Bank Capital and Financial Stability

The following post comes to us from Anjan Thakor, Professor of Finance at Washington University in Saint Louis.

In the paper, Bank Capital and Financial Stability: An Economic Tradeoff or a Faustian Bargain?, forthcoming in the Annual Review of Financial Economics, I review the literature on the relationship between bank capital and stability. Higher capital contributes positively to financial stability. On this issue, there seems to be little disagreement. There is, however, disagreement in the literature on whether the high leverage in banking serves a socially-useful economic purpose, and whether regulators should permit banks to operate with such high leverage despite its pernicious effect on bank stability, and this disagreement seems at least as strong as that over the causes of the subprime crisis (Lo (2012)). Some of the disagreement over higher capital requirements is between those who emphasize the potential benefits of this in terms of reducing systemic risk and those who believe that sufficiently high capital requirements will generate various costs (e.g., lower lending and liquidity creation and the migration of key financial intermediation services to the unregulated sector).

This disagreement is actually valuable because it raises the important issue of calibration: how high should capital requirements be before these costs exceed the stability benefits? We do not have a strong base of research to answer this question. Our theories are primarily qualitative in their characterizations, so definitive statements about the precise levels of optimal capital requirements are elusive. But more empirical research along the lines of Hanson, Kashyap and Stein (2011) and Kisin and Manela (2013) can yield useful insights on this. Theoretically, some progress may be made by settling the issue of whether capital requirements ought to be designed to protect against systematic tail risks, being cognizant of what we have learned about the potentially endogenous dependence of these risks on the capital structures of systemically important institutions.

I also discuss a variety of possible reasons why banks themselves may oppose higher capital requirements, and that we would do well to not only understand the academic arguments on this issue but also the arguments of bankers and the political economy of capital regulation. A factor of some significance in this may be that bank managers often have compensation that rewards them for ROE, suggesting that regulatory concern with the level of executive compensation may be misplaced. What matters more are the conditioning variables for compensation. Another key factor is that banks have many tax-exempt competitors, so banks are particularly sensitive to their overall effective tax burden, including the effect of capital requirements.

One point of view about capital regulation appears to be that bank capital structures are optimally chosen in equilibrium, so capital requirements that distort leverage choices away from these (private) optima will generate costs that we should try to avoid, or at least balance against the benefits of enhanced stability that come with higher capital. Although these private optima may maximize bank equity value, the distorting effects of government safety nets can create a gap between what is privately optimal for banks and what is optimal for society, so the tradeoff is between the social benefits of higher bank capital and its costs as perceived by banks. The other point of view is that even though observed capital structures may be privately optimal, these may be the private optima of bank managers, and may diverge even from bank value maximization. In this case, evidence on the positive cross-sectional relationship between bank capital on the one hand and lending, liquidity creation and bank value on the other would suggest potential benefits even to the shareholders of individual banks from capital regulation that elevates capital levels in banking.

Given the disagreement in the theoretical literature about the desirability of raising capital requirements to enhance banking stability, and the fact that the empirical evidence, while highlighting the benefits of higher capital in the cross section of banks, does not conclusively settle the issue, we could research this topic for a long time without achieving consensus. In the meantime, policymakers must decide. And their decisions have profound consequences. Higher capital in banking should be thought of as “private deposit insurance” that reduces the contingent liability of the government related to prohibitively expensive future bailouts. These bailouts are necessitated by the correlated failures of highly-leveraged banks making correlated assets choice that endogenously create systemic risk. Higher capital in banking can stanch this systemic risk by altering incentives at the individual bank level, and thereby diminish the threat of a sovereign debt crisis engendered by the need for a dramatic increase in government debt to finance a bailout of the banking industry. These benefits seem large enough to justify the possible loss of bank-level as well as social benefits associated with the replacement (in the aggregate) of some bank debt with equity. Changes in the tax code to reduce the tax disadvantage of equity would lessen the bank-level cost of reducing leverage and facilitate a transition to higher capital levels. Moreover, it would be best to achieve the transition in a phased-in manner, so that banks can build up higher capital levels via lower dividends and higher earnings retentions. This will avoid adverse selection and other costs associated with equity issues.

The full paper is available for download here.

Both comments and trackbacks are currently closed.