How Little We Know: The Challenges of Financial Reform Russell Roberts

hen an airplane crashes, to financial crashes. After all, pre- of both investors and the taxpayers, leading to expert investigators probe sumably know more than we did in the past. suggestions for caps on executive compensa- the cause of the crash. We have ever more data and ever more sophis- tion or changes in the mix between cash and Their analysis can lead ticated techniques for analyzing the data. Yet other forms of compensation. As an example, to changes in aircraft de- there is no evidence to suggest that financial the Obama Administration pay czar, Kenneth Wsign, flight procedures, and regulations in market regulation is more effective than in the Feinberg, has limited take-home pay for ex- hopes of reducing the likelihood of a crash in past. If anything, the opposite appears to be ecutives at firms receiving government assis- the future. The growth of knowledge in the the case. tance and offset the reduction with increases airline industry has been extremely produc- This discouraging empirical record does in stock that would not be accessible for two tive. Between 1989 and 2008, there was a not seem to hamper the unending stream of to four years. The hope is to encourage the seven-fold reduction in the probability of a ideas for what might make our financial sys- growth of long-term investing instead of riski- fatal crash. tem more secure, based on an analysis of what er short-term bets. There is a natural tendency for economists went wrong this time. It is obvious, for ex- Despite the spectacular failure of Fannie (and even for normal people) to presume that ample, that excessive leverage played a role in Mae and Freddie Mac, some economists insist similar analytical techniques can be applied the vulnerability of the firms that collapsed in that Fannie and Freddie need to be kept in 2008. The natural response is to increase capi- place but somehow, just made safer. This op- Russell Roberts is professor of economics at George Mason tal requirements. timistic advocacy—which assumes that Fan- University, a scholar, and a research scholar at ’s . He blogs at Cafe The incentives of management on Wall nie and Freddie are like airplanes that need Hayek and hosts the weekly podcast EconTalk. Street appear to be out of line with the interests better landing gear—is in spite of the fact that

© The Berkeley Electronic Press The Economists’ Voice www.bepress.com/ev November 2009 -- between 1992 and 2008 Fannie and Freddie Landing gear that is more reliable makes an concede that such a top-down enterprise is had their own regulator, the Office of Federal airplane safer, but reducing risk or changing inherently flawed. Housing Enterprises Oversight, that failed incentives in one part of the financial system Economists who treat the financial system to stop the meltdown of Fannie and Freddie can cause less transparent changes elsewhere like an airplane ignore the symbiotic dance that has cost the U.S. taxpayer about $100 that reduce stability. Recognizing the meager- between politicians on the one hand and fi- billion and counting. Somehow, this time ness of our understanding would be a good nancial institutions on the other. will be different. starting place for what should be a humble The revolving doors between Fannie Mae approach to financial reform. and government, and between Goldman emergent order, feedback loops and unin- For example, advocating ‘better’ super- Sachs and government, are only the most ob- tended consequences vision or ‘more vigorous’ regulation or even vious indicators that something is wrong. A am not so optimistic about reforming the something more specific such as larger capital Martian impartially observing the U.S. finan- I government sponsored enterprises or the cushions, ignores the empirical record that cial system would conclude that it is run to system of regulation and supervision of Wall regulators and politicians either for venal benefit the executives and protégés of Gold- Street. Our financial system is unlike an air- or human reasons, seem unable to maintain man Sachs. This is not a good thing even if it plane. It is complex in a way that makes even these restrictions in the face of pressure from is not true. If it is true, then any meaningful an airplane look simple. A synthetic CDO has participants. Even detailed, specific plans are financial reform must start with finding ways something of the complexity of an airplane. unlikely to succeed because of feedback loops to break that symbiosis. The financial system is much more complex. that are present in all emergent systems. Most reforms ignore that symbiosis, con- It is an emergent system, where outcomes are Hayek said: “The curious task of econom- demn capitalism as inherently unstable, and the result of a dynamic interaction between ics is to demonstrate to men how little they look for ways to artificially create the right investors, regulators, and politicians. really know about what they imagine they incentives. But the government, because Economists have a very imperfect under- can design.” There are few better illustra- of that symbiosis and other political incen- standing of this interaction. Because of the tions of this than the attempts to engineer a tives, played a significant role in reducing feedback loops between the different actors in financial system that preserves the incentives the stability of the system and perverting the the system, we also have a very imperfect un- to take risks while at the same time preserv- incentives that would naturally emerge. The derstanding of how changing one piece of the ing prudence. We imagine we can design a references provide some recent work that dis- system interacts with the rest of the system. better financial system. Perhaps it is time to cusses this in more detail.

