against change in complex contracts work to keep any shift towards the incremental and premised upon the prior structure.238 This is not a network effect in the sense of normal usage of the term – that the transactional costs of switching prevent change.239 Rather it is an agency problem – attorney incentives work to hamper change and confine it within the structure itself.240

The evolution of private equity structure, a short-term relational agreement, comports with this theory. The structure has seldom changed over two decades. The largest shift, the implementation of a reverse termination fee structure, came from an externalized source, the credit market. This is not surprising. The credit market has historically been more turbulent and prone to failed transactions.241 Moreover, the liquidity wave within the markets in 2005 was unprecedented.242 This drove a shift in the terms of debt commitment letters as lenders, knowing that the debt would subsequently be sold in the market, became remarkably unconcerned with credit risk focusing instead on the saleability of this debt. In interviews with

235 See Hill, supra note 19. 236 This appears partially at odds with the traditional view of innovative environments which find that high resource, decentralized and open communities are flash points for innovation. See Robert Drazin & Claudia Bird Schnoohoven, Community, Population and Organizational Effects on Innovation: A Multilevel Perspective, 39 ACAD. MANAG. J. 1065 (1996). But it does match with theories that posit change as an external force. See RONALD S. BURT, STRUCTURAL HOLES: THE SOCIAL STRUCTURE OF COMPETITION (1992) 237 See Choi & Gulati, supra note 216, at 993-94. 238 See W. Scott Frame & Lawrence J. White, Empirical Studies of Financial Innovation: Lots of Talk, Little Action, 42 J. ECON. LIT. 116, 122 (2004). 239 See Kahan & Klauser, supra note 223, at 350-53. 240 See Michael E. Boardman, Contra Proferentem: The Allure of Ambigous Boilerplate, 104 MICH. L. REV. 1105 (2006) (postulating that drafters of boilerplate may persist in writing ambiguous or confusing terms due to prior court interpretations which clarify this language). 241 For example, the late 1980s collapse of the high yield market almost bankrupted First Boston and forced it into an acquisition transaction with Credit Suisse. See The Burning Bed, BUSINESS WEEK, May 1990. 242 See Speech of Governor Kevin Warsh at the New York University School of Law Global Economic Policy Forum, New York, New York, Financial Market Turmoil and the Federal Reserve: The Plot Thickens (April 14, 2008). 312

8-Oct-08] Failure of Private Equity 49 market participants this shift created value for the private equity firms allowing them to negotiate firmer debt commitment letters.243 Acquirees in private equity captured some of that value in negotiations to foster the 2005 shift in the private equity structure.244

The change in the private equity structure was thus a response to events in the related credit market. The additional value accrued to private equity firms was in part seized by lawyers for acquirees.245 However, the private equity lawyer was incentivized to graft any changes onto the pre-existing structure. Thus, the reverse termination fee structure was placed onto the pre-2005 private equity structure which permitted acquirers to terminate the acquisition agreement in the event of a MAC to the acquiree company.246 The material adverse change mechanic was itself one that was initially inserted into the structure in the 1980s borrowed at that time from agreements for strategic transactions.247 In grafting these new terms onto the private equity structure, lawyers underestimated and failed to fully account for the issues surrounding this structure and its adaptability. Instead, they relied upon “good enough” solutions and extra-legal constraints and understandings to complete the new structure. The result was the flawed structure exposed by the recent wave of litigation.

Additionally, the need of attorneys to bargain and show value within the bounds of structure played out in the evolution of private equity in the pre- August diffusion of choice among reverse termination fee structures.248 This explains the bargaining and diffusion of deal structure among a pure reverse termination deal with a single or two-tiered reverse termination fee versus a specific performance termination fee. Negotiating these variations allowed private equity attorneys to show value and for private equity firms to signal their intentions within the bounds of the contract.249

Another example of negotiation with the bounds of the structure

243 See Leinwand and Goldfeld, supra note 161, at 1. See also Breaking Up is Costlier to Do, M&A: DEALMAKERS J., Apr 1, 2007 (“The prevalence of these fees is connected to the availability of capital, deal attorneys say.”) 244 See Sorkin & Swedenburg, supra note 50, at 2. 245 See Interview C and supra notes 60-68 and accompanying text. 246 See supra notes 60-68 and accompanying text. 247 See Ronald J. Gilson & Alan Schwartz, Understanding MACS: Moral Hazard in Acquisitions, 21 J.L. ECON. & ORG. 330 (2005). 248 See supra notes 60-68 and accompanying text. 249 This micro-variation is exacerbated by the “battle of the forms” which occurs in each transaction. Each major private equity law firm has their own form for the transaction, and each form is structured similarly but is written in a different style and with slightly different wording for each of the common clauses. In the digital age there is much more of a merging of the forms, rather than a dominant form based on who writes the first draft. Each lawyer will work to make the contract more familiar and comfortable to himself. The result is oftentimes pidgin. 313

50 82 S. Cal. L. Rev. __ [8-Oct-08 occurred among choice of law clauses. Private equity contracts began to more frequently use Delaware for their choice of law after an adverse decision rendered in 2005 in Consolidated Edison v. Northeast Utilities250 limited the ability of acquirees to sue under New York law for monetary damages for lost premium when a acquirer breached an acquisition agreement.251 In order to ensure certainty on this issue, Delaware subsequently became the preferred choice of law for private equity contracts.252 Lawyers thus viewed this contract change as an easy one that could show their knowledge and ability to structure contracts in light of the law. This was further illustrated on the less frequent occasions when private equity actors elected to have their contracts governed by New York law. In those instances parties sometimes added specific language to address the Consolidated Edison case.253 But again, the piecemeal and incomplete nature of change in the private equity structure came back to haunt participants – the shift to Delaware choice of law clauses failed to account for the continuing use of New York choice of law clauses in debt commitment letters resulting in multiple causes of actions under different laws in a number of private equity litigations. Lawyers simply failed to account for the interaction of these differing law and forum clauses in the event of litigation.254

The need for attorneys to show value also explains how changed terms in these complex contracts diffuses. In SunGard the transaction involved the top law firm of Simpson Thacher.255 Their involvement as the initiators can be explained by their prominent role as counsel in many private equity transactions: Simpson was counsel on approximately 21% of all public private equity deals with a value greater than $100 million from January 1, 2004 through August 1, 2007.256 The structure had been used intermittently

250 426 F.3d 524 (2d Cir. 2005). 251 The Con Ed case held that neither a acquiree nor its shareholder could sue a acquirer for the premium agreed to be paid on their shares under an acquisition agreement. Id.; see also Miller, supra note 162. 252 I find that from 2005-2007, [●]% of private equity contracts had a Delaware choice of law and [●]% a New York choice of law. Mergermetrics Database (all deals over $100 million). [My findings find the reversal of the trend found by Professors Eisenberg and Miller against Delaware law in acquisition agreements. See Theodore Eisenberg and Geoffrey Miller; Ex Ante Choice of Law and Forum: An Empirical Analysis of Corporate Acquisition agreements. 59 VAND. L. REV. 1975 (2006). – to be confirmed] 253 See, e.g., Agreement and Plan of Merger among NuCO2 Acquisition Corp., NuCO2 Merger Co. and NuCO2 Inc., at Sec. 9.02, dated as of January 29, 2008, available at http://www.sec.gov/Archives/edgar/data/947577/000092189508000276/ex21to8k01124_01292008.htm (acquisition agreement under New York law adding language that in the case of breach of the agreement “shall not be limited to the Expense Reimbursement Amount and may include the benefit of the bargain of the Merger to such party (and, in the case of the Company, its stockholders), adjusted to account for the time value of money”). 254 It was likely these complications which led Penn National Gaming to agree to terminate its own acquisition transaction with Fortress and Centerbridge. Penn had a specific performance form of acquisition agreement, but the prospect of multiple litigation in differing forums likely led it to agree to settle. 255 See Exhibit 99.1 to SunGard Current Report on Form 8-K, dated Mar 27, 2007, available at http://www.sec.gov/Archives/edgar/data/789388/000119312505061716/0001193125-05-061716-index.htm. 256 For private equity transactions over $100 million of public acquirees. Author calculations from 314

8-Oct-08] Failure of Private Equity 51 in prior transactions, but the use in such a historic transaction by these significant firms validated it for use by other parties.257 In fact, Simpson publicly promoted its work on this structure in order to highlight its innovative nature.258 The quick adoption reflected the widespread number of private equity participants in the SunGard transaction, the competitive nature of the private equity market which made this innovation seem more attractive in the market and the desire of top law firms to “compete” by offering this innovation as a product developed by them.259 Here, change in the private equity context appears to diffuse much quicker than in other studied areas of corporate innovation such as the poison pill.260 This is likely due to the smaller private equity legal community and their ability as sophisticated parties to rapidly respond to change.261 Again, though, this change was one that ultimately was incremental and brought on by external forces. So, the change was ideal for adoption. It was pronounced by the leading private equity participants and an incremental change that other market participants were likely to find within the market norm and therefore less risky to adopt.

This notion of change – that it comes from the top down rather than from external innovators in the complex contracting model – is one that is driven by the structure of the private equity industry.262 In the private equity model barriers to entry are high. The attorneys who regularly represent the private equity firms and their acquirees are sizable, prominent law firms who have the capacity to negotiate these transactions in a timely manner and are familiar with market nuances. New entrants are unable to offer these skills, lack validation in this market and are a signal of inexperience by the party hiring them and thus are largely prevented from entry. Moreover, when they do enter into the game, these new firms typically represent acquirees on a one-off basis.263 Any innovation by them is not protected

MergerMetrics Database. See Choi & Gulati, supra note 216 (citing the role of Cleary Gottleib in effecting change to unanimous consent clauses in sovereign wealth bond indentures). 257 Here, the fact that the structure was publicized by Simpson Thacher may have been as a vehicle to promote their own private equity clients as providing a superior transaction structure in competitive bidding situations. 258 See Sorkin & Swedenburg, supra note 56. 259 See Michael J. Powell, Professional Innovation: Corporate Lawyers and Private Lawmaking, 18(3) L & SOC’Y INQ. 423, 433-39 (1993) (discussing Wachtell Lipton’s marketing efforts for its poison pill innovation). 260 See David Strang & Sarah Soule, Diffusion in Organizations and Social Movements: From Hybrid Corn to Poison Pills, 24 ANN. REV. SOC. 265 (1998). See also Gerald F. Davis, Agents Without Principles? The Spread of the Poison Pill Through the Intercorporate Network, 36 ADMIN. SCI. Q. 583 (1991). 261 There is a similarity to the venture capital community in the size and cohesiveness of private equity. See Mark C. Suchman, On Advice of Counsel: Law Firms and Venture Capital Funds as Information Intermediaries in the Structuration of Silicon Valley (1994) (unpublished Ph.D. dissertation, Stanford University). 262 This jibes with those who have described innovation in a world of unpatentable ideas as driven by those who dominate the market and can therefore profit maximally from such innovation. See Frame & White, supra note 238, at 119. 263 From 2004-August 1, 2007, 16.2% of law firms represented a acquiree or acquirer more than five times while 62.6% did so only once. Author calculations from Mergermetrics database for public private equity 315

52 82 S. Cal. L. Rev. __ [8-Oct-08 intellectual property and a public good. It can be quickly taken advantage of by other, more prominent firms who can capture more value from such innovation.264 These outsiders are therefore not highly incentivized to innovate, lack economy of scale and being previously unfamiliar with the structure they resort to the market “norm”. Here, their incentives are similar to more experienced lawyers who do not want to take the risks of deviating from current structure. This is a risk that is heightened in the case of newcomers since deviation is more apt to be viewed as a result of inexperience rather than true innovation.265

C. Law Firm Centrality and the Private Equity Structure

The private equity world is a discrete one, and a few select law firms represent the majority of acquirees and private equity acquirers.266 This circumstance can also function to further hamper and forestall innovation and change. The following chart sets forth the legal representation of private equity firms and acquirees in all private equity buy-outs greater than $100 million from January 1, 2004 through August 1, 2008:

transactions greater than $100 million in value. 264 See Suchman, supra note 197, at 105. 265 The model put forth here may be at odds with events in other corporate arenas such as change and innovation in the venture capital community. There, the unique innovational mindset of the industry participants and separation from the East coast law firms nurtured a unique model erected by external law firms. See Mark A. Suchman, The Contracting Universe: Law Firms, Venture Capital Funds and the Institutionalization of New- Company Financing in Silicon Valley. See also Michael J. Powell, Professional Innovation: Corporate Lawyers and Private Lawmaking, 18(3) L & SOC’Y INQ. 423 (1993). 266 See also Mark C. Suchman & Mia L. Cahill, The Hired Gun as Facilitator: Lawyers and the Suppression of Business Disputes in the Silicon Valley, 21 L. & SOC. INQUIRY 679, 683 (1996) (Silicon Valley venture capital lawyers “[t]hrough their relations with both entrepreneurs and investors, they identify, create, transmit, and enforce the emerging norms of the community . . . . facilitating what might otherwise be prohibitively costly, complex, and unpredictable transactions.”) 316

8-Oct-08] Failure of Private Equity 53

Chart IVA267 Both Neither Either Only Acquiree and Acquiree nor Acquiree or Only Private Equity Acquirer Acquirer Acquirer Acquiree Firm Represented Represented Represented Represented Represented By Top Private By Top Private By Top Private By Private By Top Private Equity Law Equity Law Equity Law Equity Law Equity Law Firm Firm Firm Firm Firm Transaction Size >$500 Million 39.66% 3.91% 20.67% 4.47% 16.20% Transaction Size <$500 Million 12.29% 5.03% 18.44% 5.03% 13.41% Total of all Private Equity Transactions from 1/1/2004- 8/1/2007 51.96% 8.94% 39.11% 9.50% 29.61%

Chart IVA reveals that representation of private equity firms is highly concentrated and a core group of 22 law firms represented a private equity firm in approximately 81.5% of transactions and were involved in 91.07% of all transactions.268 Six firms were ubiquitous during this time period and involved in the majority of transactions:

Chart IVB269 Law Firm Law Firm Law Firm Total Percentage Represented Represented Transactions of all PE Acquiree PE Firm Involved Transactions from Jan. 1, 2004-Aug. 1, 2007 >100 MM Simpson Thacher 8 29 37 21% Skadden Arps Meagher Slate & Flom 18 16 34 19% Wachtell Lipton Rosen & Katz 16 12 28 16% Kirkland & Ellis 5 14 19 11% Latham & Watkins 11 7 18 10% Ropes & Gray 3 15 18 10%

267 Author calculations from Mergermetrics database. 129 law firms were involved in 192 public private equity transactions with a value greater than $100 million during this time period. Top Private equity firms are the 22 firms out of 129 firms involved in private equity deals and who had six or greater representations on a private equity deal during the applicable time period. 268 Id. 269 Id. 317

54 82 S. Cal. L. Rev. __ [8-Oct-08

As Charts IVA and IVB show, in this specialized market private equity firms have a stable of law firms they repeatedly retain. These law firms are highly experienced in negotiating and familiar with the structure of private equity. This additionally incentivizes these firms to represent their clients through the possibility of repeat business opportunities.

This trend persists for acquirees. Chart IVA shows that acquirees were represented by a top private equity law firm in 61.4% of transactions. However, this figure is lower than private equity firms due to acquiree preferences for their traditional counsel for representation.270 These non-top firms are not as experienced in and likely not as familiar with the private equity structure. Of the 129 law firms I found represented a party in a private equity transaction during the time period January 1, 2004-August 1, 2007, 63 provided legal advice in only one private equity transaction.271 And outside the top 22 law firms these other 107 law firms represented a acquiree 69.2% of the time and a private equity firm 30.8% of the time.272

In the case of these non-top law firms, acquirees may not obtain equivalent legal advice. The firms are often smaller and lack the skill-set and ability to make the appropriate investments to understand and innovate with respect to the structure of private equity. Moreover, as discussed supra their incentives to do so in these circumstances are low. 273 These non-top law firms are unlikely to obtain further private equity representations, and they are likely losing part of their client business as a consequence of the sale. Because of the need for this experience, acquirees still in a majority of circumstances retained a top private equity firm, versed in the structure of private equity. However, the top law firms’ economic incentives may be skewed: their primary clients are the private equity firms who provide repeat business. Accordingly, the incentives of the top law firms to innovate or otherwise challenge their clients and push for a structure that would harm their ability to further obtain private equity business are further lowered.274 Thus, the particular nature of the private equity legal field may combine with the other described forces to prevent substantive innovation to the current private equity structure, to the detriment of acquirees.275 Reliance of

270 This may be due to, among other things, relationship specific investments which incentivize acquirees in these circumstances to use their traditional counsel. See Michael Klausner et al., The Law And Economics Of Lawyering Second Opinions In Litigation, 84 Va. L. Rev. 1411 (1998). 271 Author calculations from Mergermetrics database. 272 Id. 273 See supra notes 262-265 and accompanying text. 274 See MACKLIN FLEMING, LAWYERS, MONEY AND SUCCESS (1997); Lisa Bernstein, The Silicon Valley Lawyer as Transaction Cost Engineer?, 74 OR. L. REV. 239, 248 n.43 (1995). 275 See generally Gillian Hadfield, Legal Barriers to Innovation: The Growing Economic Cost of Professional Control Over Corporate Legal Markets, 60 STAN. L. REV. 102 (2008). 318

8-Oct-08] Failure of Private Equity 55 attorneys on extra-contractual forces works to cover for this possible conflict by allowing acquiree attorneys to otherwise justify the structure of private equity. I believe that this potential conflict did not affect attorney bargaining on granular issues – law firms still attempted to show value to their clients. Rather reliance on extra-contractual forces permitted acquiree attorneys to forgo innovating and bargaining over more favorable unfamiliar terms. This additional potential agency conflict was thus likely not a determinant force in setting the structure and terms of private equity but rather a reinforcing one.276

D. The Optimality of the Private Equity Structure

The question remains as to whether the private equity contract was an “optimal one”. But the role of and forces upon attorneys and the nature of change in private equity contracts belie a term like optimal.277 In efficient bargaining without transaction costs or other external drags sophisticated, well-represented parties can negotiate optimal contracts. Here, I define an optimal contract as an efficient bargain which does not produce excess, unanticipated transaction costs or an uneconomical remedy for a foreseeable eventuality.278 The structure of private equity does not meet this test. Due to attorney agency costs, it is a path dependent structure resistant to change and created piecemeal upon the pre-existing structure.279 The private equity structure is also one that inherently and necessarily relies on extra-legal forces to complete it. If these external forces fail they are likely to expose the flaws permeated within the structure. The post-2007 August events affecting the private equity structure provided very real proof of this. The structure of private equity is thus a suboptimal bargain.

V. THE FUTURE OF PRIVATE EQUITY

The private equity attorney as agent and transaction cost engineer is more than just a scrivener, negotiating words in a contract. The attorney, together with his or her principal, weighs the wording of the contract

276 Evidence for this comes from the United Rentals litigation. In that transaction Simpson Thacher represented the acquiree; from the trial transcripts it appeared to be bargaining hard within the bounds of the current private equity structure. See supra at Part III.A.2. 277 See, e.g., Kahan & Klausner, Standardization and Innovation, supra note 216, at 750-51 (arguing that the “put at par” event risk covenant feature in bond indentures was suboptimal). 278 Others simply define “suboptimal” or “optimal” as a faulty bargain that is unfavorable to one party in its overt economics. See, e.g., Choi & Gulati, supra note 216, at 940-943 (discussing the debate over whether unanimous consent clauses in sovereign wealth bond indentures are “suboptimal”); Kahan & Klausner, Standardization and Innovation, supra note 60, at 751 (“[t]he remedy provided to bondholders in most covenants is a put at par. This, however, is a faulty remedy [since] the put would overcompensate bondholders and possibly deter efficient transactions.”) 279 See Henry T. Greely, Contracts as Commodities: the Influence of Secondary Purchasers on the Form of Contracts, 42 VANDERBILT L. REV. 133 (1989). 319

56 82 S. Cal. L. Rev. __ [8-Oct-08 against other legal and non-legal factors. The attorney then drafts the contract to reflect this weighing, confirming their impressions through the dialogue and bonding of the negotiation itself.280 The private equity contract is thus a complex contract, and like all such contracts is dependent upon outside forces to finish it. The contract terms function together and interact with external norms and constraints to form a fuller bargain. This bargain is greased by the relational bonding and signaling that occurs between transactional participants and their representatives in the negotiation of the transaction and which continues even as the relationship breaks down. But due to the nature of the private equity structure the contract is often ambiguous in nature or otherwise flawed in its terms.

Prior to August 2007, these flaws and ambiguity were accepted for a number of reasons. First, lawyers appeared to rely on norms and external and contractual constraints to complete the private equity structure. Attorneys were thus not incentivized to restructure or fix the private equity structure. Second, the old structure had been validated and the new mechanics of the post-2005 structure were “good enough” -- able to get to deal completion. Further negotiation over these particular structural terms appeared unnecessary given the past success of the structure and the perception that the new one was incrementally better and created more certainty. Third, to the extent these flaws actually created uncertainty they allowed acquirees and their attorneys to achieve a desired result – a deal -- while preserving an option for further, later negotiation. Finally, the nature of legal representation in the private equity world – where the top firms provide the bulk of legal advice and repeatedly represented both acquirees and private equity firms likely functioned with these other factors to inhibit any challenge to or innovation within the structure.

