A new microeconomics of banking and money market
These new macroeconomic developments were founded on a new microeconomics of banking which has been developed since 1970. The latter has allowed the analysis of the rationale of banks and bank risks to be reconsidered.
As regards the consequences of informational asymmetries on the default risk, that is, the probability of default by borrowers, quoting Adam Smith, Stiglitz and Weiss (1981) show that banks must ration credit. Because they have an insufficient amount of information concerning their borrowers, banks are victims of adverse selection and moral hazard. That is why they are not interested in raising their credit rate too high when faced with an excess of credit demand, because this could result in a decrease in the rate of return on their loans, due to the resulting excess of risk taken by borrowers in the case of a higher interest rate. At fairly high levels of interest rates, low-risk borrowers may be inclined to stop borrowing and, on the contrary, only high-risk borrowers (who have a greater potential return on their project) may continue to borrow. Therefore, banks will limit the total loan amount to a level corresponding to the interest rate that they consider to be the maximum that can be paid by low-risk borrowers, and will not raise the interest rate, when there is an excess demand of credit.Douglas Diamond (1984) stressed that it is optimal for final lenders to delegate the monitoring of borrowers to banks once they have deposited their funds in the bank and a loan has been granted to these borrowers by the bank itself. In this model, banks are now creating new kinds of securities, by establishing a credit contract with the borrower and a deposit contract with the individual lender who deposits his funds in the bank. Diamond showed that the total cost from this “delegated monitoring” operated by the bank (that is, the sum of the costs of the bank monitoring borrowers and of individual depositors monitoring the bank) is therefore inferior to the costs of individual lenders monitoring borrowers, when they are themselves in a direct financing relationship.
The bank will benefit from scale economies in monitoring costs. Moreover, as banks hold bigger portfolios than individual lenders, they obtain a better diversification of their portfolio and, as a result, reduce the probability of default for ultimate lenders.Contemporary economic theory also shows that it is possible to reduce informational asymmetries for banks by establishing long-term relationships between banks and their borrowers. This enables banks constantly to collect and accumulate data. The latter may be destroyed when banking crises occur. As a result, when banks are no longer able to lend, or when their informational capital has been destroyed, this will affect macroeconomic variables, because financial markets cannot replace them. This means that banking crises have an effect on the real sector of the economy not only by reducing the quantity of money, but also by destroying all the customer relationship and informational capital that banks have created and accumulated in lending (Bernanke 1983).
The microeconomics of banking also explores liquidity risks. Diamond and Dybvig (1983) showed how banks can provide a liquidity “insurance” for individual lenders that is different from the liquidity provided by financial markets, thanks to their supply of deposit contracts (which guarantees a fixed price) when this insurance cannot be provided through contracts between individual agents, owing to the fact that liquidity needs are “private” information that cannot be publicly verified. Banks will insure their depositors against random shocks that affect their intertemporal consumption preferences, while allowing them to benefit from the returns they obtain from investing their funds. The model shows that a bank is a means of improving agents’ welfare, because it helps them to obtain a better consumption equilibrium than they could achieve on their own.
As long as these shocks are not perfectly correlated, the total quantity of liquid reserves required by a bank that is defined as a coalition of depositors rises less rapidly than the number of its deposits.
As a result, the existence of a banking system with fractional reserves is possible, in which part of the deposits is used to finance illiquid investments.However, this may create fragility for banks, and there is a limit to the provision of liquidity that can be obtained from them, since there may be a “bad” equilibrium, that is to say a bank run equilibrium. The latter is due to coordination failures between depositors. A given depositor’s choice to “run” on a bank if he is not actually facing a liquidity shock is based on his anticipation of other depositors’ behaviour. If he expects a great number of other depositors to withdraw early, he will be tempted to imitate them. Because of the deposit contract rule (“first come, first served”), if he seeks to withdraw his deposit too late, he will not obtain anything. When too many depositors share this feeling and withdraw early, they themselves cause the failure of the bank (even if its fundamental value is positive), which thus cannot repay all of its customers.
This model can also be used to study contagion between banks. When there is a bank failure, this can prompt other banks’ depositors to modify their own expectations about the probability of failure of their own banks. That is why the model also shows the necessity of a deposit insurance. This takes the form of a tax on early withdrawals as a ratio of total withdrawals. As it penalises depositors, it will dissuade those who are not experiencing a liquidity shock from withdrawing early.
Liquidity problems may be solved through the interbank money market, which is used to pool such risks. Banks that have an excess of liquidity and are not experiencing early withdrawals can lend to other banks that are experiencing a run on withdrawals. On principle, when a bank cannot borrow on this market, this signifies that it is insolvent. However, for any bank, an asymmetry of information exists concerning the quality of the other banks’ assets. Doubts about one bank’s solvency may arise, even if it is perfectly sound. Moreover, in time of crisis, any bank may refuse to assist another bank if it lacks confidence in the latter’s ability to assist it in turn should it itself experience such a need in the future (Freixas et al. 2000). As a result, the interbank money market may be the site of these coordination failures. Therefore the lender of last resort, namely the central bank, must intervene by lending to individual banks.
Jerome de Boyer des Roches and Sylvie Diatkine