Not Everything Can Be Too Big to Fail

There's a lot of loose talk about 'systemic' risk.

PETER J. WALLISON

There is a really bad idea circulating in the nation's capital. Of course, that's not surprising -- but when it's endorsed by the Treasury secretary, we'd better pay attention.

This week, Henry Paulson said in an interview with the Washington Post that he wants the Federal Reserve to be able to regulate and ultimately take over any failing financial institution that it considers crucial -- including hedge funds. This echoes a notion that has been endorsed by executives of some industry associations, that to prevent a recurrence of today's financial crisis it will be necessary to impose tough new regulations on financial institutions deemed "systemically significant."

If this approach were to be adopted in the panicked Washington of today, it would substantially change our financial system and guarantee -- rather than prevent -- future crises.

For starters, because the definition of a systemically significant institution is highly dependent on context, it's impossible to identify one in advance. Consider Drexel Burnham Lambert. When it failed in 1990 it was one of the largest securities firms in the United States, not much smaller in relation to the market at the time than Lehman Brothers was when it collapsed earlier this year. Yet Drexel's collapse, which happened when the market was functioning normally, did not put the financial system or the economy at risk.

On the other hand, when Germany's Herstatt Bank failed to meet its international payment obligations in the mid-1970s, that event caused a serious globalized payment-system crisis -- even though Herstatt itself would not have been on anyone's list of major financial institutions at the time.

And then there's Northern Rock in London. When it collapsed recently, the British government was forced to bail it out even though it was not considered significant before its depositors started to run. In fact, in today's fragile markets, almost any financial institution is systemically significant -- if what we mean by the term is that its collapse will stimulate fear about the stability of others.

In short, predicting the true sources of systemic risk in advance of their actual failure is probably impossible. But even if it could be done, should we want to?

The answer is no. An institution designated as systemically significant, or "crucial," would be marked as too big to fail. After all, that's what such a designation means -- that the institution's failure must be avoided because of its potential impact on the economy or financial system. But once we designate a financial institution as too big to fail, and regulate it as such, a lot of unpleasant things follow.

First, we will have created "moral hazard" and impaired market discipline. The markets will understand that a loan to a systemically significant institution will carry less risk than a loan to an institution that does not have this status. Accordingly, systemically significant institutions will have an easier time raising funds than others, will pay lower rates and grow larger than others, and it will be able to take more risks because of the absence of market discipline.

This in a nutshell is the story of Fannie Mae and Freddie Mac. The two companies were implicitly backed by the U.S. government, which in practical terms meant that they would not be allowed to fail. As a result, they were able to borrow as much as they wanted and take risks with those funds that they couldn't have taken unless the markets believed -- correctly as it turned out -- that Uncle Sam would stand behind them.

Second, systemically significant institutions would suddenly have an unfair competitive edge that would warp the market. Why? Because their lower funding costs would make them more profitable than others, and enable them -- as it enabled Fannie and Freddie -- to drive competitors from any markets they enter.

The effect of designating certain companies as systemically significant would cause a consolidation of the industries in which they are located, with the systemically significant companies gobbling up those that couldn't survive, and others seeking mergers simply to claim designation as systemically significant or too big to fail.

Finally, the support voiced in Washington for the idea of regulating systemically significant financial institutions is based on the fundamental misperception that regulation can prevent them from taking the huge risks their protected status would permit. This New Deal notion should be discarded.

Exhibit A is the banking system, now mired in the worst financial crisis since the Great Depression -- even though it has always been the most heavily regulated. Exhibit B is the S&L debacle less than 20 years ago. Thousands of S&Ls and more than 1,500 commercial banks collapsed in another memorable regulatory failure.

It is completely inexplicable -- after the blindingly obvious failure of bank regulation -- that Washington (and European Union) policy makers would now want to spread regulation beyond banking and into other financial institutions, including hedge funds, brokerage houses and others that the government designates as systemically significant. This would give the government the opportunity to pick winners in each financial industry -- ultimately creating Fannies and Freddies everywhere. Among bad ideas, this one stands out.

Mr. Wallison is a senior fellow at the American Enterprise Institute.