Banks and security markets.

Gorton, Gary

What do banks do (that markets cannot do)? In the United States in 1845, the answer would have been that banks made loans and issued mortgages, but their most important role was to provide a medium of exchange by issuing private money.(1) By the late 19th century, U.S. capital markets were more developed, but at the same time large banks resembled German universal banks.(2) Passage of the Glass-Steagall Act in 1934 changed that by restricting banks' activities.(3) In 1996 banks face competition from money market mutual funds for deposit business and from junk bonds, commercial paper, and medium-term notes for bank loans.(4) While smaller firms continue to rely heavily on banks, banks are now engaged in many new activities, such as interest rate and currency swaps.(5) Whatever it is that defines banks as unique institutions, the pattern of bank activities has changed over the last 150 years as banking has interacted with the development of security markets. The challenge is to explain the persistence of banking as security markets increasingly develop.

Banks and Security Markets in the Organization of Capitalist Economies

One problem in understanding what banks do is that the function of securities markets is not well understood. With James Dow, I address the issue of the connection between stock market price "efficiency" and economic efficiency.(6) In this work, we assume that firms are operated by managers who must be compensated in a way that induces them to find desirable investment projects. The managers may make an effort to produce information, but are not always successful in receiving information. When they are not successful, they may be induced by their compensation contracts to rely on inferences drawn from changes in their firms' stock market prices. In such an economy, two types of information must be produced and transmitted in order to achieve the most efficient allocation of resources. First, the stock market must provide forward-looking or prospective information when informed traders produce information about the firm's investment opportunities for managers to act on. Second, the stock market must provide backward-looking or retrospective information when stock prices reflect informed traders' production of information about the outcomes of investment decisions made in previous periods. Managerial compensation based on stock prices then can induce managerial effort.

This model of the stock market seems to be what most economists have in mind when they speak of "market efficiency." Stock prices allocate resources by influencing investment decisions and by providing a way to monitor corporate managers. But, consider the same economy without the stock market but with banks instead. Banks design contracts to hire information-producing loan analysts who write prospective and retrospective reports about investment opportunities and managerial performance. Dow and I show that the banks can implement the same allocation as the efficient equilibrium of the stock market economy. Efficient security prices are neither necessary nor sufficient for economic efficiency.

That the savings-investment process might be equally well organized around banks as around security markets suggests empirically investigating the role of banks in economies where securities markets are less important than in the United States, as in Germany. For much of recent German history the stock market has been small and illiquid. German banks, though, can own stock legally. The question is whether bank block shareholding is, in some sense, a substitute for a liquid stock market. Schmid and I examine the role of banks in Germany, and show that in the 1970s firm performance was better when a bank was a large shareholder. This is consistent with the proposition that when banks obtain a block of stock (via a family selling out, or because of financial distress), they have an incentive to improve firm performance by monitoring because, effectively, the block cannot be sold (since the stock market is traded so thinly).(7) By the 1980s, however, German capital markets had developed further, and this role of banks was no longer present.(8)

Bank Uniqueness

Banks and securities markets may be substitutes but, even in economies with highly developed capital markets, banking persists as an important institution. This suggests that banks perform some tasks that markets cannot accomplish, even when they are highly developed.

On the asset side of the balance sheet, banks originate loans. To the extent that bank loans are held by the bank (so that bank equity is at risk), there is an incentive to oversee the activities of borrowers to maintain the value of the loan. Because of free-riding, it is difficult for a large number of debt holders to interact with borrowers during certain states of the world: "monitoring" them, for example, when they are distressed, or verifying that they can repay the loan. A bank, by concentrating the debt, eliminates this problem.(9) Beyond this, the details of what "monitoring" really means are fuzzy. Moreover, the argument about concentration would seem to apply to all debt and so cannot explain the role of bank loans as distinct from corporate bonds.

The renegotiability of bank loans emanates from a contract provision that gives banks the right to seize collateral, and from the fact that a single agent is in a position to renegotiate. Kahn and I consider the optimality of this contract provision.(10) We model the interaction between a bank and a borrower when: 1) the borrower may have an incentive to (at a cost) increase the risk of a project if the project goes badly; while 2) the bank may have an opportunistic incentive to threaten early termination of the project. When the borrower seeks to add risk, the bank may respond by forgiving some debt to eliminate the borrower's incentive to add risk, liquidating the loan by seizing the collateral, raising the interest rate, or doing nothing. All of these outcomes happen in equilibrium. In fact, the variance of the value of the firm is state dependent in equilibrium, a result that has implications for the application of option-pricing methods to corporate securities. The contract provision allowing the bank the right to initiate renegotiation by threatening to seize the collateral is optimal when this type of renegotiation, that is, monitoring, is valuable ex ante.

