It's a pretty strange journey for an eight-decade old law. And the enthusiasm shows little sign of tapering off. This week, McCain, Cantwell, Sen. Elizabeth Warren (D-MA), and Sen. Angus King (I-ME) proposed the 21st Century Glass-Steagall Act, which would revive sections 20 and 32 of the act, which the Gramm-Leach-Bliley Act of 1999 repealed. But is bringing those sections back a good idea? And would anything really change if we did?
What actually happened to Glass-Steagall
First, let's be clear on what exactly what we're talking about when we talk about Glass-Steagall. There are, confusingly, two Glass-Steagall acts. There was the Glass-Steagall Act of 1932, which allowed the Fed to back its notes with government securities, rather than gold, and to loan out its gold. Basically no one talks about that one any more.
Then there was the Banking Act of 1933, also jointly sponsored by Sen. Carter Glass (D-VA) and Rep. Henry Steagall (D-AL). That bill was probably most notable for creating the Federal Deposit Insurance Corporation (FDIC) and the federal deposit insurance system more broadly. It also regulated some forms of speculative activity by investors.
But what people generally mean by "Glass-Steagall" is sections 16, 20, 21, and 32 of the Banking Act of 1933. These are the sections of the law that regulated the relationship between investment banks (like Goldman Sachs or Morgan Stanley) and commercial banks (like Wells Fargo or Bank of America). Here's what each of them did:
Section 16 banned national banks (that is, banks regulated by the Office of the Comptroller of the Currency) from dealing in securities. So, say, Wells Fargo couldn't go buy some cattle futures for its own profit.
Section 20 banned state-chartered or national banks that are Federal Reserve members from being affiliated with companies that are "engaged principally" in securities-dealing.
Section 21 banned companies that primarily deal with securities from accepting deposits. So a standalone investment bank or hedge fund couldn't just start a side-project that offered checking accounts.
Section 32 banned Fed member banks from sharing corporate board members with companies "engaged principally" in securities-dealing.
Sections 16 and 21 are still in effect. Wells Fargo still can't speculate in cattle futures, and Bridgewater still can't offer checking accounts. What the Gramm-Leach-Bliley Act repealed were sections 20 and 32. That enabled holding companies to own both securities firms and commercial banks, and for the two kinds of companies to share directors on their boards.
The classic example of the kind of deal this enabled was the merger of Citicorp, a commercial bank, with the Travelers Group, a financial conglomerate which owned the investment banks Smith Barney and Salomon Brothers. Bank of America's purchase of Merrill Lynch in 2008 also likely wouldn't have been possible before Gramm-Leach-Bliley.
Was getting rid of Sections 20 and 32 a big deal?
Sort of. There are a lot of things that having a joint commercial/investment bank lets you do. The big advantage, says Jonathan Macey, a corporate finance expert who teaches at Yale Law School and School of Management, is marketing. So, for example, a stockbroker in the investment banking division can direct clients toward credit cards offered by the commercial banking divisions.
But there are also major privacy concerns. Suppose you get a mortgage loan from Citibank, where you also keep your checking account. Let's suppose further that you don't have any income coming in, and are several months late on your mortgage. There's nothing stopping a trader at Smith Barney from seeing that information, and using it to short the mortgage-backed security (MBS) into which your mortgage has been packaged. "I'm not sure they're smart enough to do that," Macey says. "But if they were, nothing would be prohibiting them from doing so."
But Macey also explains that the law was rife with loopholes by the time Gramm-Leach-Bliley passed. "By the time we got to Gramm-Leach-Bliley, the Fed under Alan Greenspan, which was vociferously opposed to Glass-Steagall, had eviscerated the statute," Macey says. "They had interpreted these provisions so broadly in favor of expanding bank powers that, by the time Gramm-Leach-Bliley came along, it was like the fire was so already out there, and all you could do is spit on it to extinguish the flame."