The Economists’ Voice www.bepress.com/ev November 2009 -- the consequences of rescuing creditors of of their companies’ stocks from earlier highs. For these natural feedback loops to emerge, we large financial institutions Surely this limits the moral hazard problem. must get rid of the doctrine of ‘too big to fail’ he rescues of the last 25 years, starting with But Cayne and Fuld could not access those or maybe ‘too connected to fail’ or at least ‘too TContinental Illinois, up through the Fed- paper profits at will. More importantly, each connected to Goldman Sachs to fail.’ orchestrated intervention to prevent the sud- man is worth in the neighborhood of $500 But even otherwise optimistic economists den collapse of Long Term Capital Management million, wealth they accumulated through understand that this prescription is unlikely and the implicit guarantee of Fannie and Fred- cash compensation and the occasional judi- to succeed given the incentives of politicians die, reduced the incentive of counterparties, cious sale of their companies’ stock in advance and policymakers to avoid short-run damage particularly lenders, to restrain the highly-lev- of the crash. Yes, there was a lot of myopia in favor of meltdowns that come to pass in the eraged bets that ended up wreaking so much and overconfidence on Wall Street. But when long-run, where they may not be dead but are financial havoc. people spend their own money, they spend it at least out of office. Yet without rescinding It’s obvious why equity holders like debt. more carefully. Cayne and Fuld doubled down the implicit policy of bailing out the creditors But why didn’t debt issuers restrain risk-taking with other people’s money. They were more of large financial institutions, there will be in- as Wall Street firms pursued riskier invest- prudent with their own. adequate incentives to restrain excessive lever- ments? Part of the reason is that they antici- Reasonable people can debate the mag- age and the imprudence that inevitably follows pated the possibility of government rescue, nitude of the responsibility that past rescues when people are allowed to gamble with other particularly as their lending made them in- played in the excessive risk-taking that de- people’s money. Attempts to repair the system creasingly entangled with each other. stroyed much of Wall Street. But the subse- from the top-down will fail. We must find ways The importance of moral hazard, de- quent rescue of Bear Stearns, Merrill, Citigroup, to let bottom-up solutions emerge. scribed presciently in 2004 by Gary Stern and Fannie and Freddie and AIG have certainly set Ron Feldman, is often greeted with skepticism significant expectations for the future. what can be done? because of what seems to be the restraints on The solution seems simple. Rather than try hese observations suggest a very modest recklessness imposed by equity holders. After to turn this dial or push that lever connected Tprogram for reform: all, the skeptics point out, Jimmy Cayne, the to some part of the system the optimal amount Don’t try to re-create the old system while CEO of Bear Stearns and Richard Fuld, the (holding everything else constant, somehow) trying to make it ‘better.’ There is a natural CEO of Lehman Brothers each lost $1 billion we should let natural feedback loops emerge wariness about securitization right now. That (yes, with a ‘b’) from the collapse of the value that encourage prudence as well as risk-taking. is good. Let it blossom. There is a natural The Economists’ Voice www.bepress.com/ev November 2009 -- wariness about zero-down mortgages. That is In addition, the Fed has played a major it, the political incentives are going to stay good. Let it blossom. role in exacerbating the moral hazard prob- in place. Economists play an important role Recognize that having every American own a lem. It needs to be restrained rather than in how people perceive what has happened. home is not the American Dream but the dream empowered. It is not good for a democracy We should stop being the enablers of such of the National Association of Home Builders to have an agency as unaccountable as the obscene transfers of wealth. and the National Association of Realtors. Any Fed acquire even more power and use it in ad Policymakers who make creditors and lend- government programs to increase home own- hoc ways. ers whole should be excoriated, condemned and ership should be funded out of current tax Capitalism is a profit and loss system. The called to account rather than praised and hon- dollars where the costs are visible. Get rid of profits encourage risk-taking. The losses en- ored. Zero cents on the dollar for bankrupt mandates such as the Community Reinvest- courage prudence. These natural feedback bets made by lenders and creditors would be ment Act. Get rid of Fannie and Freddie and loops have been distorted by the rescues of ideal but is unlikely to be a credible promise. the implicit guarantee that turned out to be the past and the present. We need to take the So let’s start more modestly. A ceiling of 50 only too real. ‘crony’ out of crony capitalism. That will not cents on the dollar for creditors and lenders Be aware that the Fed is certainly part of be easy. But economists should not make it when the institutions they fund become insol- the problem and may not be part of the solu- more difficult. The near universal praise by vent is a natural place to start. Even this may tion. The Fed created the artificially low in- economists for the actions of Bernanke, Paul- be too difficult for politicians to stomach. But terest rates that helped inflate the housing son, and Geithner and the near universal con- economists should at least preach the virtues bubble. The Fed then raised interest rates demnation by economists of the decision to of letting creditors lose money when they fi- too quickly with disastrous effects for the ad- let Lehman Brothers enter bankruptcy, greatly nance imprudent risk-takers. justable-rate mortgages encouraged by their reduces the credibility of any promise by pol- We are what we repeatedly do. Not what low-interest rate policy. Monetary policy icymakers to act differently in the future. we say we are. Not what we’d like to be. But should not be left to any self-proclaimed or Over the last 15 months, average Ameri- what we do. What we do as a body politic publicly-anointed maestro. Following an au- cans have sent hundreds of billions of dollars is rescue rich people from the consequences tomatic money growth rule or the Taylor rule to some of the richest people in human his- of their decisions. That is bad for democracy would have avoided much of the pain. Some- tory. The better the citizenry understands this and bad for capitalism. Until we fix that, we body needs to hold the Fed accountable for reality, the better chance the political incen- as citizens are playing a game of “heads— funding exuberance. tives will change. If people don’t understand Wall Street executives win a ridiculously

