Does boosting bank capital requirements limit output or growth?

Paul Krugman has an intemperate polemic against Alan Greenspan, for instance:

It’s terrible on all counts; but the most offensive thing intellectually is the incredible fallacy of claiming that higher capital requirements for banks amount to keeping resources idle…the man once revered as a demigod of finance doesn’t understand basic economics — or, more likely, that he chooses not to understand what he, amazingly, is still being paid to not understand.

There is no mention of the actual literature on this topic.  Here is one summary of some common views:

Hall (1993) presents evidence that from 1990 to 1992 American banks have reduced their loans by approximately $150 billion, and argues that it was largely due to the introduction of the new risk-based capital guidelines. He goes even so far as to say that “To the extent that a “credit crunch” has weakened economic activity since 1990, Basle-induced declines in lending may have been a major cause of this credit crunch.” Hence, it is not an overstatement to say that Basel I did have an impact on bank behavior as it forced them to hold higher capital ratios than it otherwise would have been the case.

It is a common belief, though by no means universally held, that the implementation of Basel meant a slower U.S. recovery from the recession of the early 1990s.  Here is a summary of some of the international evidence, plus there is a general literature survey in the first few pages at that link; see for instance Peek and Rosengren (the term “capital crunch” will help in Google searches).  Perhaps the literature on the early 1990s may not apply today, but Greenspan’s claim is not incoherent a priori.  Furthermore the Greenspan piece links to an FT piece which a) is consistent with his general account, and b) shows various Europeans, not all of whom are bankers, sharing the same worry for today, and c) makes it clear Greenspan is not committing the “Junker fallacy” of confusing paper holdings with real resource destructions.  The negative effects come through a tax on financial intermediation.  Maybe just maybe one could criticize Greenspan for not being clear enough on the mechanism, but he is still right on the comparative statics and certainly not spouting nonsense.

At the theoretical level, papers on Modigliani-Miller deviations, and the interrelation between production and finance, also make Greenspan’s argument acceptable, if not necessarily correct; one can even look to Joe Stiglitz here.  Krugman admits that capital requirements lower bank risk-taking but of course that can lead to less lending and, in many future world-states, lower output.  That may well be a good thing, since it lowers systemic risk and thus helps output in some world states, but it’s wrong to deny the significant possibility of a real opportunity cost.

Also, bank capital requirements are not well understood in terms of a pure Modigliani-Miller debt-equity swap.  For one thing, the capital requirements favor some asset classes over others, and arguably in a way which limits expected growth.  The capital requirements also involve a commitment to particular accounting standards.

On matters of policy, I do in principle favor significant increases in capital requirements for banks.  But should we push to impose those tougher requirements today in such a weak economy?  Perhaps Krugman would be eager but I’m not so sure.  For all his worries about repeating the mistakes of 1937-8, Krugman doesn’t seem to recognize this may well be another step down that path.