In this light, the post-2007 private equity implosion was a failure of lawyers, particularly those for acquirees, to properly assess risks and to innovate. Lawyers and other transaction participants failed to predict the effects of a mass market disruption as occurred in August 2007 and onwards.281 In other words, acquiree lawyers as agents over-relied on these external forces to forgo innovative approaches to bridge the pre-existing closing gap and “fix” the flaws in the private equity contract despite the low transaction costs to do so. This over-reliance may have been enforced by the

280 Cf. Michael J. Powell, Professional Innovation: Corporate Lawyers and Private Lawmaking, 18 LAW & SOC. INQUIRY 423 (1993). 281 See Megan Davies and Jessica Hall, Buyout spats bruise many, damage trust, Jul 7, 2008, available at http://biz.yahoo.com/rb/080707/dealtalk_buyouts.html?.v=3 (quoting Marilyn Sonnie, a partner at law firm Jones Day stating that "Boards never really thought they were signing up for a deal that could just evaporate . . . . But a lot of them did.") 320

8-Oct-08] Failure of Private Equity 57 confined nature of the private equity legal community which may have biased the structure in favor of private equity firms. There may not even have been over-reliance -- the market disruption may not have been predictable or otherwise rationally contemplated at the time these pre- August contracts were negotiated. If so, the failure to innovate and sloppy drafting practices may be explained by the perceived low probability of such mass disruption which made the costs of such further innovation not worthwhile. While these explanations no doubt played a part and are what attorneys are likely to cite as the reason, they are belied by future events in the world of private equity.

If the failure of private equity was a failure of calculus with respect to norms and externalized forces, you would predict that lawyers would reassess these forces in light of post-2007 events. You would further predict that this calculus would militate a shift towards contractual certainty in the private equity structure. Indeed, in the wake of this collapse, commentators and market participants speculated that the structure would be entirely transformed to be become significantly more favorable to acquirees.282

So far, this is not what has happened. Instead, the private equity structure has shifted in the opposite direction towards a more- private equity favorable model. Every one of the private equity transactions announced in 2008 has utilized a pure reverse termination fee structure.283 This is a telling response. The nature of this shift marks a recognition that the drivers to closing in a private equity transaction substantially exist outside the contract language.284 It also represents a collapse of the bargain between private equity firms and acquiree companies which permitted more rigorous forms of the reverse termination fee structure to exist.285

The response of acquirees and their attorneys fits within a path dependent model of private equity. In this scenario, attorney innovation may not be as appealing since it will again be subject to scrutiny and failure.

282 Interview D. See also Andrew Ross Sorkin & Michael J. de la Merced, The fine Art of Dealmaking Gathers Dust, THE N.Y. TIMES, Apr 2, 2008, at 8 (reporting that “Sixty-two percent of the respondents to the Brunswick poll said that reverse termination fees, payments that acquirers can use to walk away from deals, will be tightened or amended over the next year.”); Bird, supra note 83 (discussing seller expectations for tighter deals in the private equity market) 283 As of June 26, 2008, there were 11 of these transactions in value greater than $200 million. Mergermetrics database. See also Steven M. Davidoff, The More Things Change, N.Y. TIMES DEALBOOK, Jan. 25, 2008, available at http://dealbook.blogs.nytimes.com/2008/01/25/the-more-things-change/ 284 Moreover, within the transactions agreed to in 2008 there was an observable “cleaning-up” of language. See Email from Alan Fishbein, dated June 9, 2008. 285 This supports the view that the pre-August 2007 reverse termination fee structure represented a negotiated bargain between private equity and acquirees wherein private equity acquirees gave up contractual leverage in reliance upon external forces. The latest form of private equity structure is merely private equity firms eliminating this overlay. 321

58 82 S. Cal. L. Rev. __ [8-Oct-08

By negotiating within the prior structure and instead relying upon externalized forces lawyers allow the burden of failure to be assumed by the companies rather than lawyers themselves. Again, we have an agency cost. Because of this though, the role of relational bonding in the contract stage becomes even more important. Not only are companies self-selecting for acquisitions, they need to ensure that the external forces to closing will hold.286 Anecdotally from transaction descriptions from the post-2008 transactions this appears to be the case. It is no surprise that the 2008 private equity deals thus far have been in industries less affected by the market disruption. Acquirees justified using the reverse termination fee structure due to their stable cash generative business models which would make them less resistant to any adverse impact by the economic crisis. This would ensure that their business remained stable and the private equity acquisition would complete.

Relatedly, in the post-August 2007 litigation private equity firms appeared to repeatedly be able to find some clear or less than clear contractual or legal basis to attempt to terminate their agreements. The failure of private equity shows the importance of extra legal forces in gluing together transactions. In complex transactions there will always be limits to what attorneys can do, at some point further additions or revisions to the contract are constrained by bounded rationality or are otherwise hampered by time constraints or other transaction costs. This may always mean that there is some hook that an acquirer can find to terminate a transaction. In other words, when a dispute arises lawyers are always reasonably certain in the complex contract context that they can find some flaw to litigate.287 This type of behavior was clearly on display in the private equity failures of the past year. The consequence is that in the private equity context, and likely complex contracts generally, norms are an important and inescapable component of a contract.288

This combined with the other limitations in the private equity structure have worked to make the response to the events of August 2007 a muted one. The pure reverse termination fee model so far appears to have remained and become the sole breed of the private equity structure. In fact, it has even been enhanced in certain instances by additional conditionality such as EBITDA conditions.289 The specific performance model has been

286 This likely explains the first instance where the private equity structure was utilized in a significant strategic transaction, Mars’ agreement to acquire Wrigley. See Steven M. Davidoff, Wrigley and the Future of M&A, N.Y. TIMES DEALBOOK, May 1, 2008, available at http://dealbook.blogs.nytimes.com/2008/05/01/wrigley- and-the-future-of-ma/. 287 Interview Y. 288 The issue with private equity is thus not reliance on extra-legal norms but over-reliance. 289 An EBITDA condition is a condition that the acquiree will met a set acquiree for earnings before 322

8-Oct-08] Failure of Private Equity 59 completely abandoned. As with other change in the private equity structure, this will likely remain the case until some externalized change results in a bigger shift.290

Lawyers still act within this confined structure in order to show value to their clients. Within the model painted in this Article, you would predict that this would lead to the fixing of some of the drafting ambiguities and other smaller defects in the private equity model. This is what appears to have happened – in the 2008 transactions announced thus far the contract language appears to be clearer and more carefully drafted.291 In addition, many of the other flaws in the private equity model appear to be addressed. The scope of MAC clauses has been expanded to partially address the problem of its interaction with reverse termination fee clauses. Reverse termination fees have varied more and creeped higher in amount. “Reasonable best efforts” clauses appear to have been enhanced by making the reverse termination fees payable upon the failure of the transaction due to regulatory reasons providing a stick to force the cooperation of private equity funds. The response is an expected one based upon the forces driving private equity – the flaws in private equity that otherwise are easy to fix – the transaction costs to doing so are low. The fundamental structure, though, remains.292

CONCLUSION

The conclusions of this Article are necessarily early. The full consequences of the past years’ shock to the structure of private equity are yet to be fully known. Moreover, the complete response of the private equity participants and its effect on the structure of private equity will likely only be achieved once the market returns to stability in the coming years. Yet, the window on the private equity structure opened by this shock has already informed our understanding of how these complex contracts are negotiated, agreed and evolve. The private equity attorney as transaction cost engineer relies upon more than just the words of the contract and law,

interest, taxes, depreciation and amortization (EBITDA) before the buy-out can be consummated. See, e.g., Agreement and Plan of Merger, dated as of June 15, 2008, by and among QGF Acquisition Company Inc., QGF Merger Sub Inc. and Greenfield Online, Inc., Sec. 6.02, available at http://www.sec.gov/Archives/edgar/data/1108906/000095012308006885/y60673kexv2w1.htm. 290 This likely will come from competition from strategic acquirers willing to offer a more certain transaction structure. 291 See David Marcus, Desperately Seeking Certainty, THEDEAL.COM, July 18, 2008, at http://www.thedeal.com/newsweekly/features/desperately-seeking-certainty.php 292 Still, I believe the story of the Failure of Private Equity supports the neoformalists. [Flesh this point out] Compare William J. Woodward, Neoformalism in a Real World of Forms, 2001 WISCONSIN LAW REVIEW 971 (arguing that neoformalism lacks persuasive empirical evidence that it will "cause" people to take more care in making or reading contracts). 323

60 82 S. Cal. L. Rev. __ [8-Oct-08 but also on norms, conventions, bonding and other extra-legal factors to complete contracts. However, the structure of private equity is not an efficient transactional structure. Reliance, and perhaps over-reliance, on these extra-legal forces create a path dependent structure which dampens innovation, covers up possible attorney inefficiencies and conflicts and allows flaws to creep into the private equity contract. The failure of private equity shows the limitations and failures of lawyering. Preliminary response of attorneys to the private equity implosion confirms this and also informs our view about the nature of change and response to contract structure and terms. Only in the light of ensuing years, though, will we be able to confirm whether the forces and failures cited and discussed here continue to drag on the structure of private equity.293

293 The failure to innovate in response to this shock is not only affecting acquirees, but also private equity firms who due to their reliance on optional structures cannot now effectively compete for acquisitions. See Vipal Monga, Blackballed, THE DEAL, Jun 6, 2008 (reporting that “[w]hen Royal Bank of Scotland Group plc announced last month that it would sell its insurance business, it took the remarkable step of excluding private equity firms from the auction” due to the uncertainty embedded in their acquisition agreements). 324

8-Oct-08] Failure of Private Equity 61

APPENDIX A Selected Terminated and Renegotiated Private Equity Transactions (Aug. 2007-Aug. 2008) Ann. Date Acquiree Acquirer Outcome Nov, 16, 2006 Clear Bain/Thomas H. Channel Lee Renegotiated Terminated (Settlement Mar. 7, GE Capital Agreement) 2007 PHH Solutions / Blackstone Mar. 7, CEO led MBO / 2007 ACS Cerberus Capital Terminated Terminated Apr. 7, Sallie Mae JC Flowers (Settlement 2007 (SLM Corp) consortium Agreement) Clear Apr. 7, Channel TV Providence Equity 2007 Stations Partner Renegotiated Terminated Apr. 7, Myers Goldman Sachs (Settlement 2007 Industries Capital Partners Agreement) Terminated Apr. 7, (Settlement 2007 Harman KKR / Goldman Agreement) Terminated May 7, Acxiom ValueAct Capital / (Settlement 2007 Corp. Silver Lake Agreement) May 7, Alliance 2007 Data Systems Blackstone Terminated Jun. 1, 2007 19X, Robert X. Sillerman and Simon CKX, Inc. Fuller Terminated Accredited Jun 4, Home Lenders 2007 Holding Co. Lone Star Funds Renegotiated Jun. 7, Fortress 2007 Penn Investment Group and Terminated National Centerbridge Partners (Settlement Gaming Inc. LP Agreement) Terminated Jul. 7, Reddy Ice GSO Capital (Settlement 2007 Holdings Partners Agreement) Jul. 7, United Cerberus Capital 2007 Rentals Management Terminated Aug. 7, Capital One 2007 NetSpend Financial Terminated 325

62 82 S. Cal. L. Rev. __ [8-Oct-08

Teachers Private Capital, Providence Equity Partners Inc. and Jun 30, Madison Dearborn 2007 BCE, Inc. Partners, LLC. Renegotiated Jul, 25, Cumulus Merrill Lynch 2007 Media, Inc. Global Private Equity Terminated Sep 7, 2007 3Com Bain / Huawei Terminated

326

NBER WORKING PAPER SERIES

URBAN INEQUALITY

Edward L. Glaeser Matthew G. Resseger Kristina Tobio

Working Paper 14419 http://www.nber.org/papers/w14419

NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts Avenue Cambridge, MA 02138 October 2008

All three authors thank the Taubman Center for State and Local Government for financial support. The views expressed herein are those of the author(s) and do not necessarily reflect the views of the National Bureau of Economic Research.

© 2008 by Edward L. Glaeser, Matthew G. Resseger, and Kristina Tobio. All rights reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including © notice, is given to the source. 327

Urban Inequality Edward L. Glaeser, Matthew G. Resseger, and Kristina Tobio NBER Working Paper No. 14419 October 2008 JEL No. H0,I0,J0,R0

ABSTRACT

What impact does inequality have on metropolitan areas? Crime rates are higher in places with more inequality, and people in unequal cities are more likely to say that they are unhappy. There is also a negative association between local inequality and the growth of both income and population, once we control for the initial distribution of skills. What determines the degree of inequality across metropolitan areas? Twenty years ago, metropolitan inequality was strongly associated with poverty, but today, inequality is more strongly linked to the presence of the wealthy. Inequality in skills can explain about one third of the variation in income inequality, and that skill inequality is itself explained by historical schooling patterns and immigration. There are also substantial differences in the returns to skill, related to local concentrations in different industries, and these too are strongly correlated with inequality.

Edward L. Glaeser Kristina Tobio Department of Economics Kennedy School of Government 315A Littauer Center 79 JFK St- T347 Harvard University Cambridge, MA 02138 Cambridge, MA 02138 [email protected] and NBER [email protected]

Matthew G. Resseger ENTER POSTAL ADDRESS HERE [email protected] 328

I. Introduction

For much of the almost 2,500 years since Plato wrote that “any city however small, is in fact divided into two, one the city of the poor, the other of the rich,” urban scholars have been struck by the remarkable amount of income inequality within dense cities (Wheeler, 2005). While there is certainly plenty of rural inequality as well, the density of cities and urban regions makes the contrast of rich and poor particularly striking. Figure 1 shows the 45 percent correlation between density and income inequality, measured with the Gini coefficient, across counties with more than one person per every two acre. The tendency of dense places to be more unequal motivates this survey of inequality in metropolitan areas, multi-county units containing a dense agglomeration of population.

America is, on the whole, relatively unequal for a developed country (Alesina and Glaeser, 2004), but there are some places within the U.S. that are a lot more equal than others. While Manhattan is the physical embodiment of big-city inequality and has a Gini coefficient of .6, the Gini coefficient of Kendall County, Illinois is only about one- half that amount. Kendall is a small but rapidly growing county on the outskirts of the Chicago area that combines agriculture with a growing presence of middle-income suburbanites. In Kendall, 9.2 percent of households earned more than 150,000 dollars in 2006, and 9.3 percent of households earned less than 25,000 dollars. In contrast, 20.4 percent of households in New York County earned more than 150,000 dollars, and 26.5 percent earned less than 25,000. In Section II of this paper, we discuss the measurement of inequality across metropolitan areas.

Just as at the national level, inequality across metropolitan areas reflects the distribution of human capital, the returns to human capital and governmental redistribution. A primary difference between local and national level inequality is that local inequality is driven to a large extent by decisions of people to live in different places. According to 2006 American Community Survey, seven percent of the U.S. population lives in a different county or country than they did only one year ago, and twenty-one percent of

2 329

the population lives in a different county or country than they did five years ago. Manhattan’s inequality reflects the decisions of both rich and poor to come to the island. Since more than fourteen percent of Kendall’s population lived outside the county one year ago, the area’s reflects the fact it attracts homogenous people.

Paradoxically, local inequality is actually the inverse of area-level income segregation. Holding national inequality constant, local inequality falls as people are stratified across space so that rich live with rich and poor live with poor. A perfectly integrated society, where rich and poor were evenly distributed across space, would have highly unequal metropolitan areas that mirror the entire U.S. income distribution.

In Section III, we find that almost one-half of the variance in income inequality across space can be explained by differences in the skill distribution across metropolitan areas. Places with abundant college graduates and high school dropouts are areas that are particularly unequal. Traditional economic models try to explain the location of skilled and unskilled workers with differences in the returns to skill and differences in amenities (Dahl, 2002). We agree with this framework, but empirically, we find that history and immigration seem to be the most important determinants of inequality today.

Sixty percent of the heterogeneity in skills across larger metropolitan areas can be explained by the share of high school dropouts in the area in 1940 and the share of the population that is Hispanic. Long-standing historical tendencies are highly correlated with the location of high school dropouts and the location of Hispanic immigrants today. Historical skill patterns also play a huge role in the current location of college graduates (Moretti, 2004), and explain much of the distribution of skills across space.

Metropolitan inequality also reflects differences returns to skill. A modified Gini coefficient that holds the skill composition of each area constant, but allows the returns to skill to vary, can explain 50 percent of the variation in the actual Gini coefficient across metropolitan areas. The correlation between the raw Gini coefficient at the metropolitan area level and the estimated returns to a college degree in that area is 73 percent.

3 330

We do not fully understand why some places reward skill so much more strongly than others. In our data, the most powerful correlate of the returns to having a college degree is the share of population with college degrees, but that fact provides more confusion than clarity. The correlation could reflect skilled people moving to areas where there are large returns to skill. Alternatively, it might reflect human capital spillovers which cause the returns to skill to rise. Hopefully, future research will help us better understand the differences in the returns to skill across metropolitan areas.

In Section IV of this paper, we turn to the consequences of local inequality. We find a significant negative correlation between local economic growth and income inequality once we control for other initial conditions, such as the initial distribution of skills and temperature. Places with unequal skills actually grow more quickly, but places with more income inequality, holding skills constant, have slower income and population growth. Inequality is related to crime at both the national and city level (Fajnzylber, Lederman and Lloayza, 2002; Daly, Wilson and Vasdev, 2001). Our data also confirms a robust relationship between the murder rate and inequality. Luttmer (2005) documents that people are less happy when they live around richer people. We also find people who live in more unequal countries report themselves to be less happy.

In Section V, we discuss the policy issues surrounding local inequality. Even if we accept that local inequality has some unattractive consequences, the consequences of reducing local inequality, holding national inequality fixed, are quite unclear. Reducing local inequality, leaving the national skill distribution untouched, implies increased segregation of the rich and the poor.

Moreover, easy migration across areas severely limits the ability of localities to reduce income inequality through redistributive policies (Peterson, 1981). The ability to migrate means that when localities take from the rich and give to the poor, they will induce the rich to emigrate and attract more poor people, which in turn creates added burdens on the city’s finances. After all, the evidence in Section III suggests that the

4 331

location of high skilled people may be quite sensitive to the returns to skill. Decreasing the returns to skill at the local level with higher taxes or other policies will surely induce some skilled workers to leave.

Our final point is that America’s current schooling system puts localities at the center of any attempts to reduce national inequality through a more equal human capital distribution. While localities are inherently weak in their abilities to reduce inequality, decentralized schooling means that any attempt to equalize educational opportunities must rely heavily on localities. Not only will attempts to reduce inequality through more equal education take many years, but they will also require a tricky partnership between national and local governments.

II. Measuring Inequality Across American Metropolitan Areas

In the analysis that follows, we will use metropolitan areas as our geographic unit and the Gini coefficient as our measure of income inequality. Metropolitan areas are defined as multi-county agglomerations that surround a city with a “core urban area” of over 50,000 people. Metropolitan areas have the advantage of at least approximating local labor markets, and they are large enough to provide a certain measure of statistical precision. The disadvantage of using these areas is that they do not correspond to natural political units, which makes them awkward units for analyzing or discussing public policy.

Much of the data for this paper comes from the five-percent Integrated Public-Use Micro- Samples (IPUMS) for the 1980 and 2000 Censuses (Ruggles et al, 2008). In most cases, we will restrict ourselves to total household income, which means that we are not treating single people differently from married people. We will focus on pre-tax income inequality.1 A particular problem with using Census data to measure income inequality is that incomes in the 2000 Census are top-coded at 999,998 dollars and at 75,000 in 1980.

1 Our income measures, like many in this literature, exclude non-income returns from capital ownership, like the flow of services associated with owning a home.

5 332

In the case of top-coding, we use the income top-code, but recognize that this is understating the true degree of income inequality.

We use the Gini coefficient as our measure of inequality, mainly because it is the ubiquitous standard in the inequality literature. Many policy discussions of inequality focus on poverty, and reducing poverty levels is a natural topic of policy attention. Yet this essay is focused on extremes at both the upper and lower levels of the income distribution, and the Gini coefficient captures that heterogeneity. The Gini coefficient, 1 defined as 1 −− ))(1( 2 dyyF , where yˆ is the mean income in the sample and F(y) is yˆ ∫y the share of the population with income levels less than y. This measure has the interpretation as the area between the 45 degree curve (which indicates perfect equality) and the Lorenz curve.2

The Gini coefficient has the advantage of being invariant with respect to scale, so that larger areas or richer areas do not necessarily have larger or smaller Gini coefficients. Moreover, a ten percent increase in everyone’s income will not impact the Gini coefficient. The Gini coefficient also always rises when income is transferred from a poorer person to a richer person. One standard criticism of the Gini coefficient is that the average Gini coefficient of a number of areas will not equal the Gini coefficient calculated for those areas all together.

There are several plausible alternatives such as the variance of income within an area ( 1 ( − ˆ 2 ydFyy )() ) or the coefficient of variation ( ( − ˆ 2 ydFyy )() ). The ∫y yˆ ∫y coefficient of variation, unlike the variance, will not increase or decrease if all incomes are scaled up or down by the same percentage amount. A final type of measure is the difference in income between individuals in different places of the income distribution,

2 If we let p denote F(y), i.e. the share of the population earning less than y, then the Lorenz curve plots the share of national income going to individuals earning less than −1 pF )( as a function of p, i.e. 1 )( ydyyf . yˆ ∫ ≤ −1 pFy )(

6 333

for example the difference in income between people at the 90th and 10th, or the 75th and 25th, percentiles of the income distribution. We calculate these differences using the logarithm of income.