The Economists’ Voice www.bepress.com/ev November 2009 -- enormous amount, tails—they just win a ri- October 6. Available at http://online.wsj.com/ar diculous amount, paid for by the rest of us.” ticle/SB125478783753066235.html. Until we fix that, little else matters. Stern, Gary and Ron Feldman (2004) Too Big to Fail: The Hazards of Bank Bailouts. Brookings Institution Press: Washington D.C. (Reissued Letters commenting on this piece or others may in 2009.) be submitted at http://www.bepress.com/cgi/ Thompson, Andrea (2009) “Flying is Safer than submit.cgi?context=ev. Ever” Live Science, June 1. Available at: http:// www.livescience.com/culture/090601-air-crash references and further reading es.html. Kling, Arnold (2009) “Not What They Had in Mind: A History of Policies That Produced the Fi- nancial Crisis of 2008,” Mercatus Center, . Available at http://mercatus. org/PublicationDetails.aspx?id=28118. Lucchetti, A., David Enrich, and Joann Lu- blin (2009) “Fed Hits Banks With Sweeping Pay Limits” Wall Street Journal, October 23. Available at: http://online.wsj.com/article/ SB125623026446601619.html. Roberts, Russell (2009) “Gambling with Other People’s Money: How Policy Mistakes Created Perverse Incentives and the Financial Cri- sis of 2008,” Mercatus Center, George Mason University. Forthcoming. Solomon, Deborah (2009) “Pay Czar Tar- gets Salary Cuts” Wall Street Journal.

The Economists’ Voice www.bepress.com/ev November 2009 --