II [~IF~ MAY/JUNE 1995 R. Glenn Hubbard is Russell 1. Carson professor of economics and finance at Columbia University. The author is grateful to Allen Berger, Phillip Cogan, Richard Cantor, Mark Gertler, Simon Gilchrist, Anil Kashyop, Don Morgan, Glenn Rudebusch, Bruce Smith and participants at the confer- ence for helpful comments and suggestions. The authar also acknowledges financial support from the Center for the Study of the Economy and the State of the University of Chicaga, and the Federal Reserve Bank of New York. sidestep the credit view language per se, and Is There a instead focus on isolating particular frictions in financial arrangements and on developing “Credit Channel” testable implications of those frictions, To anticipate that analysis a bit, I argue that for Monetary realistic models of “financial constraints” on firms’ decisions imply potentially significant Policy? effects of monetary policy beyond those working through conventional interest rate channels. Pinpointing the effects of a narrow P. Glenn Hubbard “bank lending” channel of monetary policy is more difficult, though some recent models nderstanding the channels through which and empirical work are potentially promising / monetary pohcy affects economic van in that regard. ahies has long been a key research topic I begin by reviewing the assumptions in macroeconomics and a central element of and implications of the money view of the economic policy analysis. At an operational monetary transmission mechanism and by level, a “tightening” of monetary policy by the describing the assumptions and implications Federal Reserveimplies a sale of bonds by the of models of financial constraints on borrow- Fed and an accompanying reduction of bank ers and models of bank-dependent borrow- reserves. One question for debate in academic ers- The balance of the article discusses the and public policy circles in recent years is transition from alternative theoretical models whether this exchange between the central of the transmission mechanism to empirical bank and thebanking system has consequences research, and examines implications for in addition to those for open market interest monetary policy rates. Ac the risk of oversimplifying the debate, the question is often asked as whether the /7/ traditional interest rate or “money view” channel presented in most textbooks is aug- — 7 mented by a “credit view” channel.’ Before discussing predictions for the There has been a great deal of interest in effects of alternative approaches on monetary this question in the past several years, moti- policy, it is useful to review assumptions vated both by developments in economic about intermediaries and borrowers in the models (in the marriage of models of infor- traditional interest rate view of the monetary mational imperfections in corporate finance transmission mechanism. In this view, finan- with traditional macroeconomic models) and cial intermediaries (banks) offer no special recent events (for example, the so-called credit services on the asset side of their balance sheet - ‘For descriptons of the debate, see crunch during the 1990-91 recession),’ As I On the liability side of their balance sheet, Bernanke and Blinder (19f8) and elaborate below, however, it is not always banks perform a special role: The banking Bernanke (1993). straightforward to define a meaningful credit system creates money by issuing demand view alternative to the conventional interest deposits. Underlying assumptions about ‘for on analysis of the ‘credit crnnch’ episode, see Kliesen and rate transmission mechanism. Similar diffi- borrowers is the idea that capital structures Totam (1992) and the studies in culties arise in structuring empirical tests of do not influence real decisions of borrowers the Federal feserne Bank of New credit view models, and lenders, ahe result of Modigliani and York (1994). The paper by Cantar This paper describes and analyzes a broad, Miller (1958). Applying the intuition of the and Rndrigues in the New York Fed though still well-specified. version of a credit Modigliani and Miller theorem to banks, Fama studies considers the possibility of a view alternative to the conventional monetary (1980) reasoned that shifts in the public’s credit cmoch far nonbonk interme- transmission mechanism. in so doing, I portfolio preferences among bank deposits, diaries. FEDEEAL RESERVE SANK OF ST. LOUIS 63 DII~I1t~ MAY/JUNE 199$ bonds or stocks should have no effect on real in short-term rates is not obvious a priori Fama’s insight amplifies the endier outcomes; that is, the financial system is based on conventional models of the term contribution of Brainard and Tobin merely a veil.’ structure, Empirical studies, however, have that monetary policy con (1963) To keep