The U.S. financial system of the mid-20th-century had positive qualities that we should seek to restore, but the separation of commercial banks from securities dealers isn't one of them. Advocates of bringing back the so-called Glass-Steagall act of 1933 are chasing a red herring. The old system's biggest virtue was that it fractured the commercial banking system. This ensured that few lenders were capable of making the same mistakes. Individual banks were slightly more vulnerable to collapse, but these idiosyncratic failures were much less likely to affect the rest of the economy.
Soon after Glass-Steagall was passed, much of the U.S. financial system was a patchwork of small lenders that were limited on where they could operate, whom they could borrow from and whom they could lend to. There were savings and loan associations, which took retail deposits and held mortgages. There were farm banks. There were banks that specialized in lending to small business and banks that financed construction projects. Most banks could only operate within a single state.
This system was criticized almost from the beginning on the grounds that it prevented banks from diversifying and made individual firms much riskier. That's true, but these criticisms ignored the virtue of a highly fractured banking system: The safety of the whole was improved because it consisted of many different entities. To borrow a concept from biology, there was no "monoculture." The tradeoff was that each entity was less diversified than it could have been.
Policymakers paid more attention to the costs of a fractured system than to the benefits. As a result, the rules limiting bank mergers started eroding as early as the 1950s. By the early 1970s, there was a growing consensus that the distinctions separating many different types of commercial lenders were outdated and ought to be abolished. The theory was that this would benefit consumers by increasing competition. Savers could get higher returns even as borrowers would enjoy lower borrowing costs. What could go wrong? (Among other things, the S&L crisis comes to mind.)
Most of the distinctions within commercial banking were gone by the 1980s. Even the separations between commercial and investment banking were eroding thanks to favorable interpretations of ambiguously worded rules by the Federal Reserve. There wasn't much left of Glass-Steagall by the time it was repealed in 1999.
Some who advocate restoring Glass-Steagall appreciate this history. My colleague Simon Johnson and University of Chicago economist Luigi Zingales have both argued in favor of segregating commercial and investment banking on the grounds that it would force the largest institutions to shrink. It would also divide the banking system into two distinct categories with different business lines.
While this doesn't justify the recent push for a "21st Century Glass-Steagall", I'm sympathetic to this reasoning. Restoring even a little bit of the diversity that preceded the rise of big universal banks would be a good thing. I also agree with Zingales that fracturing the banking industry into competing segments makes it somewhat harder for any single group to exert undue political influence.
Fortunately, the U.S. financial system is already becoming somewhat more diverse as a result of the crisis. Insurers, private equity firms and other asset managers have started to step into the void created by the retreating banks by lending to medium-size business starved of funding. I know of at least one university endowment that provides trade financing to Latin America -- a lucrative business that many European lenders have exited. Large U.S. companies have been able to raise money by going directly to the capital markets for decades. The relative diversity of the U.S. system helps explain the economy's superior performance compared with Europe, which is more reliant on a small group of large, sickly institutions.
All of this makes me think that a new version of Glass-Steagall would only produce relatively small gains, at best. Would that be worth the enormous effort when compared with other reform options being discussed? (This assumes that the effort would pay off, instead of merely distracting attention from alternative ideas.) In my judgment, the answer is no. As my colleague James Greiff and I have noted, old-fashioned "boring" banks are quite capable of wrecking the economy if they are undercapitalized. Reducing the amount of debt and increasing the amount of equity in the financial system is the single most useful thing we can do to make it safer. Let's concentrate on that first.
(Matthew C. Klein is a writer for Bloomberg View. Follow him on Twitter.)