Banks not only provide unique services on the asset side of the balance sheet, but they also produce a medium of exchange on the liability side of the balance sheet. Pennacchi and I explain this.(11) If agents face unanticipated needs to consume, and face a cash-in-advance constraint, then they will have to dissave by selling securities to obtain cash. When they sell securities, they may sell in a market where better-informed traders take advantage of this liquidity need. The uninformed consumers lose money, on average, when they are less informed about the value of the risky securities they are selling. Their loss to informed traders is increasing in the variance of the value of the security being sold. Therefore, a low-variance security or, in particular, a riskless security with a known value, would minimize or avoid such losses. Banks produce such a riskless trading security by issuing debt, which is a claim on a diversified portfolio (of loans). If the debt is not riskless, the government can improve matters with deposit insurance.(12)

Panics and the Origin of Bank Regulation

The combination of demand deposits, which can be redeemed at par on demand, with nontraded bank loans can create the possibility of an event in which depositors en masse exercise their right to demand cash for their deposits. In the United States large numbers of relatively undiversified banks issuing demand deposits faced repeated banking panics of this type. There is nothing mysterious about banking panics. I show that in the United States banking panics occurred at the peak of the business cycle when consumers received information forecasting a recession.(13) At the peak, consumers know that they will want to dissave in the coming recession. Their savings are in banks, some of which will fail during the recession. Because of asymmetric information about the value of the nontraded bank loans, depositors cannot distinguish the banks that will fail from those that will not. As a result, when they receive information forecasting a recession, rational, risk-averse depositors withdraw from all banks. This event is a banking panic.

During the 19th century, banks formed coalitions - clearinghouses - partly to address the problem of banking panics. Clearinghouses monitored member banks by restricting their activities, conducted strict bank examinations, and enforced sanctions against members to enforce compliance. During panics clearinghouses organized suspensions of the payment of cash to honor demand deposits; instead of paying out cash, they provided a form of deposit insurance by issuing their own private money (claims on the clearinghouse) to honor deposit contracts. Since this money was a claim on the clearinghouse, depositors were insured against the failure of any particular member bank, although not against the failure of the clearinghouse (which never occurred in U.S. history).(14)

Panics can be seen in the context of the industrial organization of banking. Demand deposits are a medium of exchange that clears internally in the banking system, not externally through trade in a market. Internal clearing closed the external market in which bank claims were traded, the banknote market of the pre-Civil War Era. But this caused an information asymmetry, since there was no longer any information-revealing market about the value of banks.(15) But, without an information-revealing market, how are depositors to monitor banks? Banking panics can be seen as a monitoring mechanism and, in this sense, were desirable.(16) The information asymmetry created the necessary condition for panics, but also the incentives for the private provision of bank regulation, examination, and insurance. Ultimately, government bank regulation and insurance took over the clearinghouse functions.

Current Regulatory Issues

In the 1980s, while a number of new debt markets opened or grew significantly (including junk bonds and commercial paper), banks failed at increasing rates as they became unprofitable. One widespread explanation for the high failure rate of banks involves the moral hazard attributable to underpriced deposit insurance. In this view, bank shareholders have an incentive to take on risk when the value of the bank charter falls sufficiently.(17) This view is inconsistent with banks being run by managers and, it turns out, with empirical evidence on which types of banks want to take on inefficient risk.

Rather than assume that shareholders directly control bank actions, Rosen and I assume that bank managers, who may own a fraction of the bank, make the lending decisions.(18) If managers have different objectives than outside shareholders, and disciplining managers is costly, then managerial decisions may be at odds with the decisions that outside shareholders would like them to take.(19) We show empirically that, contrary to the moral hazard view, excessive risktaking by banks occurred at those banks controlled by managers with stockholdings well below 50 percent, but sufficiently high that they constituted important blockholders. This result suggests that a failure in the market for corporate control in banking can explain the persistence of unprofitability of banking in the 1980s.

Another explanation for the persistence of bank failures during the 1980s concerns "regulatory forbearance," that is, the unwillingness of regulators to close insolvent banks. This view raises more general welfare questions concerning bank regulation. What is the objective function of regulators? What should they do when banks become riskier? Winton and I consider the question of bank capital requirements in a general equilibrium setting.(20) General equilibrium imposes the discipline that the capital in banks must come from somewhere in the economy. We show that there are unique costs associated with bank capital, and that regulators optimally will not, indeed cannot, force banks to raise costly capital. The basic argument is that consumers need a risk-less transactions medium supplied by banks, as I have discussed here. Holding bank equity exposes consumers to possible losses should they need to sell the equity. To the extent that they must hold bank equity, and not demand deposits, they face losses if they have unanticipated needs requiring them to sell their bank equity. But this risk is priced and so imposes a cost on equity that is unique to the banking industry. We show that capital requirements never can be binding: if they are too onerous they can be avoided by exit from the banking industry. But then banks do not supply the socially valuable services that markets cannot supply. To avoid such socially undesirable exit, the regulators may take actions that resemble forbearance. This, however, is socially optimal.