A lot of these had to do with section 20's requirement that affiliates of commercial banks not be "principally" in the securities business. The Greenspan-era Fed didn't count securities trading done for clients of the firm, or on certain kinds of securities such as municipal bonds, in determining whether a business was principally involved with securities. Basically, more than 50 percent of a bank's dealings had to involve non-exempted trading on its own accounts for it to count under section 20. "Things had been interpreted such that Goldman Sachs was not in the securities business, just insanely broad," Macey says.
The other major get-around would be to lard up securities businesses with other non-bank businesses before merging them with commercial banks. So let's say Wells Fargo wanted to buy a hedge fund. It could sell the hedge fund all of its software divisions that handle its information technology, such that more than 50 percent of the fund's business was software, not securities trading. Then it could safely acquire the fund without running afoul of section 20.
So just bringing back sections 20 and 32 wouldn't do much. And, as many people have explained before, the idea that Glass-Steagall's partial repeal caused the financial crisis doesn't really stand up to muster. To say it contributed is probably fair; it increased the number of very large financial institutions, which might have necessitated more bailouts than otherwise happened. And it's certainly fair to argue, as Raj Date and our own Mike Konczal do here, that the move was of a piece with a pattern of financial deregulation that contributed to the crisis. But the real action happened before 1999.
But that's not what the bill does
However, the words "engaged principally" don't appear in Warren-McCain's legislative text. Indeed, it includes the following definition of "securities entity" for the purpose of its regulations:
(I) includes any entity engaged in —(aa) the issue, flotation, underwriting, public sale, or distribution of stocks, bonds, debentures, notes, or other securities;(bb) market making;(cc) activities of a broker or dealer, as those terms are defined in section 3(a) of the Securities Exchange Act of 1934;(dd) activities of a futures commission merchant;(ee) activities of an investment adviser or investment company, as those terms are defined in the Investment Advisers Act of 1940 and the Investment Company Act of 1940, respectively; or(ff) hedge fund or private equity investments in the securities of either privately or publicly held companies; and(II) does not include a bank that, pursuant to its authorized trust and fiduciary activities, purchases and sells investments for the account of its customers or provides financial or investment advice to its customers.
That's pretty airtight, and could be difficult for banks to get around. If enforced strictly, it could require major banks to spin off investment (and insurance) divisions, greatly reducing their size and scale.
Maybe that's how to do it, but it's worth remembering why 20 and 32 went away in the first place. The theory was that allowing different kinds of financial activity would allow companies to diversify their risk. If investment banking is tanking one month and commercial banking's doing all right, then the latter can blunt the impact of the former. That reduces risk, rather than exacerbating it. That obviously didn't happen, but that doesn't mean the underlying idea was necessarily wrong.
Macey, for one, thinks we should have repealed parts 16 and 21, rather than 20 and 32. That way individual commercial and investment banks could diversify their activities without buying each other up and becoming huge mega-banks. Combined with a bill capping the size of banks, along the lines of the Brown-Kaufman amendment, a "Glass-Steagall swap" of that sort could result in small-enough-to-fail banks while preserving whatever diversification advantage Gramm-Leach-Bliley was supposed to produce.
So it really matters what problem Warren-McCain is meant to solve. If its intent is to break up the big banks, then there's a question of whether or not we should just cap the size of them, rather than dictating what kind of banking the now-smaller banks can and can't engage in. If the intent is to split up investment and commercial banking, regardless of scale, then it's worth considering whether the privacy and other benefits of that are outweighed by the diversification advantages of merging different kinds of financial businesses.
Those are difficult, technical questions. There are sound arguments to be made that the breaking up the big banks wouldn't do all that much, or that its effects would actually be harmful; Canada, after all, had a very mild crisis despite having a very concentrated banking sector. Many of the companies that proved most fragile in the 2008 financial crisis were not mergers of commercial banking and investment banking: Bear Stearns and Lehman Brothers were pure securities firms, and AIG was an insurance company. .And maybe, as Sens. Sherrod Brown (D-Ohio) and David Vitter (R-La.) have proposed, setting higher capital requirements is better than enforcing size restrictions.
Glass-Steagall, as Warren once told Andrew Ross Sorkin, is a convenient thing to rally around. But it's worth asking if it's the best way to accomplish this particular goal.