Using the 2000 Census five percent micro-sample, we calculate these five different income inequality measures at the metropolitan area level: (1) the Gini coefficient using household income, (2) the variance of household income, (3) the coefficient of variation of household income, (4) the income difference between the 90th and 10th percentiles of the household income distribution, calculated as the difference in the logs of these numbers and (5) the income difference between the log 75th and 25th percentiles of the household income distribution. Table 1 shows the correlation between our five measures as well as the correlation between these measures and the logarithm of both median family income in the area and population size.

The most reassuring fact in the table is that these income measures are fairly highly correlated. For example, the correlation coefficient between the Gini coefficient and the coefficient of variation is 92 percent. The correlation coefficient between the Gini coefficient and the 90-10 percentile income difference is 91 percent. The variance is less correlated with these other measures because it is highly correlated with the mean level of income in the area. In general, these different measures give us a similar picture of which metropolitan areas within the U.S. are most unequal.

We are interested not only in the stability of income inequality between different measures, but also in the stability of income inequality over time. Figure 2 graphs the Gini coefficient for 242 metropolitan areas estimated from the 1980 Census against the Gini coefficient from the 2006 American Community Survey, the most recent data available. For comparison, we also plot the 45 degree line.

Overall, the correlation between the Gini coefficient in 1980 and the Gini coefficient in 2006 is .58, which suggests neither extreme permanence nor enormous change in the rankings. Places that had an unusually high level of income inequality in 1980 revert

7 334

slightly to mean and have relatively less inequality today. As the Gini coefficient in 1980 increases by .1, the growth in the Gini coefficient over the next 26 years falls by .03. Some of the impermanence (and mean reversion) surely reflects measurement error in the Gini coefficient.

The most striking fact in Figure 2 is that the points in the graph are above the line in all but one area (Ocala, Florida), which means that for almost all the MSAs the estimated Gini coefficient is much higher today than it was 26 years ago. Much of this surely reflects the real increase in inequality in this country that has been extensively documented (e.g. in Katz and Murphy, 1992, and subsequent literature). However, some of the seeming increase in Gini coefficients may reflect changing top codes, but when we look at other measures that are less subject to top-coding issues (i.e. the 90-10 differences) we continue to see large increases in inequality in almost all areas.

In 1980, the Gini coefficients ranged from .33 to .45. Wisconsin had the most equal metropolitan areas 25 years ago with the Appleton-Oshkosh-Neenah MSA’s Gini coefficient of .33, and Gainesville, Florida, was the most unequal metropolitan area with a Gini coefficient of .45. Two very poor Texas areas (Brownsville and McAllen) had the next highest levels of inequality. Wisconsin still has the country’s most equal metropolitan area in 2006 (Sheboygan with coefficient of .38), but even it is substantially less equal than the most egalitarian places were 25 years ago. New Haven-Bridgeport- Stamford, with its combination of inner-city poverty and hedge-fund entrepreneurs, is now the most unequal metropolitan area in the country with a Gini coefficient of .54. While the county of Manhattan is more unequal, there are no other metropolitan areas that are even close to that Connecticut area in income inequality. The next three most unequal areas are Gainesville, Florida; Athens, Georgia; and Tuscaloosa, Alabama. Inequality shows up both in America’s richest metropolitan areas, like New Haven, and in some of its poorer areas.

Generally, there is a negative association between area inequality and average incomes. For example, the Gini coefficient, coefficient of variation, the 90-10 percentile income

8 335

difference and the 75-25 percentile income difference are all are negatively associated with area income. Figure 3 shows the –14 percent correlation between the Gini coefficient and the logarithm of median family income.

However, this connection has been declining over time. Figure 4 shows the -59 percent correlation between the Gini coefficient and the logarithm of median family income in 1980. This correlation is far stronger than the -14 percent correlation for 2006 shown in Figure 3. 25 years ago almost all rich places were relatively equal, given the relative inequality of the United States. Today, some of America’s richest places are also among the most unequal. Some of this change may reflect changing top-coding, but it surely also reflects the enormous gains in wealth at the top end of the income distribution over the past 25 years in wealthy cities like San Francisco.

While the link between average income and inequality is becoming weaker, the link between area population and inequality is becoming stronger. In 1980, the raw correlation between population and the Gini coefficient was essentially zero (-.02). In 2000, the Gini coefficient’s 15 percent correlation with area population is shown in Figure 5. Regressions (1) and (2) in Table 2 show bivariate regressions where the Gini coefficient is regressed on contemporaneous income and population measures in 1980 and today. Between 1980 and 2000, the connection between average income and the Gini coefficient fell by more than 40 percent, and the connection between area population and the Gini coefficient increased by roughly the same percentage.

Does nominal income inequality imply inequality of real incomes or of consumption? Prices differ across metropolitan areas, but if prices were the same for every type of person in every area, then prices should not impact inequality, at least as measured by the coefficient of variation or the Gini coefficient. However, as suggested by Black, Kolesnikova and Taylor (2007), prices may be quite different for people at different places in the income distribution. New York may be much more expensive for a relatively rich person than it is for a relatively poor person. Indeed, the very fact that

9 336

poor people continue to live in New York suggests that the area may not be as expensive for them as average prices would indicate.

This issue could be addressed by developing different price indices for people at different levels of the income distribution in different metropolitan areas, but that is far beyond the scope of this paper. Instead, we have undertaken the far simpler task of asking about the inequality of consumption of one important good: housing. If places with more rich people are expensive places for the rich to live, then we should expect to see less inequality of housing consumption than inequality of income.

To calculate a housing consumption Gini coefficient, we must first calculate a measure of housing consumption for everyone in the U.S. We start with a national housing price regression, where the logarithm of housing price is regressed on the characteristics of every household. Because of the limited number of housing characteristic data available from the Census microsample, we instead use data for 46 of the largest metropolitan areas from the American Housing Survey Metropolitan Samples for 1998, 2002, 2003 and 2004. Housing characteristics include interior square footage, exterior square footage, the number of bathrooms, the number of bedrooms and several other features. We then use this regression to form a predicted housing price measure for every household. Essentially, we are using a hedonic regression to create a housing price index that enables us to aggregate across different housing characteristics.

We use this housing consumption measure to calculate a Gini coefficient of housing consumption for every metropolitan area. Figure 6 shows the 38 percent correlation between this housing consumption Gini coefficient and our income Gini coefficient. More unequal incomes also have more unequal housing consumption, but in general housing consumption inequality is much less than income inequality and housing consumption inequality is particularly below income inequality in places with large amounts of income inequality. The mean housing consumption Gini coefficient is 0.28, much lower than 0.45, which is the mean of the household income Gini for this subsample of metropolitan areas in 2000.

10 337

While some of this difference can be attributed to measurement error, since our hedonic regression omits many key housing attributes, this result still suggests that places with highly unequal income levels have less housing consumption inequality than one might expect. This fact does not prove that prices are largely offsetting incomes, but heterogeneity in local prices that impact rich and poor people differently may be important, and measuring these prices and their impact is yet another interesting topic for future research.

III. The Causes of Urban Inequality

The typical economic approach to earnings is to assume that they reflect the interaction of human capital and the returns to human capital (e.g. Katz and Murphy, 1992). Indeed, human capital is often defined as including everything that goes into earnings, in which case the relationship is essentially tautological. If human capital is reduced to being a scalar, h, then the wage associated with each value of h is i (hw ) where i represents each place. If the density of population in each area with human capital level h is i (hg ) , then the average earnings in a locality is equal to ii )()( dhhghw . The density of income ∫h

−1 −1 will be ii ywg ))(( where ii (( ywG )) denotes the cumulative distribution of income.

2 If i (hw ) α += β ii h , then the variance of wages within a place is equal to β ii hVar )( , where i (hVar ) is the variance of h within place i. The coefficient of variation is β i hVar )( , where hˆ is the mean of h within place i. The Gini coefficient is ˆ i i + βα hiii

1 − 21 1 −− ii )))((1( dyywG , and if h is distributed uniformly on the interval ∫y yˆi )( 3σβ ˆ ,5. hh ˆ +− 5. σσ then the Gini coefficient is 1− ii , which is a function of [ iiii ] ˆ 3()+ βα hiii both the distribution of skills and the returns to skill.

11 338

The inequality of after-tax, after-redistribution earnings will then also be affected by the progressivity of the tax rate and the state of the social safety net. We will turn to the heterogeneity in welfare payments later, but our primary focus is on the dispersion of before-tax earnings. We will begin by discussing the role that heterogeneous human capital plays in explaining the differences in inequality across space and the causes of that heterogeneous human capital. We will then turn to differential returns to human capital and governmental after-tax redistribution.

Human Capital Heterogeneity and Income Inequality Across Areas

To assess the role that human capital plays in explaining income inequality across areas, we will take two complementary approaches. First, we will simply regress the Gini coefficient on measures of human capital. Second, we will create Gini coefficients for each metropolitan area based on the observable measures of human capital and national wage regressions. Both measures are compromised by the fact that our measures of human capital are coarse. They capture only the years of formal schooling and years of experience. True human capital would include many more subtle factors, such as the quality of schooling and of experience.

In regression (3) of Table 2, we examine the relationship between the Gini coefficient in 2000 and two primary measures of human capital in the same year: the share of adults with college degrees and the share of adults who are high school graduates. We continue to control for area population and area income. Both of these variables are extremely significant and they increase the amount of variance explained (i.e. r-squared) from 15 percent to 49 percent. As such, more than one-third of the heterogeneity in income inequality across metropolitan areas can be explained by these two basic measures of human capital.

The coefficients are also large in magnitude. As the share of college graduates increases by 10 percent, the Gini coefficient rises by .031, a little more than one standard deviation.

12 339

As the share of high school graduates increases by 10 percent, the Gini coefficient drops by .018, about two-thirds of a standard deviation. While it may be unsurprising that even such crude proxies for heterogeneity in the human capital distribution can do so well at explaining the income distribution, this fact illustrates that much of inequality within an area reflects the heterogeneity of skills within that area.

One concern about results such as these is that perhaps the inequality of human capital within an area is itself an endogenous response to changes in the returns to skill. If places that have high returns to having a college degree attract people with college degrees, then controlling for the skill distribution in this way may also end up controlling for the returns to skill. After all, economic theory predicts that college graduates should go to places where the returns to being a college graduate are higher. Of course, this cannot explain why inequality is higher where there are more high school dropouts, but it is still worth taking the endogeneity of skills seriously.

One approach to this endogeneity is to look at long-standing historical skill patterns. We only have data on the share of the population with high school and college degrees going back to 1940. In regression (4), we show the results using those variables instead of contemporaneous college and high school graduation levels. We continue to control for contemporaneous income and population, but the results are unchanged if we remove those controls. Human capital levels in 1940 are still strongly correlated with inequality today. The overall r-squared declines to 32 percent, but these historical variables incrementally explain more than fifteen percent of the variation in the Gini coefficient.

The coefficient on the high school graduation rate in 1940 is quite close to the coefficient on 2000 high school graduation. The coefficient on the college graduation rate is more than three times higher using the older data. However, the variation in the college graduation rate is much smaller in 1940, so a one standard deviation increase in the college graduation rate has about the same effect when using 1940 or 2000 data. One way to understand why 1940 college graduation rates have such a strong impact on

13 340

modern inequality is that they strongly predict the growth in college graduation rates after that point.

In the fifth regression, we go back even further and look at 19th century measures of human capital. We do not have measures of human capital in the adult population, but we do have enrollment rates for high schools and colleges during this time period, which are found in historical Census data.3 Unfortunately, we lose over 70 metropolitan areas by using this historical data. The coefficients on these enrollment rates are not, therefore, particularly comparable with the coefficients on population-based skill measures in regressions (3) and (4). In regression (5), we find that the share of the population enrolled in college in 1850 is a quite solid predictor of income inequality today. High school enrollment rates also continue to negatively predict inequality.

In this case, the incremental r-squared created by those two variables when compared to regression (2) is quite modest (five percent). Still, we are impressed by the ability of 150-year-old educational variables, measuring something quite different from modern skill levels, to explain anything. The coefficient on college enrollment in 1850 is quite large, but again this needs to be considered together with the extremely small level of variation in this variable across space. These results continue to suggest to us that historical patterns of human capital play some role in explaining income inequality today.

In the sixth regression, we add two more historical measures of human capital: (1) the share of the population that is illiterate in 1850 and (2) the share of the population that was enslaved in that year. We interpret both measures as proxies for human capital deprivation. Learning to read is an obvious measure of human capital. Slaveowners often opposed education for their slaves. Indeed, during the century after emancipation, the former slave areas continued to provide particularly poor education for African- Americans. Both illiteracy and slavery in 1850 help predict inequality today. Adding

3 Haines, M.R., (2004) ICPSR study number #2896, Historical Demographic, Economic, and Social Data: The United States, 1790-2000.

14 341

these variables increases the r-squared of the regression to 26 percent, an additional six percent.

Another way of looking at the impact of human capital is to ask whether human capital in 1980 is associated with growth in inequality after that year. In Table 3 we regress the Gini coefficient in the year 2006 on the Gini coefficient in the year 1980 and other controls. We use the 2006 measure, rather than the 2000 measure, because it increases the time period of change by 30 percent. Regression (1) can be interpreted as a growth regression since we are asking about the determinants of inequality today, holding past inequality constant. The coefficient of .79 on the Gini coefficient in 1980 implies that there is mean reversion between 1980 and today, although this could be the result of measurement error. The result on income and population tell us that income inequality has been rising both in bigger areas and in richer areas.

In regression (2), we include our controls for human capital in 1980. Again, there is a strong positive association between college graduation rates in 1980 and inequality today. Places with more highly skilled people in 1980 have become more unequal over time, which presumably reflects both the rise in the returns to skill and the tendency of skilled people to move to more skilled areas, which we will discus later. Places with more college dropouts have also become more unequal over time. Not only do contemporaneous skill levels predict inequality, but inequality of skills in 1980 predicts an increase in income inequality since then.

We now turn to our second means of assessing the importance of skill distributions in explaining the inequality of income. In this approach, we calculate only the income inequality from males between the ages of 25 and 55. To keep sample sizes up, we look only at the 102 metropolitan areas with more than 500,000 people. We use only workers with positive earnings, and we use only labor market income. We calculate three Gini coefficients. First, we calculate the standard Gini coefficient using the earnings from these workers. Figure 7 shows the 74 percent correlation between this Gini coefficient and our household income Gini coefficient among these 102 metropolitan areas.

15 342

We then compare this Gini coefficient for male workers with Gini coefficients based entirely on the human capital of these workers, by which we mean age and years of schooling. To calculate these “human capital only” Gini coefficients, we use a nationwide earning regression to predict earnings for everyone in the sample. By using these predicted earnings rather than true earnings we can isolate the impact of the level of human capital in an area while abstracting away from the differential returns to schooling. We calculate the Gini coefficient based on these predicted earnings. Figure 8 shows a 57 percent correlation between the two Gini coefficients.

Our Gini coefficient based only on human capital explains about 33 percent of the variation in overall income inequality among working-age males. This is a considerable amount of explanatory power, but it still leaves plenty to be explained. Another way of looking at the data is to note that the average Gini coefficient calculated with human capital only is one half the size of the Gini coefficient calculated using true income. As such, human capital gets you some, but far from all, of the way towards explaining the amount of income inequality.

One reason why these human capital based measures are failing to more fully explain actual income inequality might be that our human capital measures are so coarse. Occupations may provide us with a richer means of measuring individual-level human capital. As such, we created a third Gini coefficient using the same sample but including occupation dummies in our wage regressions. These dummies are then used, along with years of schooling and age, to predict wages, and these predicted wages are used to calculate a local Gini coefficient. The mean of this occupation-based Gini coefficient is about half-way between the Gini coefficient with just education and age and the Gini coefficient of true income. If we accept that occupation is a measure of skills, then this measure goes much more of the way towards explaining income inequality across areas.

Figure 9 shows the 86 percent correlation between this occupation-based Gini coefficient and the true income Gini coefficient. The occupation-based index has the ability to

16 343

explain 73 percent of the variation in the true income Gini coefficient. There are two interpretations of this finding. First, occupation may be proxying for income more than human capital and therefore should not be used as a control in the regression. Second, occupation may indeed be a better measure of human capital. There is surely some truth in both views, but we can draw the conclusion from this section that heterogeneity in human capital across space can explain a considerable amount of the heterogeneity in income inequality across space.

The Causes of Human Capital Inequality

Why are the levels of human capital so different in different places? One theory is that current human capital levels reflect long-standing education policies formed over the last 200 years; Goldin and Katz (2008) provide a rich description of this history. Urban and economists who emphasize people’s location decisions will tend to focus on differences in economic productivity and differences in amenities that then motivate migration (e.g. Dahl, 2002). According to this view, areas that specialize in industries which are particularly good for low or high skill workers should then disproportionately attract those workers. Of course, this theory just pushes the puzzle one step backward. A more complete explanation for heterogeneity in skills would also explain why different industries are located in different places. In some cases, like cities with ports or coal mines, there are exogenous factors that explain industrial location, but observable variables tend to only explain a modest amount of industrial concentration (Ellison and Glaeser, 1999).

Alternatively, amenities may draw high-skill workers to a particular location. For example, if there is some amenity that is particularly desirable and in particularly short supply, then we would expect rich people to locate in places with that amenity. This can certainly explain why there are so many rich people in Paris (Brueckner, Thisse and Zenou, 1999) or on the Riviera. Alternatively, there can be other amenities, such as access to public transportation, which might draw poor people disproportionately to a given area.

17 344

While the economic framework that emphasizes rational location choice certainly has some ability to explain the distribution of skilled people across space, much of current skill patterns appear to be determined by long-standing historical skill patterns, as discussed above. For example, Figure 10 shows the 73 percent correlation between the share of adults with college degrees in the year 2000 and the share of adults with college degrees in the year 1940. The college share of the population in 1940 is able to explain more than 50 percent of the variation in the college share today, which suggests the enormous power of historical forces in shaping the skill composition of cities today.

We also run a regression that shows that skill growth is strongly predicted by the initial skill level. As the 1940 share of the adult population with college degrees increased by 5 percent, the growth rate in the share of the population with college degrees between 1940 and 2000 increased by 10 percent. Far from there being mean reversion in this variable, there has been a tendency of skill growth to be concentrated in places that began with more skills (Berry and Glaeser, 2005).

The relationship between today and the past is much weaker at the bottom end of the skill distribution. Figure 11 shows the 45 percent correlation between the share of the adult population who did not have a high school degree in 1940 and the share of the adult population without a high school degree in 2000. In this case, the 1940 variable can only explain one-fifth of the variation in the high school dropout rate today. The graph shows a number of places, such as Miami and McAllen, Texas, which have particularly large high school dropout rates today relative to their historical levels. Older variables, such as the percent of the population enslaved in 1850, can explain about one-tenth of the variation in the high school dropout rate today.

These outliers suggest that immigration, particularly from Latin America, is also associated with a heavy concentration of less skilled workers. Figure 12 shows the 57 percent correlation between the share of the population that is Hispanic and the share of the population without a high school degree in 2000. Together, the Hispanic share today

18 345

and the dropout rate in 1940 can explain 61 percent of the variation in the dropout rate today. The Hispanic share is particularly high in Florida, Texas and California, which are three states that were once part of the Spanish Empire and which are geographically close to Mexico and other countries in Latin America. In fact, the correlation between Hispanic share and MSAs latitude is -36 percent, which reinforces this point. Once again, history seems to play a large role in explaining the current skill distribution.

By contrast, the evidence supporting the importance of the traditional economic explanations of the location of talent is much weaker. For example, there is little evidence that highly skilled people have moved to areas with particularly pleasant temperatures. There is a no robust association between January temperature and the share of the population with college degrees. High July temperatures are associated with fewer college graduates, but even this effect is quite modest. July temperatures can explain only 6.5 percent of the variation in the share of the population with college degrees.

LeRoy and Sonstelie (1983) and Glaeser, Kahn and Rappaport (2008) provide theory and evidence supporting the view that less skilled people live in the centers of metropolitan areas because of access to public transportation. Cars are expensive, and poorer people prefer the time-intensive, lower-cost alternative of buses and subways. Can access to public transportation explain the location of less-skilled people across areas? No. There is virtually no correlation across metropolitan areas between the share of the population without high school degrees and the share of the population that takes public transportation. This absence of correlation is particularly surprising since the poor generally take public transit more, which should yield a positive relationship between less skilled people and public transit, even if the poor didn’t move across metropolitan areas in response to public transit access.

Alternatively, people of different skill levels may be drawn to particular areas because of skill-specific economic opportunities. Silicon Valley has a booming computer industry, and it attracts extremely highly skilled engineers. New York City attracts smart people to

19 346

work in finance. Certainly, there is a strong correlation between the skill level of an area and the skill orientation of the industries in the area. Using the 2000 Census, Glaeser and Gottlieb (2008) ranked industries by the share of the workers in that industry in the nation with a college degree. They then calculated the share of a metropolitan area’s employment that was in the top 25 percent industries ranked by human capital and in the bottom 25 percent of industries ranked by human capital.

