A Failure of Criticism (X): New Regulation of Banking

by Richard A. Posner

This is my last entry subbing on Andrew Sullivan's blog, but I will continue blogging about the economic situation on my Atlantic Correspondents blog, http://correspondents.theatlantic.com/richard_posner/, next week.

The movement to alter the regulation of banks, and the financial system more generally, is gathering steam. I think it is premature. Congress rarely does anything in haste without screwing up, and as I have emphasized in my previous blog entries, regulatory reform instituted before the end of a depression is likely to retard recovery because it further unsettles the business environment, and uncertainty tends to freeze investment.

One reform measure has passed Congress and is about to be signed by the President: a law to make it more difficult for credit card issuers to squeeze customers who do not pay their credit card debt on time, and in general to "protect" people who take on more credit card debt than they can afford--which means simply people who are at a high risk of defaulting because they are highly leveraged. Although Congress rejected placing a ceiling on how much interest credit card issuers can charge, anything that makes it more difficult for a creditor to collect a debt has the same effect as an interest ceiling: it increases the creditor's costs, which reduces the amount of credit and therefore of debt. (A ceiling on interest--a usury law--reduces loans to borrowers who have a high risk of defaulting, because the lender is not compensated for the risk by being able to charge an extra-high interest rate. The new credit card law is a form of usury law.)

The law is premature, since it will reduce borrowing and hence economic activity, which is the opposite of what we want in a recession or depression.

The law will have that effect even if the credit card issuer cannot identify the applicants for credit who are most likely to default, and therefore raises the interest rate to all borrowers.

But we should consider the law's effects in the next boom. The gravity of our current bust is due to excessive borrowing and lending during the preceding boom, and anything that increases interest rates, whether directly or indirectly, will reduce borrowing. But a law aimed at just one type of loan may not have a significant effect on borrowing. A person who owns a house can increase his leverage (the ratio of his debt to his equity) by taking out a home equity loan if he owns a house, by pawning items that he owns, by taking out payday loans, by borrowing from relatives and friends, and by paying his bills late.

A law that raised all consumer interest rates (or all interest rates, period) would tend to moderate the boom phase of the business cycle. But it would also increase the gravity of the bust by reducing people's access to credit during an economic downturn, since the law would apply equally whether the economy was healthy or sick. So it's probably not a good idea.

A related idea that is being bandied about is to create a financial products consumer protection agency, which would try to prevent deceptive sales pitches by mortgage brokers and other sellers of debt to the public. Such an agency would be unlikely to affect the amplitude of booms and busts, because of offsetting effects. To the extent that it reduced consumer search costs or even just made consumers feel better protected against sellers of credit, it would increase the demand for credit, while to the extent it burdened the marketing of debt to consumers, it would reduce the supply of credit.

What makes no sense is subsidizing credit, as is done by the deductibility from income tax of mortgage interest, and by the refusal to allow business firms, including banks, to deduct the cost of equity capital. There is no reason to allow homeowners to deduct the interest on their mortgages, but there is reason to allow banks and other firms to deduct their interest expense, because it is a cost that reduces their income. However, since a firm cannot operate without equity capital, the cost of obtaining such capital is also a business expense, and the average return on equity could be used to approximate the cost for the individual firm. Without any deduction for the cost of equity capital, companies have an incentive to borrow capital, and this increases their debt-equity ratio and hence their risk of bankruptcy, since debt is a fixed cost, which does not decrease when revenues decrease.

There is also cause for concern about the effects of multiplying the number of agencies that regulate financial services (well over 100 at present, as a result of state regulation of banking and insurance and the federal regulation layered over it) and with the delays and disorganization that attend the creation of a new agency.

The alternative to regulating the demand for credit in an effort to moderate busts is to regulate the supply. Too little attention has been paid to measures for improving the Federal Reserve's ability to spot bubbles, as a signal that it should increase interest rates. If I am right that the Federal Reserve has been a major culprit in our present economic crisis, the failure to explore ways of improving its exercise of its power over interest rates leaves a big gap in the reform agenda. The focus of the proposals for reform has instead been on regulating banks more stringently than in the recent past. Eliminating the financing of bank regulatory agencies by fees paid by the regulated firms should be considered. That method of financing regulation creates a conflict of interest because financial firms can, often by a minor change in their corporate structure, choose which agency to be regulated by. This gives rise to competition between regulatory agencies for firms to regulate, a competition that is akin to the competition of sellers for buyers and thus tends to reward laxity in regulation, which makes an agency more attractive to the regulated firms.

The general approach to regulating bank safety is to require that banks have reserves of a specified amount (cash that they are not permitted to lend, with the amount being a percentage of the bank's demand deposits) and also that their equity capital be a minimum percentage, usually 5 percent, of their total capital. The riskier the bank's assets (loans and investments), the more reserves and equity capital the bank can be required to have.

But the problem both of valuing, and of estimating the risk of, a bank's assets (how likely it is that they will lose so much value as to imperil the bank's solvency?) is acute. Because the prospect of being bailed out by the government--on the ground either that the bank is "too big to fail" or that it is "too interconnected" with other banks, so that its failure could have a domino effect--creates an incentive to take risks, bank regulation should, in principle, be stricter the larger or the more interconnected the bank is. But in practice there is the danger that regulation will be too strict and economies of scale or interconnection will be lost, or that less efficient nonbank competitors will eat the banks' lunch because the competitors are not regulated, or regulated as strictly.

A deeper problem is that it may not be possible, without profound changes in bank regulation, actually to reduce the riskiness of lending when low interest rates increase the demand for credit. If a bank is not permitted to make risky loans, this will make it hard for it to obtain an adequate spread between the cost of its borrowed capital and its revenue from re-lending it, but it can compensate by increasing its leverage (its debt-equity ratio) because, by assumption that interest rates are low, it can borrow at a low cost. If the bank's leverage is limited, it can make riskier loans and other investments because the regulatory agencies are not in a good position to determine the riskiness of individual transactions. Credit rating agencies cannot be relied upon because they have a conflict of interest, being paid by the companies they rate; but if instead they were compensated by investors or by the government, this would just replace one conflict of interest with another.

The difficulties involved in any ambitious program of re-regulating the banking industry bring me back to my suggestion that the focus of reform be the Federal Reserve's control over interest rates.