I don't fully agree with Robert Reich's view expressed below and elsewhere that breaking up the big banks is the key to solving the too big too fail problem. It does mean that the failure of an individual firm would be less worrisome, and hence less likely to get help from the government, but that assumes shocks are idiosyncratic rather than common. However, the shocks we should worry about are common -- systemic to use another term. The shocks that are capable of bringing down large institutions would, if the large banks had been broken up into smaller pieces, bring down all the smaller banks just as easily. Collectively they would still be too big too fail and still require a bailout. And it's certainly true that there are historical episodes where the failure of a large number of small banks was the problem.
But there is another reason to worry about big banks, the economic and political power that comes with size. Even if it's true that breaking banks into smaller pieces doesn't help much, if at all, to make the system safer, all else equal, it does make it much less likely that banks will capture regulators and legislators. And it makes it harder for banks to use their economic power in the marketplace to increase profits at the expense of consumers, or to maintain a financial marketplace where high risk, high reward, taxpayers pay the bill it things go sour strategies are allowed.
So yes, break up the big banks if for no other reason than to curtail economic and political power. And if I'm wrong and this also makes the system substantially safer, so much the better. But what I was really interested in here is the distinction Reich makes between the regulatory and structural approaches, a distinction I think is important:
Obama’s Regulatory Brain, by Robert Reich: The most important thing to know about the 1,500 page financial reform bill passed by the Senate last week — now on he way to being reconciled with the House bill — is that it’s regulatory. If does nothing to change the structure of Wall Street. The bill omits two critical ideas for changing the structure of Wall Street’s biggest banks so they won’t cause more trouble in the future, and leaves a third idea in limbo. The White House doesn’t support any of them.
First, although the Senate bill seeks to avoid the “too big to fail” problem by pushing failing banks into an “orderly” bankruptcy-type process, this regulatory approach isn’t enough. The Senate roundly rejected an amendment that would have broken up the biggest banks... You do not have to be an algorithm-wielding Wall Street whizz-kid to understand that the best way to prevent a bank from becoming too big to fail is preventing it from becoming too big in the first place. ... Because traders and investors know they are too big to fail, these banks have a huge competitive advantage over smaller banks.
Another crucial provision left out of the Senate bill would be to change the structure of banking by resurrecting the Depression-era Glass-Steagall Act and force banks to separate commercial banking ... from investment banking. Here, too, the bill takes a regulatory approach..., it would not erode the giant banks’ monopoly over derivatives trading, adding to their power and inevitable “too big to fail” status.
Which brings us to the third structural idea, advanced by Senator Blanche Lincoln. She would force the banks to do their derivative trades in entities separate from their commercial banking. This measure is still in the bill, but is on life-support after Paul Volcker, Tim Geithner, and Fed chair Ben Bernanke came out against it. Republicans hate it. The biggest banks detest it. ... Almost no one in Washington believes it will survive the upcoming conference committee. But it’s critical. ...
Wall Street’s lobbyists have fought tooth and nail against these three ideas because all would change the structure of America’s biggest banks. The lobbyists won on the first two, and the Street has signaled its willingness to accept the Dodd bill, without Lincoln’s measure. The interesting question is why the president, who says he wants to get “tough” on banks, has also turned his back on changing the structure of American banks — opting for a regulatory approach instead.
It’s almost exactly like health care reform. Ideas for changing the structure of the health-care industry — a single payer, Medicare for all, even a so-called “public option” — were all jettisoned by the White House in favor of a complex set of regulations that left the old system of private for-profit health insurers in place. The final health care act doesn’t even remove the exemption of private insurers from the nation’s antitrust laws.
Regulations don’t work if the underlying structure of an industry — be it banking or health care — got us into trouble in the first place. ... A regulatory rather than structural approach to deep-seated problems in complex industries like banking and health care is also vulnerable to the inevitable erosion that occurs when industry lobbyists insert themselves into the regulatory process. Tiny loopholes get larger. Delays get longer. Legislative words are warped and distorted to mean what industry wants them to mean. ...
Inevitably, top regulators move into the industry they’re putatively trying to regulate, while top guns in the industry move temporarily into regulatory positions. This revolving door of regulation also serves over time to erode all serious attempt at overseeing an industry.
The only way to have a lasting effect on industries as large and intransigent as banking and health care is to alter their structure. That was the approach taken to finance by Franklin D. Roosevelt in the 1930s, and by Lyndon Johnson to health care (Medicare) in the 1960s.
So why has Obama consistently chosen regulation over restructuring? Because restructuring Wall Street or health care would surely elicit firestorms from these industries. Both are politically powerful, and Obama did not want to take them on directly.
A regulatory approach allows for more bargaining, not only in the legislative process but also, over time, in the rule-making process as legislation is put into effect. It’s always possible to placate an industry with a carefully-chosen loophole or vague legislative language that will allow the industry to continue to go on much as before.
And that’s precisely the problem.
The line between structural and regulatory intervention is a bit vague in some cases, but it's still conceptually useful to categorize the types of intervention in this way. As I've said many times, we should try as hard as we can to make the system safe, but we'll never be able to guarantee that the financial system is immune to sudden collapse. Thus, as we think about the structural and regulatory changes that are needed, we should be sure to make changes that minimize the fallout when another collapse occurs, as it eventually will. Much of the change that is needed is structural in nature, but not all, e.g. I'd categorize leverage limits, which I view as critical to minimizing the fallout when problems occur, as regulatory.
However, as noted above structural change is harder than imposing new regulations. The fact that legislators are shying away from the harder to impose types of change out of fear of losing reelection support from the financial industry points to the political power the industry still has, and to the need for structural change to reduce this political (and economic) power. If we cannot muster the political will to make such changes in light of the most devastating financial collapse since the Great Depression, that does not bode well for the future.