Figure 13 shows the 79 percent correlation between this measure of high skilled industries and the share of the population with college degrees. Figure 14 shows 49 percent correlation between the share of the adult population without a college degree and the measure of low-skill industries. Certainly, there is a robust relationship between the skill orientation of the industries in an area and the skill distribution of the area. But which way does the causality run? Are skilled industries moving into an area because there are an abundance of skilled workers, or are skilled workers moving to areas because of skill-oriented industries?

While surely both phenomena occur, we think that the evidence supports the view that industries are responding to the area’s skill distribution more than the view that the skill distribution is responding to the area’s industries mix. For example, the share of the population with college degrees in 1940 can explain 35 percent of the variation in the skill mix of industries today. By contrast, the skill composition of the industries in the metropolitan area in 1980 can only explain seven percent of the variation in growth of the population with college degrees since that date. The complex two-sided nature of this relationship makes it difficult to accurately assess the direction of causality, but there are reasons to think that much of the industrial mix in the area is actually responding to the skill distribution.

One variable that seems more plausibly predetermined is the concentration in manufacturing during the first half of the 20th century. The location of factories does not seem likely to be particularly driven by the presence of highly skilled workers 100 years ago. Yet an industrial orientation, as late as 1950, is negatively correlated with the share

20 347

of the population with college degrees today, perhaps because those manufacturing cities tended not to reinvent themselves as centers of idea-oriented industries or perhaps because manufacturing employers were less disposed towards high schools earlier in the century (Goldin and Katz, 2008). As the share of the workforce in manufacturing in 1950 increases by 10 percent, the share of the population with college degrees drops by about 1 percent. This may explain why, as shown in Table 2, Regression (7), manufacturing in 1950 is negatively associated with inequality today.

These industrial measures are essentially proxies for differential returns to human capital across metropolitan areas. Using wage regressions, we are able to estimate such differential returns directly by running regressions of the form:

BA HS (1) WageLog )( MSA += βα MSA * BAGrad + β MSA * HSGrad + Other Controls

BA where α MSA is an area specific intercept, β MSA is an area specific return to having a

HS college degree and β MSA is an area specific return to having a high school diploma. This regression estimates a differential return to different levels of schooling for each metropolitan area. We estimate this regression only for prime age males, and include controls for experience. We focus only on those areas with more than 500,000 people so that the returns to schooling are estimated with reasonable precision.

BA Figure 15 shows the 24 percent correlation between our estimate of β MSA and the share of the adult population with college degrees in the 102 metropolitan areas in our sample with more than 500,000 people. One interpretation of this fact is that skilled people are moving to places where the skill levels are higher. A second interpretation is that agglomerations of skilled people raise the returns to skill. Any interpretation of this relationship is compromised further by the fact that an abundance of skilled people would normally reduce the returns to skill in a typical model of labor demand. Still, this does suggest that highly skilled people are living in places where the returns to skill are higher.

21 348

Differential Returns to Human Capital

The fact that the returns to capital differ across space can also potentially explain the inequality that we see across metropolitan areas. As the formula discussed above illustrates, pre-tax income inequality will reflect both differences in the distribution of skills and differences in the returns to those skills. Certainly, the framework predicts a strong link between places with higher returns to college and income inequality.

Figure 16 shows the 73 percent correlation between our estimated return to a college degree and the Gini coefficient across the 102 areas with more than 500,000 people. The measured return to a college degree is much better at explaining area inequality than the number of people with college degrees. Of course, these returns are directly based on the same income data that is being used to generate the Gini coefficient. Still, this finding seems to confirm the view that heterogeneity in returns to skill can help us to explain differences in income inequality across space.

To look at this further, we again calculate Gini coefficients for each metropolitan area using wage regressions. However, in this case, we allow the coefficients on skills to differ across metropolitan areas as shown in equation (1) above. We again run these regressions only for prime aged males. We then use these regressions to predict the amount of inequality in an area if the skill distribution of the area were the same as the skill distribution in the country as a whole. When we calculated Gini coefficients using wage regressions above, we were calculating local Gini coefficients based only on differences in the skill composition, holding the returns to skill constant across space. Now we calculate Gini coefficients holding the skill composition constant, but allowing the returns to skill to differ across space.

Figure 17 shows the 71 percent correlation between these predicted wage Gini coefficients and actual Gini coefficients in our sample of prime age males across metropolitan areas with more than 500,000 people. The relationship is tighter than it was when we looked at Gini coefficients that assumed a constant return to skill. Moreover,

22 349

this Gini coefficient holding the skill composition constant explains 50 percent of the variation in the actual Gini coefficient, whereas our constant return to skills Gini explained only 33 percent of the difference. We interpret these results as suggesting that differential measured returns to human capital can explain area-level income inequality somewhat better than differences in measured human capital.

One potential concern with interpreting these results is that measured returns to human capital may not be measuring higher returns to human capital, but instead measuring high levels of true human capital associated with each coarse category of observed human capital. For example, if people with college degrees in some areas went to higher quality schools or have had better work experience, then this would cause the measured return to a college education to increase, even if the true returns to human capital were constant across space. We have no way of dealing with this hypothesis, and we will continue referring to the measured returns to human capital as the returns to human capital, understanding that it can also reflect other things.

While differences in the returns to skill do seem to explain a significant amount of the differences in inequality, we do not know what explains differences in returns to skill across space. For example, the positive correlation between the share of the population with college degrees and returns to skill might suggest that being around other skilled people increases the returns to being skilled. Alternatively, Beaudry, Doms and Lewis (2006) suggest that places with abundant skilled workers invested in computerization, which then had the effect of raising the returns to skill. While we are certainly sympathetic to these interpretations, it is hard to distinguish between this view and the view that more skilled people are moving to areas where the returns to skill are higher.

Moreover, the share of the population with college degrees in 1940 does little to explain the returns to college today. If we thought that higher returns to skill reflected the power of agglomerations of skilled people, then an abundance of skills in 1940, which predicts skills today, should also predict higher returns to college today. This is not the case, as we find only a 5 percent correlation between the share of the population with a college

23 350

degree in 1940 and our measure of the return to a college degree. Instead, a higher return to college today is strongly associated with recent growth in the share of the population with college degrees. The correlation between our return to college measure and the change in the proportion of the adult population with college degrees is 26 percent. These facts support the idea that skilled people are moving to areas where the returns to skill are higher.

Other variables also do a relatively poor job of explaining the returns to skill. For example, our measure of skill intensive industries doesn’t explain the returns to skills. Looking at Figure 15 shows that some of the places with the highest returns to skill are usual suspects. The financial agglomeration in Southwestern Connecticut and the technology agglomeration in San Francisco Bay have very high returns to human capital. But there are also areas, like Houston and Birmingham, that are more of a surprise.4

Figure 18 illustrates the 34 percent correlation between our estimated returns to college and the share of workers in finance among the 102 cities in our sample with more than 500,000 people.5 Figure 19 shows the 27 percent correlation between the returns to college and the share of workers in the computer industry among the same sample of cities.6 This seems to support the results of Beaudry, Dom and Lewis (2006) who show a link between computerization and inequality. A related and interesting hypothesis is that wage inequality is linked to the former concentration of low skilled workers in routine tasks that have now been made obsolete (Autor and Dorn, 2007).

Still, we are much more confident that differences in the returns to skill can explain a significant amount of income inequality across metropolitan areas than we are in explaining why areas have such different returns to human capital. A number of recent papers, like Autor and Dorn (2007); Black, Kolesnikova and Taylor (2007); and Beaudry,

4 Like Black, Kolesnikova and Taylor (2007), we find a modest positive relationship between cost of living and returns to college. 5 “Finance” is defined using IPUMS 2000 Occupation Codes for the 5% sample at http://usa.ipums.org/usa/volii/00occup.shtml. Finance codes are 12, and 80-95. 6 “Computers” is also defined using IPUMS 2000 Occupation Codes for the 5% sample at http://usa.ipums.org/usa/volii/00occup.shtml. Computer codes are 11, 100-111, and 140.

24 351

Dom and Lewis (2006) have brought some understanding to this question, but it remains a pressing topic for future research.

IV. The Consequences of Urban Inequality

At the national level, income inequality has been linked to low levels of economic growth, perhaps because inequality leads to political strife (Persson and Tabellini, 1994; Alesina and Rodrik, 1994), but local inequality is not the same thing as national inequality. No one should be surprised if the political and economic effects of inequality are different at the local level. After all, local political outcomes are far more constrained by state constitutions and easy out-migration.

More generally, local inequality, as opposed to local poverty, is not necessarily a bad thing. If people of different income levels mix throughout the country, then local inequality will be higher than if people segregate into homogenous, stratified communities. A large number of studies suggest economic mixing, i.e. local inequality, benefits the less fortunate by giving them more successful role models (Wilson, 1987) or employers (Mazzolari and Ragusa, 2007). Others suggest that the wealthy develop empathy for the poor through spatial proximity (Glaeser, 1999). Egalitarians can simultaneously hope for policies that would reduce inequality at the national level, such as increasing the schooling levels for least fortunate, while opposing policies that would reduce local income inequality by moving rich people away from poor people.

Persson and Tabellini (1994) found a strong negative relationship between national income inequality and economic growth. Some facts about urban growth are quite similar to facts about country growth. For example, schooling predicts growth at both the country and the city level. Does the connection between inequality and growth carry over to the metropolitan level?

In Table 4, we look at the relationship between inequality and growth across our sample of metropolitan areas. We use 1980 as our start date and look at the growth of both

25 352

income and population after that year. While country-level regressions typically look only at income growth, city level growth regressions look at population and income (and sometimes housing values as well), since increases in productivity should show up both in higher wages and in more people.

The first regression of Table 4 shows that the raw relationship between income inequality and local area growth is positive. Places with more inequality have been gaining population. However, as the second regression in Table 4 shows, this result is not robust to including a number of other city level controls, such as human capital variables and January temperature. With these controls, inequality has a significant negative effect on area population growth. The next two regressions in Table 4 show the relationship between inequality and area income growth. After including area level controls, inequality has a significant negative impact on income growth. The coefficients here should be interpreted while keeping in mind the relatively low level of variation in the 1980 Gini variable. Going all the way from the bottom of the inequality distribution at 0.33, to the top of the distribution at .45, would cause population growth to fall by about 1 standard deviation.

These results do suggest that income inequality is only negatively correlated with area growth once we control for skills. Increases in the skill distribution that make a place more unequal by increasing the share of highly educated citizens are associated with increased, not decreased, growth. However, growth of both income and population was lower in places where the income distribution is particularly unequal, holding skills constant.

A second adverse consequence of inequality at the country level is the connection between inequality and crime (Fajnzylber, Lederman and Lloayza, 2002). These results have also been found at the metropolitan area level (Daly, Wilson and Vasdev, 2001). We duplicate them here.

26 353

In regression (5) of Table 4, we show the strong positive relationship between income inequality and murder rates across metropolitan areas. We focus on murder rates because they are the most serious crime outcome and the outcome that is least likely to be impacted by reporting differences across areas. The 35 percent correlation, shown in Figure 20 is quite strong. Regression (6) of Table 4 shows that the inequality-crime relationship is robust to a number of other controls.

Why do murder rates increase with inequality? One view is that inequality is just proxying for poverty, but both at the country and city level, the impact of inequality on crime survives controls for the mean income level and the poverty rate. A second explanation is that inequality leads to less focus on providing community-wide public goods, like policing. A third explanation is that inequality breeds resentment which then shows up in higher murder rates.

All of these explanations remain speculation, but there is some evidence that links unhappiness to envying richer neighbors. Luttmer (2005) looks at the self-reported happiness of individuals as a function of the wealth of their neighbors. He finds that people who have richer peers are more likely to say that they are unhappy. The existence of envy can, under some conditions, suggest that sorting by income is preferable to highly unequal areas.7

In Figure 21, we show the -47 percent correlation between the Gini coefficient and the average self-reported happiness in the metropolitan area taken from the General Social Survey. These happiness data span the last 25 years, and they represent the share of people who say that they are very happy. Inequality can explain 22 percent of the variation in this unhappiness measure, and this result is robust to a reasonable number of other controls such as average area income and population size.

V. Government Policy

7 We have also looked at whether there is a correlation between income inequality and racial segregation, using dissimilarity measures of segregation (see Cutler, Glaeser and Vigdor, 1999 for details of the measure). We find no evidence of any such connection.

27 354

There are two ways in which government policy interacts with the study of urban inequality. First, government policy may itself be a cause of that inequality. Second, if policy makers seek to reduce local inequality, then the study of that inequality may improve the quality of decision-making. We start with a discussion of the role that governmental actions might have on the level of inequality and then turn to a discussion of potential policy implications.

Government Policies and Local Inequality

There are at least three channels through which government policy might impact the degree of income inequality across space. First, education is largely a government service, and government policies towards education could either widen or narrow the distribution of skills within an area. Second, government policies can also impact migration in ways that might increase or decrease the skill distribution. Third, the government engages in taxes and redistribution, which would impact the after-tax income distribution. Some of these policies are explicitly intended to impact inequality, and other policies are intended to achieve different results but still could end up changing the level of local inequality.

Investment in school certainly appears to impact the distribution of skills within an area. For example, Moretti (2004) shows that the presence of a land-grant college in a metropolitan area prior to 1940 is positively correlated with the skill level of the area today. When we regress the area Gini coefficient on Moretti’s land-grant college indicator variable, controlling for area population, income and the share of adults who are high school dropouts, we find a positive, but statistically insignificant, impact of land grant colleges on inequality. This effect disappears when we control for share of adults with college degrees, which implies that this variable (weakly) increases inequality because it increases the share of more skilled people.

28 355

Conversely, we find a modest negative relationship between current high school enrollment rates and inequality when we control for area income and area population. This result may reflect the tendency of poverty to lead to low enrollment rates or the tendency of middle income people to move to areas with fewer dropouts or the ability of high school graduation to reduce inequality. We will not try to distinguish these hypotheses but just point out that correlations of this form suggest that education policy can surely impact inequality, both by its direct effect on the skill distribution and by shifting migration patterns.

There is a long economic literature that suggests that different local level government policies have the ability to induce selective migration. For example, Borjas (1999) argues that heterogeneity in welfare policies across space has had a huge impact on the location patterns of less skilled immigrants, and especially their tendency to locate in California. Blank (1988) also found that higher welfare levels impact the location decision of unmarried women with children. This type of effect can explain the poverty of East St. Louis, which traditionally had higher welfare payments because it lies on the Illinois side of the Mississippi River within the St. Louis metropolitan area.

Less work has been done on the impact of redistribution on the location decisions of the rich, but what evidence does exist supports the view that the wealthy are quite mobile and respond to attempts at redistribution. Certainly, within metropolitan areas, there are good reasons to think that the rich are sensitive to local tax rates and the bundle of local public goods. The strong tendency of richer people to live outside of city borders suggests that they are voting with their feet within certain areas. Haughwout et al. (2004) argue that these migration tendencies are quite strong and can mean that areas can actually lose revenue by raising taxes. Feldstein and Wrobal (1998) argue that the migration elasticities are so strong that states cannot effectively redistribute income at all. This type of result supports the view that local governments can affect local inequality by moving people more than they can by classical redistribution.

29 356

The connection between government policies and local income inequality is in many ways an understudied topic. We certainly know something about the impact of education, and we know much about migration responses to local policy differences, but we do not understand the full contribution that government has played in making some places more or less equal. For example, local land-use controls that prevent housing for lower income people can create less inequality within an area, but we do not know how empirically important this might be.

Local Governments and Local Inequality

Localities do have tools with which they could reduce local income inequality, but it is not obvious that such tools would enhance welfare. For example, the preceding analysis suggested that much of the heterogeneity in income inequality across metropolitan areas was associated with differences in returns to skill. Localities could equalize local incomes by reducing the returns to skill through more redistributive taxation. Redistribution would both directly reduce inequality, and is likely to also reduce inequality by inducing wealthier people to leave the area. However, few localities would actually find it attractive to increase equality by getting rid of the biggest tax-payers. While this migration effect might reduce inequality, if it eliminated the richest people in the city, there are many reasons to think that it would also hurt the area’s economy.

The returns to skill were not closely tied to industrial mix, so trying to attract particular industries doesn’t seem likely to lead to significant changes in inequality. Moreover, the historical track record of local industrial policy is decidedly mixed. A long tradition of urban analysis suggests that localities have a very limited ability to make society more equal (Peterson, 1981). The ability of wealthy people to flee is just too great.

Greater welfare gains would seem to be associated with policies that enhanced the skills of the less fortunate. Improvements in school districts and reductions in the size of the criminal sector could have two possible benefits. First, they might increase the skill levels at the bottom end of the income distribution. Second, they might attract middle-

30 357

income people into the area. Local policies that strengthened the bottom of the income distribution without targeting the top of the income distribution seem most likely to reduce income inequality without creating other problems.

However, while such policies might well be beneficial, local governments have again only a limited ability to make the nation-wide skill distribution more equal. Areas with many poor parents have fewer resources with which to educate their children. These places have lower tax revenues, holding everything else constant, but they also have less parental human capital on which to draw. The long-noted power of peers means that places with lower initial skill distributions inevitably have difficulty creating first rate public schools.

National Governments and Inequality

Even if one accepts egalitarian ideas that inequality is itself a bad thing, it certainly does not follow that reducing local inequality is clearly desirable. Would it be sensible for national government policies to artificially segregate rich people into some cities and poor people into others? Such segregationist policies would increase equality at the lower level, but it is hard to see how they would increase local welfare levels. Some of our regressions have suggested that localities will grow more quickly and have less crime if they are more equal, but even these results must be treated gingerly. National policies that created equality by removing high skill, high earnings workers from power areas would also be likely to reduce the economic performance of those areas. Policies that reduce the numbers of poor people by eliminating urban attributes that attract those people, like low-cost housing, would also have negative consequences, most notably the destruction of a valuable asset that is providing some benefit for the least fortunate members of society.

At the national level, egalitarianism suggests simple, non-spatial policies, such as classic income redistribution and policies that support human capital accumulation among the least fortunate. National policies can also reduce income inequality through

31 358

redistributive taxation. Those policies involve costs and benefits, and their connection with cities and localities is relatively modest, since local attempts at redistribution are likely to create emigration of the wealthy.

Long term reductions in inequality are most likely to be achieved through a more egalitarian distribution of human capital. Naturally, changes in the distribution of human capital will take years, if not decades. Those changes will also involve the often uncomfortable cooperation of national and local governments. Attempts to reduce inequality by changing the skill distribution must considerably involve localities, given our current decentralized schooling system. Yet, localities rarely have the resources to significantly upgrade their schools on their own.

The current structure of local public schooling creates incentives for middle income people to leave big cities to get better schools for their children. The poor and the very rich, who send their children to private schools, remain. There could be welfare gains from an education system that kept the advantages of choice and competition that are associated with the current system but that also reduced the incentive for middle class parents to leave big cities.

This fact is the great challenge facing attempts to reduce inequality through schooling. Our schools are local and localities have a great deal of trouble dealing with inequality. Poor places have fewer resources to allocate to their schools. Yet if there is going to be a more equal education distribution, then their schools must be improved. Creating equality in human capital requires the difficult cooperation of national level education policy, and schools that often operate at a very local level.

VI. Conclusion

In this essay, we have reviewed the economic causes of metropolitan-area income inequality. Differences in income inequality across areas can be explained well by both differences in the skill distribution and differences in the returns to skill. If anything,

32 359

differences in the returns to skill appear to be more important in explaining the variation in the Gini coefficient across American metropolitan areas. Differences in the skill distribution can be well explained by historical tendencies towards having more skilled people and by immigration patterns. Differences in the returns to skill are far more difficult to explain, but today the returns to a college degree are highest in areas that specialize in finance or computing.

There are some negative correlates of area-level inequality. More unequal places have higher murder rates, and people say that they are less happy. More unequal places grow more slowly, at least once we control for the skill distribution in an area. The raw correlation between area-level inequality and population growth is positive.

Area-level income inequality does not create the same policy implications as national income inequality. At the nation level, an egalitarian, Rawlsian social welfare function implies the need to reduce income inequality. However, egalitarianism does not provide the same implications about local inequality. Shuffling people across the country in a way that creates more homogeneity at the local level would not seem like a natural means of increasing social welfare given standard social welfare functions. Instead, such functions would instead push towards a focus on policies like human capital development that would promote equality nationwide.

We concluded by noting that localities are poorly poised to reduce inequality on their own. Any attempt at local redistribution is likely to lead to out-migration of the wealthy. Poor localities don’t have the resources to improve failing schools.

However, if national policies are going to try to reduce inequality by making the distribution of human capital more equal, then inevitably localities must be involved. Schools are run at the local level. The combination of national resources and local operation seems most likely to improve the quality of the poorly performing schools. Unfortunately, bringing together such different levels of government is inevitably quite difficult. Moreover, the strong correlation between human capital today and human

33 360

capital more than fifty years ago suggests that any change will not happen overnight.

34 361

References Alesina, A., Glaeser, E. L., 2004. Fighting Poverty in the U.S. and Europe: A World of Difference. Oxford University Press, New York, NY.

Alesina, A., Rodrik, D., 1994. “Distributive Politics and Economic Growth.” Quarterly Journal of Economics 109(2), 465-490.

Autor, D., Dorn, D., 2007. Inequality and Specialization: The Growth of Low-Skill Service Jobs in the United States, MIT mimeograph.

Beaudry, P., Doms, M. E., Lewis, E. G., 2006. Endogenous Skill Bias in Technology Adoption: City-Level Evidence from the IT Revolution. NBER working paper.

