The case for narrow banking?

John Quiggin makes the case for narrow banking in the pages of The National Interest:

Attempts to regulate the market for derivatives have been stymied by a mixture of determined resistance from the industry and the technical difficulties of defining and regulating such complex and opaque financial instruments. The “shadow-banking” system, associated with investment banks, hedge funds and other speculative financial institutions, is as large and dangerous as ever …

The problem with the shadow banks is not that they are “too big to fail” but that they are too interconnected to fail. The failure of an investment bank, no matter how large, is not a problem if it does not imperil the core functions of the financial system—taking deposits and lending to finance housing and business investment.

The answer is to separate these functions as completely as possible. Banks that benefit from publicly guaranteed deposit insurance should be excluded from any form of financial activity beyond the core functions of saving and lending.

Guaranteed banks should be precluded from operating as part of the shadow-banking system, not just through direct speculation but also from sharing ownership through holding companies and from exposing themselves to risk through loans or other forms of credit to shadow banks. Such institutions should be required to raise all their funds (not merely their equity) from high-wealth private investors capable of assessing the associated risks and bearing whatever costs result from failures.

On the other side of the fence, governments should give a binding guarantee not to rescue investment banks and hedge funds that get into trouble. Everyone involved in these institutions should be fully liable for losses, just as they get the full benefit of profits.

This is not a new idea. Economists as diverse as Milton Friedman and James Tobin advocated it under the name “narrow banking.”

I think the credibility of any ‘non-bailout guarantee’ is the central issue here. As long as such guarantees are not seen as politically credible there may be an unacceptable element of moral hazard present in the system and as long as this is so and can lead to crises such as the one we’ve just experienced, governments may find it difficult to commit to not bailing out the system which sets up the next round of moral hazard induced crises and so on …

While these function separation proposals may seem heretical to supporters of a market based approach, as a matter of pragmatism if non-bailout guarantees cannot be credibly committed to then they may be the least worst alternative.