By Ricardo J. Caballero

Ben Bernanke, US Federal Reserve chairman, and Tim Geithner, US Treasury Secretary, have reminded us recently that it is time to start thinking about the financial system of the future. This is important not just for the future itself, but also because doing so helps to narrow the types of policies that are desirable during the current crisis. The measures taken today should not hamper, and ideally should facilitate, the transition into the financial system of the future.

There is a natural unifying principle that offers an opportunity for a smooth transition: The financial system of the future should be built on public-private partnerships where, for a premium, the government explicitly assumes most of the tail aggregate risk, while the private sector provides the capital necessary to deal with microeconomic risk and small aggregate shocks.

This new principle differs from the conventional wisdom, which instead points toward boosting capital requirements for financial institutions. The problem with the standard recipe is that it does not distinguish between micro and macro risk. The core of business for financial institutions is the management and redistribution of microeconomic risk. This activity however requires much less capital than does managing aggregate risk. Therefore, basing capital requirement on the latter activity’s demands would be enormously wasteful.

Weighting capital requirements by the amount of macro-risk banks take is reasonable, but it does not substitute for the public guarantee. It is still the case that banks would be over-capitalised with respect to their main activities, and possibly undercapitalised with respect to extreme aggregate shocks, such as the current one. In fact, insurance and macro-based-weights are complementary measures. With the right macro-risk weighting-function, the capital requirement effect of keeping substantial amounts of uninsured macro-risk on its balance sheet should be nearly prohibitive for a leveraged financial institution.

Of course a system that is based on insurance would require strong supervision, but there is consensus already that such supervision is needed regardless of the system we adopt. Moreover, in practice the government ends up operating as an ex-post insurance company against crises anyway, why not make the arrangement explicit and charge a premium for it?

Put simply, the deepest pocket in the economy (the government) should be the insurer of last resort, and behave as would any insurance company by accurately assessing its clients’ risks. Having acquired intimate knowledge of these risks, it can also assess the potential systemic impact of these private risks, and impose minimum-insurance requirements to internalise these systemic risks. That is, the institutions deemed “too big” or “too complex” to fail would be required to have more aggregate insurance than do the non-systemically important institutions.

As it turns out, this insurance system could also play a central role in the solution to the current crisis. The most devastating mechanism at work at this time is the reinforcing feedback between bad news in the real economy and bad news in the financial system. The aggregate-insurance arrangement severs this feedback-loop, and hence it establishes the conditions necessary to restore confidence in the financial system even as the real economy goes through the natural -but now shorter- lags of a recessionary episode.

In sharp contrast, the deleveraging process (the equivalent to raising capital requirements) acts as a contractionary device before it can turn into a stabilisation mechanism through enhanced confidence. Aggregate insurance rather than deleveraging is the right remedy at this time if one of the main goals is for a quick recovery in lending.

Ricardo J. Caballero is the head of the department of economics at Massachusetts Institute of Technology

April 22, 2009 6:58pm in Banks, Crisis, Regulation | Comment