Berry, C., Glaeser, E. L., 2005. “The Divergence of Human Capital Levels Across Cities.” Papers in Regional Science 84(3), 407-444.

Black, D., Kolesnikova, N., Taylor, L. J., 2007. Earnings Functions When Wages and Prices Vary by Location. Federal Reserve Bank of St. Louis working papers.

Blank, R. M., 1988. “The Effect of Welfare and Wage Levels on the Location Decisions of Female-headed Households.” Journal of Urban Economics 24(2), 186-211.

Borjas, G. J., 1999. “Immigration and Welfare Magnets.” Journal of Labor Economics 17(4), 607-637.

Brueckner, J.K., Thisse, J-F., Zenou, Y., 1999. “Why is central Paris rich and downtown Detroit poor? An amenity-based theory.” European Economic Review 43, 91-107.

Cutler, D., Glaeser, E. L., Vigdor, J., 1999. “The Rise and Decline of the American Ghetto.” Journal of Political Economy 107, 455-506.

Dahl, G., 2002. “Mobility and the Return to Education: Testing a Roy Model with Multiple Markets.” Econometrica 70(6), 2367-2420.

Daly, M., Wilson, M., Vasdev, S., 2001. “Income inequality and homicide rates in Canada and the United States.” Canadian Journal of Criminology, 219-236.

Ellison, G., Glaeser, E. L., 1999. “The Geographic Concentration of Industry: Does Natural Advantage Explain Agglomeration?” American Economic Review Papers and Proceedings 89(2), 311-316.

Fajnzylber, P., Lederman, D., Loayza, N., 2002. “Inequality and Violent Crime.” Journal of Law and Economics 45(1), 1-40.

35 362

Feldstein, M., Wrobel, M.V., 1998. “Can State Taxes Redistribute Income?” Journal of Public Economics 68(3), 369-396

Glaeser, E. L., 1999. “The Future of Urban Research: Nonmarket Interactions.” Brookings-Wharton Papers on Urban Affairs, 101-149.

Glaeser, E. L., Gottlieb, J. D., 2008. “The Economics of Place-Making Policies.” Brookings Papers on Economic Activity.

Glaeser, E. L., Kahn, M. E., Rappaport, J., 2008. “Why do the Poor Live in Cities? The role of Public Transportation.” Journal of Urban Economics 63(1), 1-24.

Goldin, C. and L. Katz (2008) The Race between Education and Technology. Cambridge: Harvard University Press.

Haines, M. R., 2004. “Historical, Demographic, Economic, and Social Data: The United States, 1790-2000.” Inter-University Consortium for Political and Social Data, http://www.icpsr.umich.edu/cocoon/ICPSR/STUDY/02896.xml.

Haughwout, A., Inman, R., Craig, S., Luce, T., 2004. “Local Revenue Hills: Evidence from Four U.S. Cities.” The Review of Economics and Statistics 86(2), 570-585.

Katz, L., Murphy, K., 1992. “Changes in Relative Wages, 1963-1987: Supply and Demand Factors.” Quarterly Journal of Economics 107(1), 35-78.

LeRoy, S. F., Sonstelie, J., “1983. Paradise Lost and Regained: Transportation Innovation, Income and Residential Location.” Journal of Urban Economics 13, 67- 89.

Luttmer, E. F. P., 2005. “Neighbors as Negatives: Relative Earnings and Well-Being.” Quarterly Journal of Economics 120(3), 963-1002

Mazzolari, F., Ragusa, G., 2007. “Spillovers from High-Skill Consumption to Low-Skill Labor Markets.” IZA Discussion Paper 3048.

Moretti, E., 2004. “Estimating the Social Return to Higher Education: Evidence From Cross-Sectional and Longitudinal Data.” Journal of Econometrics 121(1-2), 175-212.

Peterson, P. 1981. City Limits. Chicago: University of Chicago Press.

Persson, T., Tabellini, G., 1994. “Is Inequality Harmful for Growth?” American Economic Review 84(3), 600-621.

Ruggles, S., Sobek, M., Alexander, T., Fitch, C. A., Goeken, R., Hall, P. K., King, M., Ronnander, 2008. Integrated Public Use Microdata Series, http://usa.ipums.org/usa

36 363

Wheeler, C., 2005. “Cities, Skills and Inequality.” Growth and Change 36(3), 329-353.

Wilson, W. J., 1987. The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy. Chicago University Press, Chicago, IL.

37 364

Table 1 Table of Correlations Between Income Inequality Measures for 2000

Difference of Difference of the Log of the Log of Coefficient of Income for Income for the Variance of the Variation of the 90th and 75th and the Gini Household Household the 10th 25th Log of Median Log of Coefficient Income Income Percentiles Percentiles Family Income Population Gini Coefficient 1

Variance of Household Income 0.36 1

Coefficient of the Variation of 0.92 0.20 1 Household Income

Difference of the Log of Income for the 90th and the 10th 0.91 0.29 0.74 1 Percentiles

Difference of the Log of Income for the 75th and the 25th 0.86 0.13 0.72 0.93 1 Percentiles

Log of Median Family Income -0.25 0.68 -0.44 -0.17 -0.28 1

Log of Population 0.15 0.57 0.04 0.12 0.00 0.44 1

Source: The Gini coefficient, variance of household income, the coefficient of the variation of household income, the difference of the log of income for the 90th and 10th percentiles and the difference of the log of income for the 75th and 25th percentiles are calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org. Median family income and population are from the 2000 Census.

38 365

Table 2 Causes of Urban Inequality

1980 Gini Coefficient 2000 Gini Coefficient (1) (2) (3) (4) (5) (6) (7)

Ln(1980 Population) 0.0055 [0.0012]** Ln(1980 Median Family Income) -0.1078 [0.0084]** Ln(2000 Population) 0.0081 0.005 0.0075 0.0085 0.0078 0.008 [0.0016]** [0.0013]** [0.0014]** [0.0019]** [0.0018]** [0.0015]** Ln(2000 Med. Family Income) -0.0625 -0.1054 -0.0773 -0.0602 -0.0392 -0.0525 [0.0096]** [0.0112]** [0.0091]** [0.0131]** [0.0139]** [0.0096]** 2000 Pct. Of 25+ Pop. With BA 0.3122 [0.0233]** 2000 Pct. Of 25+ Pop. With HS -0.1842 [0.0257]** 1940 Pct. Of 25+ Pop. With BA 1.1176 [0.1337]** 1940 Pct. Of 25+ Pop. With HS -0.1906 [0.0329]** 1850 Pct. Of Pop. Enrolled in College 3.0137 2.4933 [0.8351]** [0.8285]** 1850 Pct. Of Pop. Enrolled in HS -0.0499 -0.0129 [0.0173]** [0.0194] 1850 Illiteracy Rate 0.061 [0.0327] 1850 Pct. Of Pop. Enslaved 0.0372 [0.0102]** Share of labor force in manufacturing, 1950 -0.0447 [0.0109]** Constant 1.3835 1.013 1.5937 1.1751 0.9877 0.7564 0.9174 [0.0782]** [0.0960]** [0.1045]** [0.0917]** [0.1311]** [0.1411]** [0.0960]**

Observations 258 282 282 281 208 208 281 R-squared 0.39 0.15 0.49 0.32 0.2 0.26 0.19

Notes: Standard errors in brackets. * significant at 5%; ** significant at 1%

Source: The Gini coefficients are calculated using the 5% Integrated Public Use Microdata Series for 1980 and 2000, at usa.ipums.org. Other 1980 variables are from the 1980 Census, and other 2000 variables are from the 2000 Census. All other variables are from Haines, M.R., ICPSR study number 2896, Historical, Demographic, Economic, and Social Data: The United States, 1790-2000.

39 366

Table 3 Changes in Inequality over Time

2006 Gini Coefficient (1) (2)

1980 Gini Coefficient 0.7934 0.6189 [0.0734]** [0.0781]** Ln(1980 Population) 0.0049 0.0037 [0.0014]** [0.0014]** Ln(1980 Median Family Income) 0.0293 0.0359 [0.0126]* [0.0143]* 1980 Pct. Of 25+ Pop. With BA 0.1773 [0.0371]** 1980 Pct. Of 25+ Pop. With HS -0.1366 [0.0246]** Constant -0.2165 -0.1352 [0.1367] [0.1508]

Observations 242 242 R-squared 0.42 0.49

Notes: Standard errors in brackets. * significant at 5%; ** significant at 1%

Source: The Gini coefficient for 2006 is from the American Community Survey, 2006. The Gini coefficient for 1980 is calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 1980, at usa.ipums.org. Other 1980 variables are from the 1980 Census.

40 367

Table 4 Consequences of Urban Inequality

Population Growth 1980-2000 Income Growth 1980-2000 Murder Rate per 100,000 (1) (2) (3) (4) (5) (6)

1980 Gini Coefficient 1.0301 -2.0335 -0.264 -1.3893 50.3572 55.3637 [0.7147] [0.6512]** [0.3524] [0.3635]** [14.5159]** [17.1951]** Ln(1980 Population) 0.0329 0.017 0.0255 0.0275 1.1389 0.9193 [0.0141]* [0.0112] [0.0069]** [0.0062]** [0.3450]** [0.3282]** Ln(1980 Median Family Income) -0.2484 -0.6385 -0.1137 -0.4195 0.8419 8.9701 [0.1228]* [0.1141]** [0.0606] [0.0637]** [2.6157] [2.9379]** 1980 Pct. Of 25+ Pop. With BA 1.0094 1.2358 -12.388 [0.2995]** [0.1671]** [7.1261] 1980 Pct. Of 25+ Pop. With HS 0.8556 -0.0066 -9.5897 [0.1957]** [0.1092] [4.9076] 1994 Mean Jan. Temp. 0.0084 -0.0003 0.0316 [0.0009]** [0.0005] [0.0269] Constant 1.8515 6.0646 1.814 5.0625 -36.9898 -109.0205 [1.3318] [1.1903]** [0.6567]** [0.6644]** [29.0251] [31.1609]**

Observations 258 258 258 258 120 120 R-squared 0.05 0.4 0.04 0.18 0.2 0.29

Notes: Standard errors in brackets. * significant at 5%; ** significant at 1%

Source: The Gini coefficient for 1980 is calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 1980, at usa.ipums.org. Other 1980 variables are from the 1980 Census. January temperatures are from the 1994 County and City Data Book from the U.S. Census.

41 368

Figure 1: Relationship Between the Gini Coefficient and Log Population Density, 2006 .6 New York NY

Orleans LA .55 Fairfield CT District of ColumbiaDC Westchester NY Charleston SC Fulton GA Essex NJ Clarke GA Richmond VA Lafayette LAHamiltonDenver OH CO Suffolk MA .5 ChathamMarinJeffersonNew CA Hanover GALynchburg AL NC VA San FranciscoKings CA NY CameronPulaskiEast TXOklahoma TravisBatonARDavidsonShelby TXRougeMercer OK TN TN DallasLA NJ TXSt.Cook LouisPhiladelphia IL MOHudson NJPA CabellHidalgoSullivanPalm WV TX TNBeach FL St.LosHarris Louis CuyahogaAngeles TX MO CA OH Bronx NY MobileNuecesGreenvilleHamiltonRichmond ALEl TXMuscogee Paso MecklenburgSCTulsa TN GATXLucasAllegheny OK GA OH NC PA Baltimore MD BoulderWashtenawLeonVolusiaCampbellSarasotaMaVanderburgh FLho COFLKnoxningFayette MI FLKY TNOH Hennepin KYIN PinellasBergen MNUnion FL NJ NJ SantaSt.Guilford HampdenBibbJosephForsythHillsborough Cruz GA MultnomahNCBrowardCAINSan NC MANorfolkFranklin Mateo FL FL MA ORCA OH Alexandria VA CapeKalamazoo MayLeeDurham KingNJFL MIBexar DouglasOaklandWA NCJeffersonLake RoanokeTX PassaicIL NEMIOrange KY VA NJ CA WashingtonScottInghamArapahoe IA TNNewport OrangeMIContraErieSantaDuvalHartford CO EssexNYMiddlesex RI Clara FLCostaFL CT MADelaware CA CAWayne MA PAMI MaricopaManateeGreeneLackawannaRichlandEscambiaBernalilloOnondagaGalvestonSan MOAZ FL DiegoSC FLNM PA SummitNY TX CATarrantAlamedaMarion DeKalbOHRamsey TX CA INNorfolkNassau GA MNArlington VA NY VA .45 GreggGastonChesterStark TX JacksonNCMorris MonroeNewPAOHMontgomeryProvidenceJeffersonRockland Haven NJ MONY CTLA Milwaukee NY RIPA WI MadisonLuzerneDaneErieSedgwickAtlanticBrevard PA WI ALPAKent NJ Lehigh FLMontgomeryKSSeminole MonmouthMIMontgomery PA FL OHNJ MD WinnebagoRockHarrisonCatawbaOzaukeeFortVentura WinnebagoIsland BendCollin PlymouthMS NCWI Polk CA WIILTXSomersetNewBristol TXHenrico IA IL MACastle MACobb NJVACamden DE GA NJ WashingtonSanCumberlandHallWorcesterHamilton JoaquinPasco GAGeneseeWake FLRI CAINSacramento BucksMANCLakeBaltimore NC MI INPA MDDuPage CA IL WestmorelandClevelandMadisonBooneElkhartLancasterSchenectadyWaukeshaAllenLorain KY ButlerILIN OK WyandotteJohnson INPA OHPA OH HonoluluWI NY KS KS HI Richmond NYQueens NY Gini Coefficient, 2006 Coefficient, Gini CherokeeSt.St. Lucie Clair FLGA ILOceanKaneKenton ILNJSuffolk KY NY NewRoanokeCabarrusTrumbullClermontBeaver DauphinBarnstableLondonDentonRacineJefferson VAOHNC PAOH CTTXPA WISaltKent MA CO Lake RINewport FairfaxUT News VA VA CumberlandHillsboroughPorterNiagaraPierceBrownWashingtonKenoshaNorthamptonClark IN WA WINY PAWA NHLake WI OR OHPA RockwallDutchessGreeneClarkBerksSolano OH OHTX NY PA AnneCA Arundel MD .4 MuskegonRockinghamWarrenFloydKitsapGloucester MIOH IN NH WAGwinnettMacomb NJ Portsmouth GA MI VA TollandWeberPutnamOrangeOttawaBurlington UTCT NYMIDavis NJVirginia UT Middlesex Beach VA NJ JohnsonMiddlesexMcHenryDouglasChesapeakeYork PAIN CTGAILHoward VA MD LoudounMedinaIslandWashingtonFayetteRockdale WA VAOHWill GA IL GA MN Hampton VA LebanonCalvertHenrySt.Clay CharlesDakota MDGAPA MO PrinceMO MN George's MD CarroHarfordChesterfieldll PrinceMD MD William VAClayton VA GA ForsythAnoka GA MN StaffordSarpy VA NE .35 -2 0 2 4 6 Log of Population Density

Source: 2006 American Community Survey

42 369

Figure 2: Gini Coefficient in 2006 and Gini Coefficent in 1980

.55 New Haven-

Tuscaloosa Athens, GA Gainesvill

.5 Wichita Fa New York, Brownsvill Trenton, N McAllen-Ed

San Jose, .45

Ocala, FL

Janesville .4 Gini Coefficient 2006 Appleton-O Wausau, WI York, PA Sheboygan, .35

.3 .35 .4 .45 Gini Coefficient 1980

Notes: The line shown is the forty five degree line, not a fitted regression. Only some datapoints are labeled with their MSA names to aid readability. Source: 1980 Gini coefficients are calculated from the 5% Integrated Public Use Microdata Series (IPUMS) for 1980, at usa.ipums.org. 2006 Gini coefficients are from the 2006 American Community Survey.

Source: 2006 American Community Survey

43 370

Figure 3: Relationship Between Gini Coefficient and Log Median Family Income, 2006

.55 Bridgeport

Sebastian- TuscaloosaMidland, T Athens-ClaGainesvill

.5 Lafayette, Naples-Mar Hot SpringWichita CollegeFa St New York-N Brownsvill Dothan,Morgantown AL McAllen-Ed Pine BluffColumbus,Miami-FortBrunswick,Auburn-OpeSanta Fe, Trenton-Ew Albany,Monroe, GreenvilleGA LABloomingtoDurham,SantaLos Angele BarbNC El Paso, T AlexandriaMobile,CorpusHattiesburTyler, ALChrJackson, TXCharlestonBirminghamHouston-Su M Charlottes Kingsport-HuntingtonShreveportLubbock,Deltona-DaLakeMemphis,BowlingSavannah, Charl T Gr T Iowa City, AnnBoulder, Arbor, C Laredo, TX Sumter,Jackson, SCBeaumont-PMissoula,GreenvilleSiouxOwensboro,Baton T CityPort Roug StateSt.San L Coll Luis O San Franci Cleveland,RockyMuncie, MounChattanoogPascagoula OklahomaINWilmingtonSarasota-BToledo,CapeNewCharlotte-Dallas-For C OrleanOceanCoral OHChicago-Na CityPhiladelphSantaBoston-Cam Cruz Las CrucesValdosta,Madera,Waco,Pueblo,Longview,SanMacon, Texarkana,CATXTulsa, AngeloCOFlagstaff,Tampa-St. LittleGAKnoxville, PittsburghOKCleveland-Tallahasse RocNashville-CincinnatiAustin-RouIthaca,Napa, NY CA JohnsonRome,Visalia-PoAnniston-OFresno, Ci GAAsheville,BlacksburgRedding,Amarillo,Augusta-RiSanTucson,Greensboro MontgomeryCA AntoniSouthEvansvilleKalamazoo- AZCLexington-SpringfielColumbus, Bend San Jose-S CharlestonCumberlandBakersfielDecatur,FarmingtonSpartanburLynchburg,HarrisonbuLafayette,Winston-Sa A BellinghamJacksonvilPhoenix-MeSalinas,St.Fargo,Milwaukee- Detroit-WaLouis, SanND- CFort Diego- Colli .45 Merced,MorristownMyrtleParkersbur CA BeaModesto,MountOrlando-KiIndianapol CVernAtlanta-SaDenver-Aur El Centro,Florence,YoungstownAbilene,FayettevilHouma-BayoFlorence-MCanton-MasNiles-BentBattleEugene-SprAlbuquerqu TSyracuse,Buffalo-NiCreColumbia,Champaign-Akron, OH New Haven- Danville,Wheeling,FortYakima, Jonesboro,SmithHickory-LePrescott,Bangor,Eri WAScranton--eChico,Waterloo-C, Decatur,PAMELouisvilleSalisbury, CAPeoria,LasHuntsvilleSpringfielFort IPortland-V Vegas-Providence WaltoAmes, IL SantaSeattle-Ta Baltimore-IAHartford-W Rosa Yuma,Gadsden, AZGoldsboro, ABurlingtonVictoria,Saginaw-SaYubaGrandJackson,Spokane,Davenport-Wichita,Dubuque,Columbia,Kennewick- City,Palm JuncRochester,Oshkosh-Ne AtlanticM W Bay-MK I Sacramento CRaleigh-CaMadison, W Washington Joplin,Weirton-StJohnstown, MOGulfport-BSpringfielWilliamspoElmira,Dover,Pensacola- NYLima,GrandDuluth,Dayton,Rockford,Roanoke, DE OH Rapi MNOmaha-Coun DesKansasOHRichmond, V Moines Cit Oxnard-Tho Lawton,Danville,Medford,ClarksvillPocatello, OKFlint,Gainesvill OBinghamton MICasper,Stockton,Lawrence,PittsfieldGreeley,Olympia,Kingston, Reno-SparkWYBloomingto C WHonolulu,Worcester, TerreOdessa,Ocala,FayettevilCoeurUtica-Rome PuntaHautKilleen-TeBayPanamaKokomo, FL Elkhart-GoTXCity, d'AlGordEauBoise CitRiverside- Bismarck,INHagerstownLancaster,La SiouxClaireAllentown- CityCrosse,Corvallis, Fall Anchorage,Minneapoli Lakeland,Altoona,GreatAnderson, FallWenatchee,RapidBend, Topeka,FortPWinchesterProvo-OremGrandCedarLansing-Ea Sandusky,Wayne CityORSt.Lincoln, ForkKS Cloud,Colorado Rapi BarnstableNRochester, S FondBillings,Racine,Portland-SVirginiaSaltColumbus, duHarrisburg Lake La WIFairbanks, B Norwich-Ne Dalton,Salem, GAVineland-M ORWarner Rob Manchester

Gini Coefficient, 2006 Coefficient, Gini Longview,MaHanford-ConsIdahofield,Sherman-De FallGreen Bay,Albany-SchVallejo-Fa St.ElizabethtLogan, Lewiston-AJosephSpringfiel UT-Reading,Janesville P .4 Anderson, Poughkeeps Muskegon-NJeffersonGlensOgden-Clea FallBurlingtonBremerton- Jacksonvil MichiganKankakee-BCheyenne,Wausau, CAppleton,Holland-Gr WI York-Hanov Lebanon, P St. George Sheboygan,Monroe, MI