Recent Developments in Banking

Banking has been transformed in the last 15 years. One major change has been the opening and growth of the market for loans. According to theory, bank loans are not liquid: no one should buy a loan, because then banks will lack incentives for monitoring. Moreover, if loans and bonds are substitutes, a direct contract with a firm dominates purchasing a loan, since the buyer of a loan relies on the bank for representation if the firm goes bankrupt. Yet the bank, having sold the loan, would appear to have little incentive to perform. Despite this, the market for such loans is now enormous.(21) Pennacchi and I empirically search for implicit contract features that make loan sales compatible with incentives.(22) We find some evidence that banks selling loans keep a portion of the loan, and that the price of the loan being sold reflects this.

Another major change in banking has been the advent of the derivatives market, a market with commercial banks at its center. Derivatives have been controversial because of the difficulty in valuing them. Rosen and I investigate the involvement of U.S. commercial banks in the market for interest rate swaps.(23) We find that banks, generally speaking, do not appear to be taking on excessive risk in this market. Banks seem to have small net positions in derivatives.

1 See G. Gorton, "Reputation Formation in Early Bank Note Markets, "Journal of Political Economy 104 (1996), pp. 346-397; and "Pricing Free Bank Notes," NBER Working Paper No. 3645 (1990, out of print).

2 See C. Calomiris, "The Costs of Rejecting Universal Banking: American Finance in the German Mirror, 1870-1914," in The Coordination of Activity Within and Between Firms, N. Lamoreaux and D. M. G. Raff, eds. Chicago: University of Chicago Press, 1995.

3 On the Glass-Steagall Act, see R. Kroszner and R. Rajan, "Is the Glass-Steagall Act Justified? A Study of the U.S. Experience with Universal Banking Before 1933, "American Economic Review 84 (1994), pp. 810-832; "Organizational Structure and Credibility: Evidence from the Underwriting Activities of Commercial Banks Before Glass-Steagall, "NBER Working Paper No. 5256, March 1995: and M. Puri, "Commercial Banks in Investment Banking: Conflict of Interest or Certification Role?" Journal of Financial Economics 40 (1996). pp. 373-401.

4 See G. Gorton and G. Pennacchi, "Money Market Funds and Finance Companies: Are They the Banks of the Future?" in Structural Change in Banking, M. Klausner and L. J. White, eds. Homewood, IL: Business One-Irwin, 1993. Also see L. Beneveniste, M. Singh, and W. J. Wilhelm, "The Failure of Drexel Burnham Lambert: Evidence on the Implications for Commercial Banks," Journal of Financial Intermediation 3 (1993), pp. 104-137. For surveys of recent developments in banking, see A. Berger, A. Kashyap, and J. Scalise, "The Transformation of the U.S. Banking Industry: What a Long Strange Trip It's Been, "Brookings Papers on Economic Activity 2, (1995); and J. Boyd and M. Gertler, "U.S. Commercial Banking: Trends, Cycles, and Policy," in NBER Macroeconomics Annual 1993, O. Blanchard and S. Fischer, eds. Cambridge, MA: MIT Press.

5 On loans to small business, see M. Petersen and R. Rajan, "The Benefits of Lending Relationships: Evidence from Small Business Data, "Journal of Finance 49 (1994), pp. 3-37; and "The Effect of Credit Market Competition on Lending Relationships, "Quarterly Journal of Economics 110 (1995), pp. 407-443. On swaps, see G. Gorton and R. Rosen, "Banks and Derivatives," in NBER Macroeconomics Annual 1995, B. S. Bernanke and J. J. Rotemberg, eds. Cambridge, MA: MIT Press.

6 J. Dow and G. Gorton, "Stock Market Efficiency and Economic Efficiency: Is There a Connection?" NBER Working Paper No. 5233, August 1995.

7 See also F. Allen and D. Gale, "Financial Markets, Intermediaries, and Intertemporal Smoothing," Working Paper #5-95, The Wharton School, University of Pennsylvania.