the story simple, suppose that documented a significant, positive relationship its effects oe 4 he analyzed through there are two assets—money and bonds, between changes in the (nominal) federal investor portfalias. In a monetary contraction, the central bank funds rate and the 10-year Treasury bond rate Mare fenerolly, in a model with reduces reserves, hmiting the banking system’s (see, for example, Cohen andWenninger, 1993; many assets, this description would ability to sell deposits. Depositors (house- and Escrella and Hardouvelis, 1990). Finally assign to the money view of the holds) must then hold more bonds and less although many components of aggregate effects transmission mechanism on money in their portfolios. If prices do not demand are arguably interest-sensitive (such spending arising fmm any changes instantaneously adjust to changes in the money as consumer durables, housing, business fixed in the relative prices of assets, supply the fall in household money holdings investment, and inventory investment), output While this simple tmoasset-mndel represents adechne in real money balances, To responses to monetary innovations are large description of the money view is highly stylized, it is consistent with restore equilibrium, the real interest rate on relative to the generally small estimated effects a number of alternative models bonds increases, raising the user cost of capital of user costs of capital on investment.’ beyond the texihank IS-IM model for a range of planned investment activities, I shall characterize the money view as (see, for example, Hubbard, 1994), and interest-sensitive spending falls.’ focusing on aggregate, as opposed to distribu- including dynamic-equilibrium cash- While the money view is widely accepted tional, consequences of policy actions. In in-advance madels (far example, as the benchmark or “textbook” model for this view, higher default-risk-free rates of Rotemberg, 1984; and Christiano analyzing effects of monetary policy on eco- interest following a monetary contraction and fichenbaum, 1992). nomic activity, it relies on four key assump- depress desired investment by firms and ‘Far an empidral descripton of this tions: (1) The central bank must control the households. While desired investment falls, transmission mechanism in the con- supply of “outside money,” for which there are the reduction in business and household text of the Federal Reserve’s forecast- imperfect substitutes; (2) the central bank can capital falls on the least productive projects. ing model, see Mauskapf (1990). affect real as well as nominal short-tenn interest Such a view offers no analysis of distribu- 6 See, for example, “limited partici rates (that is, prices do not adjust instanta- tional, or cross-sectional, responses to policy ponien’ models as in Lucas (1990) neously); (3) policy-induced changes in real actions, nor of aggregate implications of this and Chrisfana and lichenbaum short-term interest rates affect longer-term heterogeneity I review these points not to (1992). interest rates that influence household and suggest that standard interest rate approaches See, for example, analyses of business spending decisions; and (4) plausible to the monetary transmission mechanism are incorrect, but to suggest strongly that one innentory irvestnnent ir Kashyap, changes in interest-sensitive spending in Stein and Wilcox (1993) and response to a monetary policy innovation ought to expect that they are incomplete. Gerfer and Gilchist (1993). See match reasonably well with observed output also the review of empirical studies responses to such innovations. HOW REASONABLE IS THE of business fined investment in In this stylized view, monetary policy is Chirinko (1993) and Cammins, represented by a change in the nominal supply CREDIT VIEW? Hassett and Hubbard (1994). of outside money Of course, the quantity The search for a transmission mechanism a This current fashion actually has of much of the monetary base is likely to be broader than that just described reflects two long pedigree in macroecenemics, endogenous.° Nonetheless, legal restrictions concerns, one “macro” and one “micro,” The with iniporlont cantrihutions by Fisher (for example, reserve requirements) may macro concern, mentioned earlier, is that (1933), Gudey and Show (1955, compel agents to use the outside asset for cyclical movements in aggregate demand— 960), Minsky (1964, 19/5) and some transactions, In practice, the central particularly business fixed investment and Walnilawer 11980), Some econa- bank’s influence over nominal short-ternn inventory investment—appear too large to metric ferecosfing medels have also interest rates (for example, the federal funds be explained by monetary policy actions that focosed on financial factors