Hinesville .35 10 10.5 11 11.5 Log Median Family Income, 2006

Source: 2006 American Community Survey

44 371

Figure 4: Gini Coefficient and Log of Median Family Income, 1980 Gainesvill McAllen-Ed .45 Brownsvill Columbus,AlexandriaMonroe,New LA York, Ocala, FL Waco, TXMiami, FL Gadsden,Tallahasse A WilmingtonShreveportNew Orlean Athens,Memphis,Abilene, GA T T FortChico-Para SmithMobile,HuntingtonJerseyMontgomeryBirmingham AL CitWest Palm FayettevilSavannah,HuntsvilleLos Angele El Paso,YubaTuscaloosa SanT City,Medford-AsOrlando,Jackson, Antoni Austin-SanF M Tampa-St.Texarkana,Knoxville,JacksonvilStockton-LStateCharlottes CollChampaign-Baton Roug Sarasota-BLexington,Longview-MBentonNashville,BellinghamColumbia, Har Bergen-Pas Joplin,JohnsonFlorence,Visalia-TuAnniston, MOYakima, Florence,CiAlbany,Macon,Fresno,Pueblo,CorpusRoanoke, WA Raleigh-DuGA GASalinas, CA CO Chr V C .4 Pensacola,Lubbock,BakersfielTucson,Tyler,Santa TXAZTSanta CruzBloomingto BarbNewark, NJ Daytona BeRedding,Lafayette,TerreModesto, CSiouxRiverside-Fort HautSanEvansville Tulsa,Laude CityOlympia,CBillings,Philadelph Diego, OK W San Franci Lakeland-WAsheville,FortBiloxi-GulAugusta-Ai MyersAlbuquerquProvidenceWheeling,OklahomaLouisvilleSacramentoAtlanta, C Trenton,G New Haven-N SpringfielChattanoogGreenvilleGreensboroGreeley,Norfolk-ViHagerstownMelbourne-Eugene-Spr CBoston-Wor CumberlandCharlestonUtica-RomeJohnstown,Portland,Atlantic-CSpokane,SpringfielCharlestonLasFargo-MoorSantaPittsburgh Vegas, Kokomo,WWilmingtonCleveland- Galveston-Rosa IN Jacksonvil Danville,GlensSt.Wichita Little JosephFall BinghamtonRoc Tacoma,FaKankakee,Amarillo,Wichita,Akron,Baltimore,St.Portland-VDallas, WA Louis, Reno,OH K TXChicago, NV I Fayettevil ColoradoParkersburBoiseSalem,Columbia,Charlotte-Phoenix-MeLafayette, Buffalo-NiCity Omaha,SORCincinnatiMonmouth-OSaginaw-BaOdessa-MidSpringfielKansas NE-Detroit, Cit M Killeen-Te NewSyracuse,Kalamazoo-Columbus, Toledo,LondonMadison, Honolulu,OHHAnnouston, W Arbor, T Scranton--Vineland-MMuncie,Albany-SchSouthRichmond-PDayton-Spr IndianapolIN BendDenver, CO Hickory-MoEau ClaireMansfield,Newburgh,Duluth-SupYoungstownTopeka,Jackson,LakeBeaumont-PSteubenvil CharlRochester, KS MFlint, MI Lynchburg,Erie,Reading,Fort PA ColliLansing-Ea PHartford,Seattle-Be Provo-Orem York,FortLincoln,GreenDesHamilton-M WaynePA Bay,MoinesVentura, N CWashington Altoona,Williamspo BurlingtonP SiouxLancaster,Allentown- Decatur,FallVallejo-FaMilwaukee- MinneapoliIGary, IN Elkhart-GoSharon,Wausau,GrandSalt PA LakeWI RapiDutchessPeoria-PekDavenport-Orange C Cou Clarksvill Lima,HarrisburgCanton-MasSheboygan, OH Bremerton,Racine, WI Anchorage, St. Cloud, Waterloo-CRockford, Nassau-SufSan Jose, .35 Cedar Rapi Gini Coefficient 1980 Janesville Middlesex- Appleton-O Kenosha,Richland-K W .3 9.4 9.6 9.8 10 10.2 10.4 Log Median Family Income, 1980

Source: Gini coefficients are calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 1980, at usa.ipums.org. Median family income is from the 1980 Census.

45 372

Figure 5: Relationship Between Gini Coefficent and Log Population, 2000 .55 Bryan-Coll New York, Gainesvill New Haven- Miami, FL Athens, GA .5 Savannah,BrownsvillMcAllen-Ed Auburn-OpeAlexandriaMonroe,Tuscaloosa Naples,LA FL West Palm HattiesburIowaBloomingtoGreenville City, Lubbock, TLexington, New Orlean Los Angele Albany, GA TallahasseJackson, M San Franci Jackson, TLongview-MShreveportLafayette, BirminghamMemphis,Newark, T NJ Chico-ParaWaco,Odessa-Mid TXSantaMobile, BarbJersey AL Cit Hartford, Houston, T Anniston,StateFlorence,Champaign-Tyler,Laredo, Coll TXColumbus, TXFortTrenton, PiercJohnsonCharlestonBatonSarasota-B NElKnoxville, Paso,Ci RougFresno, T CAFort Laude Boston-Wor Lake CharlMontgomeryFort MyersBakersfiel Bergen-PasPittsburgh Philadelph Gadsden,Abilene,Redding, A TWilmingtonSanSanta C LuisBeaumont-P Cruz O Tucson, AZCincinnati Dallas, TX Chicago, I Dothan,Columbia,YubaLasYolo, City, Cruces ALAsheville, CABoulder-LoMacon,ChattanoogAugusta-AiFlint, GA MI OklahomaLouisvilleBuffalo-NiNashville,San CAntoniTampa-St.San Diego, Danville,WichitaCharlottesFortYakima, FaGalveston- SmithSpringfielReno, WAVisalia-Tu NV Tulsa,ProvidenceGreenvilleMonmouth-O Austin-SanOK Cleveland-Oakland, C .45 Flagstaff,SantaHouma, Fe,Amarillo,Portland, LAHuntsvillePensacola, Orange Cou Muncie,RockyPuntaPanamaMedford-As INGordMounBinghamton Cit Eugene-SprAtlantic-CCorpusDaytonaStockton-L Toledo,LittleChrScranton--AlbuquerquSyracuse, BeRoc OHJacksonvilGreensboro Detroit, M GrandSiouxPuVineland-MDecatur,Bentone Juncblo, Greeley,Yuma,Lafayette,CityMerced,Springfiel Roanoke,CO HarA AZEvansville C CABiloxi-GulSpokane, Modesto,V Columbia,SpringfielAkron, W C OHRaleigh-DuColumbus,Charlotte-Orlando,Baltimore,St. FPhoenix-Me Riverside-Louis,Atlanta, G Sumter,Altoona,Jamestown,TerreJoplin,Bellingham Fargo-MoorSC Haut BarnstableP MOLincoln,Ocala,FayettevilSaginaw-Ba FLNLakeland-WYoungstownRichmond-PRochester,Milwaukee-LasSacramentoFort Vegas, WorthSeattle-BeNassau-Suf WilliamspoLaBillings, Crosse,Lynchburg,Utica-RomeDavenport-Salinas,Kalamazoo-Lansing-EaMelbourne-Ann C Arbor,Albany-SchDayton-SprIndianapolSanDenver, Jose, CO St.Goldsboro,Decatur, Waterloo-CJoseph I Johnstown,Duluth-SupSouthErie, Bend Madison,PASantaWilmington RosaOmaha,Gary, Ventura,W IN NE- C Kansas Cit Washington Richland-KMyrtle BeaKilleen-TeCanton-MasDesBoise Moines HarrisburgCity Honolulu, Portland-V Kankakee,Sharon,Rochester,HagerstownMansfield, PA DutchessHickory-MoPeoria-Pek ColoradoC Allentown- S Middlesex-Norfolk-Vi BloomingtoJackson,Topeka,Bremerton,Fort MKSFayettevilHamilton-M Salem,ColliRockford,Newburgh,Fort Wichita,OR Tacoma,Wayne K WA Minneapoli Kokomo,GlensFort SiouxINElkhart-Go Fall Walto Brazoria,Fall Reading,Vallejo-Fa P Grand Rapi Gini Coefficient 2000 Wausau,Lima,CedarOlympia, WI OH Rapi W Salt Lake Eau Claire Provo-Orem .4 Kenosha,St.Racine, Cloud,Green W WI Bay, Dover, DEClarksvillAnchorage,Lancaster, JacksonvilJanesville York, PA Sheboygan, Appleton-O .35 11 12 13 14 15 16 Log of Population

Source: Gini coefficients are calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org. Population data is from the 2000 Census.

46 373

Figure 6: Gini Coefficient of Housing Consumption and the Gini Coefficient, 2000 Baltimore, Birmingham Miami, FL .32 Kansas Cit Columbus, Boston-Wor Charlotte-

.3 ProvidencePhiladelph Pittsburgh New OrleanLos Angele Tampa-St. Detroit, MCincinnati IndianapolSt. Louis,Oklahoma C New York, Norfolk-ViFort WorthBuffalo-NiDallas, TX San Franci .28 SanSan Antoni Diego, Rochester, Houston,Memphis, T T Salt Lake Portland-V WashingtonRiverside- Milwaukee-Cleveland-Hartford, .26 Phoenix-MeChicago, I MinneapoliDenver,Seattle-Be COOakland, C Atlanta, G .24 San Jose, Orange Cou

Sacramento GiniCoefficient Housing of Consumption, 2000 .22 .4 .45 .5 .55 Gini Coefficient 2000

Source: The 2000 Gini coefficient is calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org. The housing Gini coefficient was calculated using housing values and controls from the American Housing Survey Metropolitan Samples for 1998, 2002, 2003 and 2004.

47 374

Figure 7: Gini for Household Income and Gini for Male Workers Ages 25-55, 2000 .55 New York,

New Haven- Miami, FL

.5 McAllen-Ed Los Angele West Palm New Orlean San Franci BirminghamMemphis, T Jersey Cit Newark, NJ Mobile, ALEl Paso, THartford,Fort Sarasota-BLaudeHouston,Fresno, CA T Baton RougKnoxville,CharlestonBoston-Wor PittsburghPhiladelph Bergen-Pas Cincinnati Bakersfiel Dallas, TX Louisville Tucson,Chicago,Tampa-St. AZ I Buffalo-NiCleveland-OklahomaNashville, CSan AntoniOakland,San Diego, C Providence Tulsa, OK Austin-San .45 Greenville Monmouth-O Toledo, OHSyracuse,Detroit, M Stockton-LLittleAlbuquerqu Roc Orange Cou Scranton-- GreensboroColumbia,Jacksonvil Springfiel St.Baltimore,Columbus,Akron, Louis, OHRaleigh-DuCharlotte-Phoenix-MeOrlando,Atlanta, G F Rochester,Milwaukee-Riverside-Las Vegas,Seattle-BeFort WorthNassau-Suf YoungstownDayton-SprAlbany-SchIndianapolAnn Arbor,Richmond-PSacramentoDenver, CO San Jose, Gary, IN Kansas WashingtonCit Honolulu,WilmingtonOmaha,Portland-V NE- Ventura, C HarrisburgNorfolk-ViAllentown- Colorado S Middlesex- GiniCoefficient HH of Income, 2000 Tacoma,Wichita,Fort Wayne KWA Minneapoli Grand RapiVallejo-Fa Salt Lake .4 .35 .4 .45 .5 Gini Coefficient for Male Workers, 2000

Source: Gini coefficients are calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org.

48 375

Figure 8: Gini Coefficient and Human Capital Only Gini Coefficient, 2000 New Haven- .5 New York,

San Franci Miami, FL Los Angele Orange Cou West Palm McAllen-Ed Bergen-PasNewark, NJ

.45 Dallas,Fresno,Houston, TX CA T Sarasota-BAustin-San FortSan Laude Oakland,Jose, CSan Diego,Ventura, C Hartford, Bakersfiel Nassau-SufBirminghamMemphis,Tampa-St.Orlando,Chicago, TAtlanta, F IG Boston-Wor SanJersey Antoni CitPhoenix-Me Nashville,Tucson,Albuquerqu AZ JacksonvilNewCharleston OrleanCharlotte-Denver,El Paso, FortCO Worth T Monmouth-OSeattle-BeCincinnatiPhiladelphLittleRaleigh-Du Roc Stockton-L Middlesex-WashingtonRichmond-PColumbia,Tulsa,Sacramento OK Pittsburgh Knoxville,Portland-VOklahomaLas Vegas, C .4 Mobile,Louisville AL Greensboro Ann Arbor,Akron, OHColumbus,Omaha, NE- Riverside- MinneapoliCleveland-St.Baltimore,KansasDetroit, Louis,Baton MCit RougGreenville Buffalo-NiWilmingtonColorado S Syracuse,Milwaukee-Indianapol Salt Lake Albany-Sch Vallejo-Fa Rochester,Honolulu, Toledo,Allentown- OH Providence Harrisburg Norfolk-ViGrand Rapi Gini Coefficient for Male Workers, 2000 Workers, Male for Coefficient Gini Scranton-- Dayton-SprSpringfiel Gary, IN Tacoma,Fort Wayne WAWichita, K Youngstown .35 .18 .2 .22 .24 .26 Human Capital Only Gini Coefficient, 2000

Source: Gini coefficients are calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org.

49 376

Figure 9: Gini Coefficient and Human Capital Only Gini Coeff. Using Occupations, 2000 New Haven- .5 New York, Los Angele San FranciMiami, FL Orange Cou West Palm McAllen-Ed Bergen-PasNewark, NJ

.45 Houston,Dallas,Fresno, TX T CA Austin-SanSarasota-B San Jose,Oakland,Fort Laude C San Diego,Ventura, C Hartford,Bakersfiel BirminghamTampa-St.Nassau-SufMemphis,Atlanta,Orlando, GT FChicago, I Boston-WorSan AntoniPhoenix-Me Tucson,Jersey AZNashville, Cit JacksonvilNewEl Paso, OrleanFortAlbuquerquDenver,Charlotte- TWorthCharleston CO Seattle-BeMonmouth-OLittleCincinnati Raleigh-DuRoc PhiladelphStockton-L Middlesex-Richmond-PColumbia,Tulsa,WashingtonSacramento OK PittsburghKnoxville,Portland-VOklahoma C Las Vegas, .4 Mobile, AL Greensboro AnnAkron, Arbor,Riverside- OHColumbus,LouisvilleOmaha, NE- Baton RougDetroit,GreenvilleMinneapoliCleveland-St. MBaltimore, Louis,Kansas Cit Buffalo-NiColoradoWilmington S Syracuse,Milwaukee-IndianapolSalt Lake Albany-Sch Vallejo-Fa Rochester, Honolulu, Toledo, OH Providence Norfolk-ViAllentown-Harrisburg Scranton--SpringfielDayton-SprGrand Rapi Gary, IN Tacoma,Wichita,Fort WAWayne K

Gini Coefficient for Male Workers, 2000 Workers, Male for Coefficient Gini Youngstown .35

.24 .26 .28 .3 .32 Human Capital Only Gini Coefficient Using Occupation, 2000

Source: Gini coefficients are calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org.

50 377

Figure 10: Relationship Between Share of Adults with College Degrees 2000 and Share of Adults with College Degrees 1940 Boulder-Lo .5 IowaCorvallis, City,

San Franci Lawrence, WashingtonColumbia, San Jose, Madison, W .4 Raleigh-DuBloomingtoFort Colli CharlottesGainesvill Champaign- TallahasseMiddlesex- Santa Fe, AnnBryan-Coll Arbor, BloomingtoAustin-SanSeattle-BeState Coll Burlington Rochester,Oakland, C Portland,Trenton,NewSantaBoston-WorYolo,Barnstable Athens,Haven- CruzN CA GADenver, CO Bergen-PasMinneapoli Missoula,Lincoln, N Atlanta,Provo-Orem G Newark,Nassau-SufColorado NJ S Huntsville Hartford,Chicago,Orange I Cou .3 Olympia,Dallas, WTX Baltimore,Portland-VNewColumbus,Flagstaff,Fargo-Moor York,San SantaDiego,Columbia, Barb SpringfielSantaOmaha,KansasAlbany-SchRichmond-P RosaLansing-EaJackson,Albuquerqu CitDesNE-Lafayette,Lexington, Moines M Naples, FLBellinghamDutchessMonmouth-OPhiladelphRochester,We WilmingtonCHouston,Auburn-OpesCedart PalmHonolulu, Rapi T NewGreenvilleNashville, LondonMilwaukee-Charlotte-SanBillings,Ventura,Boise Luis City OSaltC Lake Tucson, AZ Jersey CitPittsfieldWilmingtonBremerton,Bismarck,SiouxIndianapolSacramento FallEugene-SprTopeka, KS LaMontgomerySt. BirminghamCrosse,LittleProvidence CharlestonLouis,CincinnatiSpringfiel FortRocPocatello,Asheville, Spokane,BatonWorthOrlando,Sarasota-BRapidFortPhoenix-MeWichita, RougLaudeOklahoma CityLos FW KAngele C Fort Walto GrandSouthNorfolk-ViPittsburghMelbourne-Akron,TuscaloosaHattiesburSyracuse, Fork Bend OH Lubbock, Reno,T NV Savannah,Memphis,GrandMonroe,Detroit,Galveston-Hamilton-MJacksonvilKnoxville,Buffalo-NiGreensboroRichland-KKalamazoo-Vallejo-Fa Rapi Waterloo-C TLACleveland- M Tulsa,Cheyenne, OK BinghamtonEauFayettevilGreenAppleton-OSpringfielLouisvilleNe Tyler,ClaireNewDayton-SprHarrisburg wbBay,Roanoke,GrandMedford-Asurgh, SanOrleanTXChico-ParaLas Salinas, AntoniJuncCruces V Abilene, C T Jonesboro,Augusta-AiSt. Allentown-Pensacola,Cloud,Peoria-PekToledo,Duluth-SupDubuque,Tampa-St. OHFortGreeley,Great IMyers FallMiami, C Amarillo, FL Jackson,Racine,Lancaster,Bangor,Davenport-Erie,Muncie, T Tacoma,WIGreenvillePACharleston MESalem, INCasper, WAOR WY .2 Mobile,Brazoria,Kenosha,FortBentonChattanoogAtlantic-CMacon, AL FayettevilWayneShreveport Waco,WFort HarLawton, GAEnid, Pierc TXSan OK OK Angelo Wausau,MyrtleEvansvilleYork,Saginaw-BaReading,Dover,Columbus, Pueblo,GlensBeaRockford,Lynchburg, Terre PAWI Florence,Elmira,DE Fall WichitaHautP CO NY Fa Odessa-Mid Lafayette,Sheboygan,PanamaScranton--Canton-MasKilleen-TeAlbany,Utica-RomeGary,PuntaSherman-DeSharon, Cit INSiouxGACorpusBiloxi-Gul Gord PA City ChrDaytona Be Dothan,ClarksvillFlorence,Joplin,Jamestown, St.ALOwensboro,Lake JosephAlexandriaDecatur,Longview-M MOJanesville CharlRedding,Fresno, I El Paso,Las C CA Vegas, T Decatur,PineFlint, BluffVictoria, JohnsonElkhart-GoA Jackson,MI Sumter, Ci M SCRiverside- JacksonvilKankakee,Texarkana,HagerstownKokomo,ParkersburYoungstownGoldsboro,Anniston,Stockton-LWilliamspoBeaumont-PYakima, INLakeland-W WA FortLaredo,Altoona,Lewiston-A RockySmithHickory-MoOcala,Huntington TX P Moun Modesto,FL C Gadsden,CumberlandJohnstown,Lima,McAllen-Ed BrownsvillAOHYubaBakersfiel City, Houma,Vineland-MSteubenvilWheeling, LAMansfield,Visalia-Tu Share of Adults with College Degrees, 2000 Degrees, College with Adults of Share Danville,Merced,Yuma, CA AZ .1 0 .05 .1 .15 Share of Adults with College Degrees, 1940

Source: Share of adults with college degrees in 2000 is from the 2000 Census. Share of adults with college degrees in 1940 is from Haines, M.R., ICPSR study number 2896, Historical, Demographic, Economic, and Social Data: The United States, 1790-2000.