8 In a series of papers Hoshi, Kashyap, and Scharfstein have investigated the role of banks in Japan. They find that the main banks of keiretsus interact with member firms in ways that the market apparently cannot reproduce. See T. Hosbi, A. Kashyap, and D. Scharfstein, "Corporate Structure, Liquidity, and Investment: Evidence from Japanese Industrial Groups, "Quarterly Journal of Economics 106 (1991), pp. 33-60; "The Role of Banks in Reducing the Costs of Financial Distress in Japan, "Journal of Financial Economics 27 (1990), pp. 67-88; and "Bank Monitoring and Investment: Evidence from the Changing Structure of Japanese Corporate Banking Relationships," in Asymmetric Information, Corporate Finance, and Investment, R. G. Hubbard, ed. Chicago: University of Chicago Press, 1990.

9 The idea of banks as monitors is attributable to D. Diamond, "Financial Intermediation and Delegated Monitoring," Review of Economic Studies 51 (1984), pp. 393-414.

10 Also see R. Rajan, "Insiders and Outsiders: The Choice Between Informed and Arm's-Length Debt, "Journal of Finance 47 (1992), pp. 1367-1400; S. Sharpe, "Asymmetric Information, Bank Lending, and Implicit Contracts: A Stylized Model of Customer Relationships, "Journal of Finance 45 (1990), pp. 1069-1087.

11 G. Gorton and G. Pennacchi, "Financial Intermediaries and Liquidity Creation, "Journal of Finance 45 (1990), pp. 49-72.

12 Other notions of liquidity production by banks are given by B. Holmstrom and J. Tirole, "Private and Public Supply of Liquidity," MIT mimeo, 1995; and D. Diamond and P. H. Dybvig, "Bank Runs, Deposit Insurance, and Liquidity," Journal of Political Economy 91 (1983), pp. 401-419.

13 G. Gorton, "Banking Panics and Business Cycles, "Oxford Economic Papers 40 (1988), pp. 751-781. Also see G. Gorton and C. Calomiris, "The Origins of Banking Panics: Models, Facts, and Bank Regulation," in Financial Markets and Financial Crises, R. G. Hubbard, ed. Chicago: University of Chicago Press, 1991.

14 See G. Gorton, "Clearinghouses and the Origin of Central Banking in the United States, "Journal of Economic History 45 (1985), pp. 277-283; and G. Gorton and D. Mullineaux, "The Joint Production of Confidence: Endogenous Regulation and Nineteenth-Century Commercial Bank Clearinghouses, "Journal of Money, Credit and Banking 19 (1987), pp. 458-468. See also G. Gorton, "Bank Suspension of Convertibility, "Journal of Monetary Economics 15 (1985), pp. 177-193.

15 Banknotes may have survived in some form but faced a prohibitively high tax when the federal government began printing money under the National Banking Acts passed during the Civil War. In any case, the trend toward the use of demand deposits in place of notes was already clear.

16 See G. Gorton, "Self-Regulating Bank Coalitions," The Wharton School mimeo, 1989; and F. Allen and D. Gale, "Optimal Financial Crises," The Wharton School mimeo, 1996.

17 See M. Keeley, "Deposit Insurance, Risk, and Market Power in Banking," American Economic Review 80 (1990), pp. 1183-1200; and A. Marcus, "Deregulation and Bank Financial Policy," Journal of Banking and Finance 8 (1990), pp. 557-565.

18 G. Gorton and R. Rosen, "Corporate Control, Portfolio Choice, and the Decline of Banking, "Journal of Finance 50 (1995), pp. 1377-1420.

19 If a bank's (market-value) capital ratio is sufficiently low, then both managers and outside shareholders may agree that the bank should maximize the value of deposit insurance. Rosen and I do not dispute this argument. Rather we focus on the prior question of how the bank came to have a low capital ratio.

20 G. Gorton and A. Winton, "Bank Capital Regulation in General Equilibrium," The Wharton School mimeo, 1996.

21 See G. Gorton and G. Pennacchi, "Are Loan Sales Really Off-Balance Sheet?" Journal of Accounting, Auditing and Finance 4, 2 (1989), pp. 125-145; G. Gorton and J. Haubrich, "The Loan Sales Market," in Research in Financial Services, Volume 2, G. Kaufman, ed. Greenwich, CT: JAI Press, 1990.

22 See G. Gorton and G. Pennacchi, "Banks and Loan Sales: Marketing Nonmarketable Assets, "Journal of Monetary Economics 35 (1995), pp. 389-411.

23 See G. Gorton and R. Rosen, "Banks and Derivatives," in NBER Macroeconomics Annual 1995, op. cit.

Gary Gorton is a research associate in the NBER's Program in Corporate Finance and a professor at The Wharton School. His "Profile" appears later in this issue.