51 378

Figure 11: Relationship Between Share of Adult HS Dropouts, 2000 and Share of Adult HS Dropouts, 1940

.5 McAllen-Ed Laredo, TX Brownsvill

.4 Visalia-Tu Wheeling, Merced, CA El Paso,Yuma, T AZ Fresno, CA Houma, LA Miami, FL Salinas,Yakima,Bakersfiel CWA Vineland-MDanville, Los Angele Kokomo, IN .3 Modesto, C Las CrucesJohnsonHickory-MoJersey Ci Cit Stockton-L Rocky MounLafayette, Yuba City, Florence, Odessa-Mid Corpus ChrNew York,Sumter,FortLynchburg,Gadsden, Anniston,Smith SC Decatur,A A Riverside-Wichita Fa Lakeland-WAlbany,Alexandria GAFlorence,Pine Bluff Houston, TElkhart-GoCharlottesHuntingtonGreenville San Angelo Waco, TXChattanoogLakeProvidenceNewVictoria, CharlTexarkana, OrleanColumbus, Dothan, AL San Antoni Jackson,Lancaster,HagerstownReading,Goldsboro,Jonesboro, T P Lubbock, T ShreveportBeaumont-PLongview-MMonroe,Mobile,GreensboroMacon,Ocala,Tuscaloosa LAAugusta-Ai AL FLGA OrangeSantaLas Barb Vegas,CouDallas, TX Greeley,Yolo,Richmond-P CABaton CLewiston-AJoplin,Memphis,Atlantic-CKnoxville,Montgomery RougBrazoria,Utica-RomeSavannah, MOKankakee,Dover,Greenville TJohnstown, DE .2 Amarillo, SiouxVentura,Richland-K CityMansfield, C Tyler,Biloxi-GulSherman-De TXFayettevilAthens,Cumberland GA FortSalem, WorthAsheville, Chicago,ORHattiesburFortRoanoke,WilliamspoGalveston-Charleston Charlotte-BirminghamPierc IOwensboro,Panama V Allentown- CitYork, PAMyrtle Bea Abilene,FortPhoenix-Me T LaudeTampa-St.Bryan-CollSpringfielMuncie,Newark,Jackson,Rockford,Pueblo,St.Benton Joseph INJamestown,NJTrenton,GlensCincinnatiNashville, TerreLouisvilleM CharlestonCO Har Baltimore,FallAuburn-Ope HautScranton--Steubenvil N ClarksvillNaples, FL San Diego, Enid,DaytonaOrlando, OKVallejo-FaChico-ParaCleveland-Fort BeLexington,Elmira, F Detroit,MyersSouthGary, Bergen-Pas PuntaYoungstownNYBend ParkersburMPhiladelphRacine, IN NeGordSt.wb WILouis,urgh, Tucson,SanOklahoma Jose, AZTulsa,SantaRedding,We CDayton-Spr OK Cruzst JacksonvilPalmNorfolk-ViDecatur, CLittleLima,Sharon,NewFlint,HarrisburgWilmington Roc Hamilton-M Canton-MasHaven-OHMI I Buffalo-NiPAAltoona,Saginaw-BaKenosha,EvansvillePensacola, PWausau, W Huntsville WI Oakland,Reno,San NV C Franci Sarasota-BIndianapolSantaJanesvilleAlbuquerquColumbia,Flagstaff,Jackson, Syracuse,Fe,Davenport-Peoria-PekErie,Hartford,Toledo,SpringfielRochester,Atlanta,Milwaukee-Grand PA MBinghamton OHKilleen-Te DutchessRapi G Sheboygan, C Jacksonvil Wichita,Medford-AsSacramento SanK Boston-WorGrand LuisKalamazoo- OJuncSantaHonolulu,Austin-SanFortLawton,PittsfieldMonmouth-OWilmingtonDubuque, Albany-SchWayneRosaRaleigh-Du PittsburghOKFayettevil I Denver,BoiseWaterloo-C CO CityColumbus,KansasBangor,Lafayette,Akron,Melbourne-Nassau-Suf CitBismarck, ME OHGreenNew Bay,Middlesex-Tallahasse LondonSt. Cloud, Salt LakeWashingtonPortland-VGreat FallTacoma, WADuluth-SupGrandEauAppleton-O ClaireFork ShareAdult of HS Dropouts, 2000 Pocatello,Eugene-SprBellingham DesCasper, MoinesBillings,Topeka,Rapid WYCity Omaha,KSSiouxLansing-Ea NE-FallStateBurlingtonSpringfiel BloomingtoCollGainesvill Fort Walto Cheyenne,Spokane,Columbia,Olympia,Fargo-Moor W W La Crosse, .1 Lincoln, NSeattle-BeCedarPortland,Provo-OremAnnBremerton, Rapi Arbor,MinneapoliBloomingto ColoradoChampaign-Missoula,BarnstableRochester, S Lawrence,Madison, WBoulder-LoFort Colli Corvallis,Iowa City, .5 .6 .7 .8 .9 Share of Adult HS Dropouts, 1940

Source: Share of adult high school dropouts in 2000 is from the 2000 Census, and share of adult high school dropouts in 1940 is from Haines, M.R., ICPSR study number 2896. Historical, Demographic, Economic, and Social Data: The United States, 1790-2000.

52 379

Figure 12: Relationship Between Share of Adult HS Dropouts, 2000 and Share of Hispanic Population, 2000

.5 McAllen-Ed Laredo, TX Brownsvill

.4 Visalia-Tu Wheeling, Merced, CA Yuma, AZ El Paso, T Houma, LA Fresno, CA Danville, Vineland-M Yakima,Bakersfiel WASalinas, CMiami, FL Kokomo, IN Los Angele .3 JohnsonHickory-Mo Ci Modesto,Jersey C Cit Las Cruces Lafayette,Rocky Moun Stockton-L Florence, Yuba City, Lynchburg,Anniston,Gadsden,Sumter,Decatur,Fort Smith SC A New York,Odessa-Mid Corpus Chr PineFlorence,Albany,Alexandria Bluff Lakeland-WGAWichita Fa Riverside- HuntingtonCharlottesGreenvilleElkhart-Go Houston, T LakeChattanoogDothan,Texarkana,Columbus,New Charl ProvidenceOrlean AL Waco, TX San AngeloVictoria, HagerstownJackson,Jonesboro,Goldsboro,Lancaster,Reading, T P San Antoni Monroe,TuscaloosaMobile,ShreveportMacon,Augusta-AiGreensboroOcala,Longview-MBeaumont-P AL GALA FL Lubbock, T Johnstown,Lewiston-AMontgomeryKnoxville,BatonSRichmond-PMemphis,Utica-RomeJoplin,GreenvilleDover,aKankakee,vannah, RougAtlantic-C MODE T LasBrazoria,Dallas, Vegas,Yolo,Greeley, OrangeTX CASanta C Cou Barb .2 CumberlandMansfield,Biloxi-GulAthens,Sherman-DeFayettevilTyler,Sioux GA TXCityAmarillo,Richland-K Ventura, C CharlestonWilliamspoOwensboro,Roanoke,HattiesburBirminghamPanamaMyrtleAsheville,York,Charlotte-Allentown-Fort PABea CitV PiercSalem,Chicago,Galveston-Fort WorthOR I SteubenvilJackson,TerreMuncie,CincinnatiScranton--Auburn-OpeGlensLouisvilleBaltimore,St.CharlestonBentonNashville,Jamestown, ClarksvillJoseph HautRockford, Fall Trenton,IN Tampa-St.MHarSpringfielNewark,FortAbilene,Bryan-Coll NNaples, Laude NJPhoenix-Me T FL Pueblo, CO ParkersburSt.Elmira,YoungstownLexington,Detroit,PuntaCleveland-Enid, SouthLouis,PhiladelphDaytonaRacine, FortNYGordOKGary, Chico-Para Newburgh,MBend Myers Be Orlando,WIINBergen-PasVallejo-Fa FSan Diego, Altoona,Sharon,Wausau,EvansvilleCanton-MasDecatur,Lima,Dayton-SprHamilton-MHuntsvilleLittleWilmingtonFlint,Pensacola,Buffalo-NiHarrisburgNorfolk-ViJacksonvilTulsa,Saginaw-BaRedding,Oklahoma Kenosha, OHRocMI PAP IWINew OKWest C Haven- W CPalmSanSanta Jose,Tucson, Cruz AZ Peoria-PekSpringfielBinghamtonSyracuse,Jackson,Erie,Columbia,IndianapolSheboygan,JanesvilleRochester,Toledo,Davenport-Milwaukee-GrandDutchessSarasota-BAtlanta, JacksonvilPAHartford,Flagstaff, M OH Rapi GKilleen-Te Reno,SanCOakland, Franci NV C AlbuquerquSanta Fe, PittsburghPittsfieldDubuque,Albany-SchFortKalamazoo-WilmingtonMonmouth-OBoston-WorRaleigh-DuMedford-AsHonolulu,FayettevilWichita, WayneLawton,Grand I Sacramento SanK JuncOKSanta Luis Rosa OAustin-San Bangor,Bismarck,Akron,St.Waterloo-CColumbus,GreenTallahasse Melbourne-Cloud,NewKansasLafayette, OHBoise ME Nassau-Suf LondonBay,Middlesex- Cit CityDenver, CO EauDuluth-SupAppleton-OGreatGrand Tacoma,ClairePortland-V WashingtonFall ForkSalt WALake ShareAdult of HS Dropouts, 2000 Eugene-SprPocatello,Bellingham BurlingtonSpringfielStateBloomingtoSiouxRapidBillings,DesFortLansing-EaCasper,Omaha,Gainesvill Topeka, Coll Fall WaltoMoinesCity WY NE- KS LaColumbia,Fargo-MoorSpokane, Crosse,Olympia,Cheyenne, W W .1 Portland,CedarBloomingtoAnnLincoln,MinneapoliBremerton,Seattle-BeAnchorage,Provo-Orem Arbor,Rapi N BarnstableMissoula,Rochester,Champaign-Colorado S Lawrence,Madison,FortBoulder-Lo ColliW IowaCorvallis, City, 0 .2 .4 .6 .8 1 Share Hispanic, 2000

Source: 2000 Census.

53 380

Figure 13: Relationship Between Share of Employment in High Capital Industries and Share of Adults with College Degrees, 2000 .7

San Jose, Springfiel Rochester, .6 TallahasseGainesvillColumbia, Iowa City, Trenton,Austin-San N Washington BloomingtoCharlottesMadison, W Boulder-Lo Wilmington StateBryan-CollSantaMiddlesex-Champaign-Raleigh-Du Coll Fe, San Franci Galveston-Albany-SchBoston-Wor DutchessNew York, C Oakland, C BinghamtonSacramentoSiouxPhiladelph Fall HuntsvilleNewark,Yolo, NJ CA .5 Lafayette, Jackson,Baltimore,Hartford,Provo-OremNassau-Suf MLincoln,New Haven- N Bloomingto Brazoria, Monmouth-OCedarRichmond-PDesAlbuquerquLexington,Olympia, MoinesRapi Athens, WPortland, GA Corpus Chr Melbourne-BatonLubbock,BirminghamVentura, RougLansing-Ea TColumbus,Columbia,Dallas, CBergen-Pas TXDenver,SantaAnn Cruz CO Arbor,Fort Colli Odessa-MidLasHarrisburg OklahomaCrucesJerseyGreenvilleOmaha,Kansas Cit C ColoradoNE- CitMinneapoli S AlexandriaUtica-RomeNewburgh,Buffalo-NiPittsburghMontgomeryLosPhoenix-MeTopeka, Tucson,AngeleHouston, Flagstaff,KSChicago,Orange AZ TAtlanta, Cou I Seattle-Be G Beaumont-PLake Macon,CharlTampa-St.NewSanMonroe, GASyracuse, AntoniOrleanLittleSt. Boise LASaltLouis, Auburn-OpeRoc LakeSanta SanCity Diego, Barb Columbus,Richland-KTulsa,TuscaloosaProvidenceSpokane,SpringfielCincinnatiIndianapol OKRochester,Nashville,SantaLafayette,Anchorage,Portland-V W Rosa Houma,Brownsvill LALongview-MAlbany, Allentown-GAChico-ParaMedford-AsRoanoke,Abilene,JacksonvilCleveland-Knoxville,HattiesburFortAsheville, SanLaudeMilwaukee-T V Luis O McAllen-EdLaredo,Redding, TXWaco, Duluth-SupCAmarillo,Miami, TXVallejo-FaTyler,Fort FL TX WorthWestFargo-Moor Palm Barnstable .4 BakersfielJohnsonScranton--SiouxPuLynchburg,Shreveporteblo, Ci City Dayton-SprCOLouisvilleSpringfielSouth BendBillings, Fresno,El DaytonaSaginaw-BaPaso, CA TPensacola,Greeley,Grand BeEauMemphis,GreensboroHamilton-M Claire JuncEugene-SprBremerton, CCharlotte- THonolulu, HagerstownStockton-LKankakee,Reading,TerreKenosha,GlensChattanoogMobile,Racine,Muncie,Salem,Augusta-AiPeoria-PekSt. Haut Fall Cloud,PKalamazoo-Savannah, ALWNorfolk-Vi WIAkron,ORLaINOrlando,Charleston WilmingtonCrosse, OH F Rocky MounSt.Punta Dover,JosephJackson, GordErie, DEWaterloo-C PA Sarasota-BTWichita,Bellingham K Johnstown,YubaWilliamspo City,Riverside-Sharon,PanamaWausau,EvansvilleYork,Fort PAWayne Toledo,PACit WISalinas,Detroit,Fort OH Walto MC Altoona,Jamestown, PGary,WichitaFortTacoma,Davenport-Greenville FortIN Pierc Fa Myers WA Vineland-MVisalia-TuGadsden,Modesto,Ocala,Lakeland-WKokomo,Goldsboro,Anniston,Decatur,Joplin,Dothan,Canton-Mas FL A CRockford, IN MO ALancaster, ALAppleton-OGreenReno, Bay, NV Merced,Yuma,Lima,Fort Yakima,CAAZSumter, Smith OHBiloxi-GulMyrtle WA SCAtlantic-C BeaGrand Rapi YoungstownJackson,Flint,Florence,Decatur,Killeen-Te BentonMI M I Fayettevil Har .3 Naples, FL Danville,Mansfield,JanesvilleFayettevil LasClarksvill Vegas, Hickory-MoSheboygan, Elkhart-Go Jacksonvil .2

Share of Employment in High Capital Industries, 2000 Industries, Capital High in Employment of Share .1 .2 .3 .4 .5 College Completion Among Population 25 or Above, 2000

Source: 2000 Census

54 381

Figure 14: Relationship Between Share of Employment in Low Capital Industries and Share of Adults without College Degrees, 2000 .6 Jacksonvil

.5 Hickory-Mo

Fayettevil Yuma,Merced, AZ CA Gadsden,Visalia-Tu A .4 Naples, FL Gary,Florence, INModesto, C MyrtleJoplin, Anniston,BeaLakeland-WFortOcala, MO Danville,Smith FL Medford-AsGreeley,Fort MyersSheboygan,PuntaSharon, C Yakima, GordRocky McAllen-EdPA WA Moun GreensboroEvansvilleWausau,WilliamspoYoungstownKankakee, WIVineland-M Waterloo-CGreenvilleLancaster,FortSioux PiercRedding,Riverside-Elkhart-Go Stockton-LCityBakersfielYuba C City, WilmingtonEugene-SprRichland-KGEaurChico-ParaandSt.Salem, Claire Cloud,RapiJamestown, Dothan,Fresno,ORSumter,Goldsboro, CAAL SC BellinghamBillings,Sarasota-BTyler,GrandToledo,Erie,Davenport-Benton TXTerre Reading,Pu JuncDaytonaPA Canton-MasOHeJanesvilleblo,Decatur, HarHautHagerstown CO PBeJohnstown, A Charlotte-HattiesburSalinas,GreenLasAmarillo,Muncie, CrucesChattanoogFort Bay, C St.Decatur,WayneElJohnson INJosephPaso,Beaumont-PLima,Brownsvill IT Ci OH Auburn-OpeBoiseAsheville, CitySpringfielLouisvilleJackson,Kenosha,Waco,GlensLynchburg,York,Panama FallPA TXT W Cit .3 Fargo-Moor Pensacola,Mobile,Dover, AL DELaredo,Mansfield, TX BarnstableSantaSanGreenville BarbCharlestonLa Akron,LuisPittsburghSouthHamilton-M Crosse,Memphis,Roanoke, OAllentown- Tacoma, OHBendBrazoria,WichitaSaginaw-BaScranton-- TLongview-MVJackson,Flint, WA FaAltoona, MIHouma, M P LA Bryan-CollAthens, GAFlagstaff,Portland-VSantaWestHouston,Nashville, Phoenix-MeSpokane,Rosa Orlando,BatonPalmTuscaloosaKnoxville,Kalamazoo-Monroe,SanAppleton-O T RougAugusta-Ai Antoni FWRockford, LAAlbany,LakeLas Vegas,Charl GA Fort Colli Yolo,Atlanta,Provo-Orem CA Jackson, GTucson,SaltSiouxFortLittleFort Lubbock,LakeReno, Savannah,JacksonvilM Vallejo-FaWorthFall LaudeAZRocMiami,Tampa-St. NVMacon, TOdessa-Mid FLAlexandria GA Denver,Santa CruzDallas, Olympia,COColumbus,Omaha, IndianapolTXCincinnatiSt. LosBirminghamWFort Melbourne-Cleveland-Louis,NE- HarrisburgAngeleAbilene,New WaltoDuluth-SupPeoria-Pek OrleanShreveport CorpusT Chr State CollHuntsvilleOrangeSanAlbuquerquLafayette,Monmouth-O Diego,Milwaukee- CouSacramentoMontgomeryOklahomaProvidenceDetroit,Dayton-SprNewburgh,Racine, Columbus,CMBiloxi-Gul WI BloomingtoRaleigh-DuLincoln,Chicago,Columbia,Anchorage,Richmond-PDesLexington,Lansing-EaKansas NHonolulu,CedarVentura, MoinesJersey ISpringfiel Syracuse,RapiCitTulsa,Buffalo-NiGalveston- C Cit OKUtica-RomeLafayette,Kokomo, IN Iowa City, CharlottesChampaign-SantaSeattle-BeOakland,Portland, Fe,Minneapoli CBaltimore,Wichita,Norfolk-Vi KAtlantic-CFayettevilClarksvill GainesvillAustin-SanBergen-PasColoradoNewark,Nassau-SufPhiladelphDutchessRochester, NJSTopeka,Bremerton, CBinghamton KS Columbia,Madison,AnnTallahasse WRochester, Arbor,New Haven-NewAlbany-SchWilmington York, Killeen-Te Boulder-Lo San Franci Middlesex-BloomingtoBoston-WorHartford, .2 Washington Springfiel San Jose, Trenton, N

Share of Employment in Low Capital Industries, 2000 Industries, Capital Low in Employment of Share .5 .6 .7 .8 .9 Share of Adults without a College Degree, 2000

Source: 2000 Census

55 382

Figure 15: Returns to College and the Percent of Residents with College Degree, 2000 New Haven-

.7 McAllen-Ed

San Jose, New Newark,York, NJ San Franci

.6 Birmingham BakersfielEl Paso, T Houston, T PittsburghCincinnatiDallas,Hartford, TX Memphis, T GreenvilleLittleCharlotte- RocWest PalmBergen-Pas Washington Mobile,Tampa-St. AL OrangeAtlanta,Boston-Wor Cou G Miami,Orlando, FL Philadelph F Austin-San San AntoniLos AngeleRichmond-PSan Diego, Middlesex- JacksonvilAkron,Ventura, WilmingtonOH Chicago, C IOakland, C

.5 Sacramento Fresno, CA Charleston Colorado S Allentown-HarrisburgGreensboroCleveland-FortSarasota-B LaudeAlbuquerquBaltimore, Dayton-SprDetroit,Knoxville,Phoenix-Me MRochester,Monmouth-O Raleigh-Du Stockton-L LouisvilleNewNorfolk-Vi OrleanSt.FortIndianapol Louis, WorthOmaha,Kansas NE-Cit Denver, CO Vallejo-FaTulsa,Oklahoma OKAlbany-SchColumbia, C JerseyNashville,Milwaukee- Cit Portland-VColumbus,Nassau-SufMinneapoli Buffalo-NiTucson, AZ Syracuse,Providence Scranton-- Seattle-Be .4 YoungstownLas Vegas,Fort Wayne Salt Lake BatonWichita, Roug K Ann Arbor, Grand RapiHonolulu,

Log College WagePremium, 2000 Toledo, OH Riverside- Springfiel Tacoma, WA

.3 Gary, IN .1 .2 .3 .4 .5 College Completion Among Population 25 or above, 2000

Source: Data is calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org.

56 383

Figure 16: Gini Coefficient and Returns to College, 2000 New Haven- .5 New York,

Los Angele Miami, FL San Franci Orange Cou West Palm Bergen-Pas Newark, NJ McAllen-Ed

.45 Fresno, CA Dallas,Houston, TX T Sarasota-BAustin-San Fort Oakland,LaudeVentura,San Diego, C San Jose, Hartford,Bakersfiel Nassau-Suf Chicago,Orlando,Atlanta,Tampa-St.Memphis, I F G TBirmingham JerseyPhoenix-Me Cit SanBoston-Wor Antoni Tucson,Nashville, AZ Albuquerqu Denver,NewFort WorthOrleanCharleston JacksonvilCO PhiladelphCharlotte-Little RocEl Paso, T Seattle-BeColumbia,Stockton-LMonmouth-ORaleigh-Du Cincinnati Tulsa, OKSacramentoRichmond-PMiddlesex-Washington Las Vegas,Portland-VOklahomaKnoxville, C Pittsburgh .4 LouisvilleOmaha,Greensboro NE- Mobile, AL Riverside-Ann Arbor, Columbus,Cleveland-Akron, OH Baton RougMinneapoliKansasSt.Detroit,Baltimore, Louis, Cit M Greenville Buffalo-Ni ColoradoWilmington S SaltSyracuse, Lake Milwaukee-Indianapol Vallejo-Fa Honolulu, Albany-Sch

Gini Coeff. Malefor Workers, 2000 Rochester, Toledo, OH Providence Allentown- Grand Rapi Norfolk-ViHarrisburg Springfiel Scranton-- Dayton-Spr Gary,Tacoma, IN WA Wichita,Fort Wayne K Youngstown .35 .3 .4 .5 .6 .7 Log College Wage Premium, 2000

Source: Data is calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org.

57 384

Figure 17: Gini Coefficient and the Gini Coeff. Holding Skills Constant, 2000 New Haven- .5 New York,

Los Angele Miami, FL San Franci Orange Cou West Palm McAllen-Ed Bergen-PasNewark, NJ

.45 Fresno, CA Dallas,Houston, TX T Sarasota-BAustin-San Fort Oakland,LaudeVentura,San Diego, CSan Jose, Bakersfiel Hartford, Orlando,Tampa-St.Nassau-SufChicago,Atlanta, FMemphis, IBirmingham G T Jersey Cit Phoenix-MeSan AntoniBoston-Wor Tucson,AlbuquerquNashville, AZ Denver,NewFort JacksonvilCOElCharlotte- WorthOrleanCharlestonPaso,Philadelph TLittle Roc Stockton-LSeattle-BeColumbia,Monmouth-ORaleigh-DuCincinnati SacramentoTulsa, OKMiddlesex-WashingtonRichmond-P Las Vegas,OklahomaPortland-V C Knoxville,Pittsburgh .4 LouisvilleOmaha,Mobile,Greensboro NE- AL Riverside- Columbus,Ann Arbor,Cleveland-Akron, OH BatonKansas RougDetroit, CitMinneapoliBaltimore,St.Greenville MLouis, ColoradoBuffalo-NiWilmington S Salt LakeSyracuse,Milwaukee-Indianapol Vallejo-Fa Honolulu,Albany-Sch

Gini Coeff. Malefor Workers, 2000 Rochester, ProvidenceToledo, OHAllentown- Grand Rapi Norfolk-ViHarrisburg SpringfielScranton--Dayton-Spr Tacoma,Gary, INWAWichita,Fort Wayne K Youngstown .35 .15 .2 .25 .3 .35 Gini Coeff. Holding Skill Constant, 2000

Source: Gini coefficients are calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org.

58 385

Figure 18: Returns to Schooling and Share of Workers in Finance, 2000 New Haven-

.7 McAllen-Ed

San Jose, New York,Newark, NJ San Franci

.6 Birmingham Bakersfiel El Paso, T Houston, T PittsburghCincinnati Dallas,Hartford, TX Memphis, T Mobile, ALGreenvilleLittle RocTampa-St.WestCharlotte- Palm Bergen-PasWashington Atlanta,OrangeBoston-Wor G Cou Orlando,Miami, F FLAustin-SanPhiladelph San AntoniSanLos Diego, Angele Richmond-PMiddlesex- Akron,Ventura, OH Jacksonvil C Chicago,Oakland, I C Wilmington .5 Fresno, CA Sacramento Allentown-CharlestonColoradoHarrisburg S GreensboroCleveland-Baltimore, Dayton-SprKnoxville,Rochester,Sarasota-BDetroit,Albuquerqu Raleigh-DuM Fort LaudeMonmouth-O Norfolk-Vi New OrleanSt.Indianapol Louis,Phoenix-MeOmaha, NE- Stockton-L LouisvilleFort WorthTulsa, KansasOK CitDenver, CO Vallejo-FaOklahomaMilwaukee- Nashville,CColumbia,Albany-Sch Jersey CitNassau-Suf Portland-V Columbus,Minneapoli Tucson,Providence AZBuffalo-Ni Syracuse, Scranton-- Seattle-Be .4 YoungstownLas Vegas,Fort Wayne Salt Lake Wichita, K AnnBaton Arbor,Honolulu, Roug

Log College WagePremium, 2000 Grand Rapi Toledo, OH Riverside-Springfiel Tacoma, WA

.3 Gary, IN .01 .02 .03 .04 .05 Share of Employment in Finance, 2000

Source: Gini coefficients are calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org.

59 386

Figure 19: Returns to Schooling and Share of Workers in Computers, 2000 New Haven-

.7 McAllen-Ed

San Jose,

New York,Newark, NJ San Franci

.6 Birmingham El Paso,Bakersfiel T Houston, T PittsburghCincinnatiHartford, Dallas, TX Memphis, T Mobile,Greenville ALWestLittle Tampa-St. RocPalmCharlotte-Bergen-Pas Washington Orange CouAtlanta,Boston-Wor G Miami, FLOrlando,Philadelph F Austin-San LosSan AngeleAntoniSanRichmond-P Diego, Middlesex- Akron,JacksonvilVentura, OH Chicago, WilmingtonC I Oakland, C .5 Fresno, CA Sacramento CharlestonGreensboroAllentown-Cleveland-HarrisburgBaltimore, Colorado S Sarasota-BKnoxville,FortAlbuquerquDayton-Spr Detroit,Laude MRochester,Monmouth-O Raleigh-Du New OrleanNorfolk-ViIndianapolPhoenix-MeSt. Louis,Omaha, NE- Stockton-LLouisvilleTulsa,FortKansas OK Worth Cit Denver, CO Vallejo-FaOklahomaNashville,Columbia,Milwaukee-Albany-Sch C Nassau-Suf Jersey Cit Portland-VColumbus,Minneapoli Buffalo-NiProvidenceTucson, AZ Syracuse, Scranton-- Seattle-Be .4 YoungstownLasFort Vegas, Wayne Salt Lake Wichita, K BatonHonolulu, Roug Ann Arbor,

Log College WagePremium, 2000 Grand Rapi Toledo, OH Riverside-Springfiel Tacoma, WA

.3 Gary, IN 0 .02 .04 .06 .08 Share of Employment in Computers, 2000

Source: Data is calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org.

60 387

Figure 20: Murder Rate and the Gini Coefficient, 2000 15

Flint, MI Mobile, AL Kokomo, IN Macon,Baton GA Roug Goldsboro, ShreveportJackson, M Laredo, TX Savannah, Danville, 10 Kansas Cit RockyYakima, Moun WA Merced,Albuquerqu CABakersfielMontgomeryColumbus, Greenville LittleHuntsville Roc Fresno, CA Pueblo, CO Racine, WI JacksonvilTulsa,Tucson, OK AZ Sumter,St. Louis, SCBeaumont-P Alexandria Columbia,YubaDothan, City, ALJackson, T Roanoke, V Charleston FayettevilFort WayneYuma,Columbus, AZGadsden,Lafayette, A ChattanoogLakeWaco, Charl TX Reading, PRochester,Modesto,Columbia, C Knoxville,Tallahasse Buffalo-NiOklahomaSan AntoniTyler, C TX Anchorage, Altoona,Ocala,Vineland-MAkron,Wichita FL P OH Fa Lubbock,McAllen-Ed T 5 Lima, OH Joplin,Toledo, MO PittsburghOH Albany,Auburn-Ope GA Topeka, KS Amarillo, Lynchburg,SantaFortLasTampa-St.WilmingtonAbilene, Smith Fe, Cruces T Monroe, LA Jackson, M Spokane,Asheville, W Green Bay, Salinas,Lansing-EaUtica-RomeWilliamspoBillings, CCharlottes Elkhart-Go Albany-SchGreeley,Syracuse,Binghamton C Dover,St. Cloud,Sioux DESalColoradoem, Fall Erie,OR Bellingham PA S Corpus Chr Gainesvill Ann Eugene-SprArbor,SpringfielEl Paso, T Lafayette,Redding,Santa Cruz C FortMansfield, Honolulu,Colli Muncie, IN Bloomingto Lancaster,Provo-Orem Lincoln,Decatur,Punta N Gord A Murder Rateper 100,000 inhabitants,2000 Olympia, WLa Crosse, State CollIowa City, EauCedar Claire RapiMadison,Waterloo-C W

0 Sheboygan,Wausau,Rochester, WI Grand Junc .35 .4 .45 .5 .55 Gini Coefficient, 2000

Source: Gini coefficients are calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org. Murder rates are from the FBI’s Uniform Crime Reports.

61 388

Figure 21: Happiness and the Gini Coefficient, 2000 1

Grand Rapi .95

Portland-V Syracuse, Richmond-PAustin-San Minneapoli Gary,Milwaukee- IN Knoxville,Houston, T .9 Tacoma, WA Atlanta, Nashville,G Allentown-Norfolk-ViSeattle-Be Fresno, CA KansasPhoenix-MeCharlotte- Cit Tampa-St.Dallas, TX IndianapolAkron,Little OH RocSan Diego, West Palm Dayton-Spr Fort Wayne Baltimore,St.Columbia, Louis, Charleston Cincinnati SanNew Franci Orlean Washington Chicago,Fort I Laude Detroit,Cleveland-Buffalo-Ni M Philadelph Los Angele .85 Riverside- Orlando,Columbus, F Pittsburgh New York,

Share Self-reporting as Happy, 2000 Happy, as Self-reporting Share Tucson, AZ

Jackson, M .8 .4 .45 .5 .55 Gini Coefficient 2000

Source: Gini coefficients are calculated using the 5% Integrated Public Use Microdata Series (IPUMS) for 2000, at usa.ipums.org. Average level of happiness is calculated using data from the General Social Survey.

62 389 Plan B Luigi Zingales

fter pointing a gun to the head of infusion of equity capital can reassure the mar- implies transforming the existing, outstanding Congress, and threatening a fi- ket that the major banks will not fail, recreating debt (roughly two trillion dollars if we just count nancial meltdown in case his plan the confidence for banks to lend to each other. the long-term bonds) into safe debt. A large frac- was not approved, Treasury Sec- The piecemeal approach of $100 billion today, tion of the equity injected will not go to gener- retary Hank Paulson has finally $100 billion tomorrow used with AIG will not ate new loans, but to provide this insurance to Aarrived at the only logical conclusion: his plan work. It will only eat up the money, without the existing debtholders. How much? We can will not work. achieving the desired effect—without reassuring estimate it by looking at the credit default swaps Desperate for a Plan B, Paulson has slowly the market that the worst is over. (CDS), which provide us with the cost of insur- warmed to the suggestion of many economists: Simply stated, nothing short of a 5% in- ing the debt against default. At 10/9/08 prices, inject some equity into the banking system. Un- crease in the equity capital of the banking the cost of transforming these two trillion of fortunately, it is too little and too late. system will do the trick. We are talking about highly risky long-term bonds into bonds as risky The confidence crisis currently affecting the $600 billion, and probably more later, particu- as Barclay’s ones would be roughly $300 billion.1 financial system is so severe that only a massive larly if nothing is done to arrest the foreclosure Consequently, half of the capital the Government crisis that started all this. Unfortunately, even if will invest in banks will not go to increase new Luigi Zingales is the Robert C. McCormack Professor of the government is willing to spend this kind of loans, but to bail out Wall Street investors. Entrepreneurship and Finance, University of Chicago Graduate School of Business, has won the 2003 Bernacer money, there are three problems. Second, a capital infusion does not address Prize for the best European young financial economist, the First, to restore the necessary confidence, the root of the problem, which stems from the 2002 Nasdaq award for best paper in capital formation, and is author of Saving Capitalism from the Capitalists, a capital infusion needs to reduce the financial housing market. If homeowners continue to together with Raghuram G. Rajan. institutions’ risk of default to trivial levels. This default and walk away from their houses, the

© The Berkeley Electronic Press Economists’ Voice www.bepress.com/ev October, 2008 -- 390 banking sector will continue to bleed and ad- down on an interest-only loan. Unfortunately, what we economists call underinvestment: the ditional equity infusions will be needed in the during the last two years the value of your house failure to maintain the house. future. More importantly, the very bailout plan, dropped by 30%; thus, you now find yourself In the old days, when the mortgage was and the animosity it generates, will induce more with a mortgage worth $380,000 and a house granted by your local bank, there was a sim- homeowners who are sitting on a house with worth $280,000. Even if you can afford your ple solution to this tremendous inefficiency. a negative equity value to walk away. Many of monthly payment (and you probably cannot), The bank forgave part of your mortgage; let’s them will think: “Why do I have to play by the why should you struggle to pay the mortgage say 30%. This creates a small positive equity rules when Wall Street does not?” when walking away will save you $100,000, value—an incentive—for you to stay. Since This leads us to the third and most impor- more than most people can save in a lifetime? you stay and maintain the house, the bank tant problem. If we bail out Wall Street, why When the homeowner walks away, the gets its $266,000 dollars of the new debt back, not bail out Detroit (probably another $150 mortgage holder does not recover $280,000. which trumps the $140,000 that it was getting billion) and Main Street? In fact, Senator Mc- The foreclosure process takes some time dur- through foreclosure. Cain has already talked about buying out the ing which the house is not properly main- Unfortunately, this win-win solution is not defaulted mortgages to keep people in their tained and further deteriorates in value. The possible today. Your mortgage has been sold and homes. Even if we limit ourselves only to the recovery rate in standard mortgage foreclo- repackaged in an asset-backed security pool and subprime mortgages, we are talking about $1.3 sures is 50 cents per dollar of the mortgage. sold in tranches with different priorities. There trillion. Where do we stop? As these are not standard times, I am gener- is disagreement on who has the right to renego- We need a different solution: a “Plan B.” ous in estimating that under the current con- tiate and renegotiation might require the agree- Plan B should minimize the money the Gov- ditions it might recover 50 cents per dollar of ment of at least 60% of the debt holders, who are ernment uses in bailing out Wall Street and the appraised value of the house; right now; spread throughout the globe. This is not going Main Street to save our precious dollars for the that makes the recovery only 37 cents per to happen. Furthermore, unlike your local bank, stimulus package that will be essential to re- dollar of the mortgage, which given a house distant debt holders cannot tell whether you are starting the economy. appraised at $280,000 equals only $140,000 a good borrower who has been unlucky or some- for the mortgage holder. body just trying to take advantage of the lender. rescuing main street Foreclosing is costly for both the borrower In doubt, they do not want to cut the debt for fear uppose that you bought a house in California and the lender. The mortgage holder gains only that even the homeowners who can easily afford Sin 2006. You paid $400,000 with only 5% half of what is lost by the homeowners, due to their mortgage will ask for debt forgiveness. Economists’ Voice www.bepress.com/ev October, 2008 -- 391 Here is where government intervention can homeowners will behave as if they own 100% homeowners, who will have to announce their help. Instead of pouring money to either side, of it. It is only at the time of sale that 50% of intention in a relatively brief period of time. the government should provide a standardized the difference between the selling price and the The great benefit of this program is that it way to re-negotiate; one that is both fast and fair. new value of the mortgage will be paid back to provides relief to distressed homeowners at Here is my proposal. the mortgage holder. It seems a strange contract, no cost to the Federal government and at the Congress should pass a law that makes a re- but Stanford University has successfully imple- minimum possible cost for the mortgage hold- contracting option available to all homeowners mented a similar arrangement for its faculty: the ers. The other great benefit is that it will stop living in a zip code where house prices dropped university finances part of the house purchase defaults on mortgages, eliminating the flood of by more than 20% since the time they bought in exchange for a fraction of the appreciation houses on the market and thus reducing the their property. Why? Because there is no rea- value at the time of exit. downside pressure on real estate prices. By sta- son to give a break to inhabitants of Charlotte, The reason for this sharing of the benefits is bilizing the real estate market, this plan can help North Carolina, where house prices have risen twofold. On the one hand, it makes the renego- prevent further deterioration of financial insti- 4% in the last two years. tiation less appealing to the homeowners, mak- tutions’ balance sheets. But it will not resolve How do we implement this? Thanks to ing it unattractive to those not in need of it. For the problem of severe undercapitalization that two brilliant economists, Chip Case and Rob- example, homeowners with a very large equity these institutions are currently facing. For this ert Shiller, we have reliable measures of house in their house (who do not need any restructur- we need the second part of the plan. price changes at the zip code level. Thus, by ing because they are not at risk of default) will using this real estate index, the re-contracting find it very costly to use this option because they rescuing wall street option will reduce the face value of the mort- will have to give up 50% of the value of their eq- he plan for Wall Street follows the same gage (and the corresponding interest payments) uity. Second, it reduces the cost of renegotiation Tmain idea: facilitating an efficient renego- by the same percentage by which house prices for the lending institutions, which minimizes tiation. The key difference between the Main have declined since the homeowner bought (or the problems in the financial system. Street and Wall Street plans is in the ease of as- refinanced) his property. Exactly like in my hy- Since the option to renegotiate offered by sessing the current value of the troubled assets. pothetical example above. the American Housing Rescue & Foreclosure It is relatively easy to estimate the current value In exchange, however, the mortgage hold- Prevention Act does not seem to have been stim- of a house by looking at the purchase price and er will receive some of the equity value of the ulus enough, this re-contracting will be forced at the intervening drop in value (per the Case house at the time it is sold. Until then, the on lenders, but it will be given as an option to and Shiller index). In banks, however, the lack

Economists’ Voice www.bepress.com/ev October, 2008 -- 392 of transparency makes this estimation very dif- By transforming all banks’ debt into equity this voluntarily initiate these special bankruptcy ficult. To avoid having to come up with this esti- special Chapter 11 it would make banks solvent proceedings? One way is to harness the power mate, which would be a difficult process and one and ready to lend again to their customers. of short-term debt. By involving the short- fraught with potential conflict of interests, we are Certainly, some current shareholders might term debt in the restructuring, this special going to use a clever mechanism invented twen- disagree that their bank is insolvent and would bankruptcy will engender fear in short-term ty years ago by Lucian Bebchuk, a lawyer and feel expropriated by a proceeding that wipes creditors. If they think the institution might economist who teaches at Harvard Law School. them out. This is where the Bebchuk mecha- be insolvent, they will pull their money out as The core idea is to have Congress pass a law nism comes in handy. After the filing of the soon as they can for fear of being involved in that sets up a new form of prepackaged bank- special bankruptcy, we give these sharehold- this restructuring. In so doing, they will gener- ruptcy that would allow banks to restructure ers one week to buy out the old debtholders ate a liquidity crisis that will force these insti- their debt and restart lending. Prepackaged by paying them the face value of the debt. tutions into this special bankruptcy. means that all the terms are pre-specified and Each shareholder can decide individually. If he An alternative mechanism is to have the banks could come out of it overnight. All that thinks that the company is solvent, he pays his Fed limit access to liquidity. Both banks and in- would be required is a signature from a federal share of debt and regains his share of equity. vestment banks currently can go to the Federal judge. In the private sector the terms are gener- Otherwise, he lets it go. Reserve’s discount window, meaning that they ally agreed among the parties involved; the in- My plan would exempt individual deposi- can, by posting collateral, receive cash at a rea- novation here would be to have all the terms tors, who are federally ensured. I would also sonable rate of interest. Under my plan, for the pre-set by the government, thereby speeding up exempt credit default swaps and repo con- next two years only banks that underwent this the process. Firms who enter into this special tracts to avoid potential ripple effect through special form of bankruptcy would get access to bankruptcy would have their old equityholders the system (what happened by not directing the discount window. In this way, solid finan- wiped out and their existing debt (commercial Lehman Brothers through a similar procedure). cial institutions that do not need liquidity are paper and bonds) transformed into equity. This It would suffice to write in this special bank- not forced to undergo this restructuring, while would immediately make banks solid, by provid- ruptcy code that banks who enter it would not insolvent ones would rush into it to avoid a ing a large equity buffer. As it stands now, banks be considered in default as far as their con- government takeover. have lost so much in junk mortgages that the tracts are concerned. Another problem could be that the institu- value of their equity has tumbled nearly to zero. How would the government induce in- tions owning the debt, which will end up own- In other words, they are close to being insolvent. solvent banks (and only insolvent ones) to ing the equity after the restructuring, might be

Economists’ Voice www.bepress.com/ev October, 2008 -- 393 prohibited by regulation or contract to hold- to see public money thrown at the problems. value of the loans as a result of the equity infusion. These estimates depend crucially upon the level of ing equity. To prevent a dumping of shares that However, the cost is quickly escalating. If we do riskiness we assume bank debt will have after the would have a negative effect on market prices, not stop, we will leave an unbearable burden of equity infusion. As a benchmark I use the level of we could allow these institutions two years to debt to our children. riskiness of Barclay’s debt (which on 10/9/08 had a sell their unsought equity. This was the stan- The time has come for the Treasury secre- CDS of 95.8 bp). dard practice in the old days when banks, who tary to listen to some economists. By under- could not own equity, were forced to take some standing the causes of the current crisis, we can references and further reading in a restructuring. help solve it without relying on public money. Bebchuk, Lucian (1998) “A New Approach to The beauty of this approach is threefold. Thus, I feel it is my duty as an economist to Corporate Reorganization,” Harvard Law Re- First, it recapitalizes the banking sector at no provide an alternative: a market-based solu- view, 101:775–804. cost to taxpayers. Second, it keeps the govern- tion, which does not waste public money and ment out of the difficult business of establishing uses the force of the government only to speed the price of distressed assets. If debt is convert- up the restructuring. It may not be perfect, but ed into equity, its total value would not change, it is a viable avenue that should be explored only the legal nature of the claim would. Third, before acquiescing to the perceived inevitabil- this plan removes the possibility of the govern- ity of Paulson’s proposals. ment playing God, deciding which banks are allowed to live and which should die; the market will make those decisions. Letters commenting on this piece or others may be submitted at http://www.bepress.com/cgi/ tomorrow is too late submit.cgi?context=ev. he United States (and possibly the world) is facing the biggest financial crisis since the T notes Great Depression. There is a strong quest for the 1. By using the maturity structure of the debt outstand- government to intervene to rescue us, but how? ing, I estimate what the cost savings (in terms of reduced insurance costs) for the lenders are. Even if Thus far, the Treasury seems to have been fol- not all lenders fully insure, the reduction in the cost lowing the advice of Wall Street, which is happy of insurance can be considered as an increase in the

Economists’ Voice www.bepress.com/ev October, 2008 --