Ben Bernanke, chairman of the Federal Reserve, has stayed carefully on the sidelines while a major argument has broken out among and around senior policymaking circles: Should our biggest banks be broken up, or can they be safely re-regulated into permanently good behavior? (See the recent competing answers from WSJ, FT, and the New Republic).
But the issues are too pressing and the stakes are too high for key economic policymakers to remain silent or not have an opinion. On Cape Cod last Friday, Mr. Bernanke appeared to lean towards the banking industry status quo, arguing that regulation would allow us to keep the benefits of large complex financial institutions.
Note, however, that Bernanke’s quote making this point in the NPR story (at the 45 second mark) is from his spoken remarks; the prepared speech does not contain any such language. And Mr. Bernanke is wise to be wary of endorsing the benefits of size in the banking sector – the evidence in this regard is shaky at best.
There are three main types of evidence: findings from academic research on the returns to size in banking; current and likely future policy in other countries; and actual practices in the banking industry.
First, while academic research is not always the primary driver of policy choices, it is relevant when we can readily see the costs of big banks (in the crisis around us) but the supporters of those banks claim they bring important benefits. In fact, the available research indicates that in the banking sector, economies of scale exist only up to a (relatively low) level of total assets, while economies of scope are elusive. The benefits from diversification across countries or lines of business are also small; moreover over the last few months we learned that correlations among different markets and asset classes increase rapidly during a crisis – thus reducing even more the benefits of diversification. [See “Consolidation and efficiency in the financial sector: A review of the international evidence,” by Dean Amel, Colleen Barnes, Fabio Panetta, Carmelo Salleo; Journal of Banking & Finance 28 (2004) 2493–2519. Note that one of the authors works at the Federal Reserve Board, and all four work in a central bank or ministry of finance.]
Second, policy in other countries matters because some fear that breaking up big US banks would somehow put us at a competitive disadvantage vis-à-vis big European or other banks. But on this issue the European Commission spoke loudly this week – ordering the break-up of ING, and the presumption is that they will also soon put similar pressure on big UK banks.
Interpretations of this action vary – some see it as an implementation of competition policy, while others feel the Commission is (rightly) concerned about the unfair subsidies implicit in government ownership and support for large banks. The Commission itself is being somewhat enigmatic, but the exact official motivation doesn’t matter – the important point is that the leading pan-European policy setting organization, which does not rush into decisions, has determined that whatever the benefits of size in banking, the public interest requires smaller banks.
Third, in terms of actual business practice, any big investment banking transaction is done with a syndicate or group of banks – there is sometimes a lead bank with a favored relationship, but that role is definitely shopped around.
Take, for example, General Electric’s October 2008 share offering, in which there were seven lead managers. Or look at the prominent Microsoft bond offering, which had Bank of America, Citi, JPM, Morgan Stanley as lead managers and Credit Suisse, UBS, and Wachovia as “joint lead” (in this context, “joint lead” is the junior partner). If a nonfinancial corporate entity takes out a large bank loan, this is also shopped around and syndicated – even for medium sized companies – so as to divide up the risk.
Similarly, if a company wants to do a foreign exchange transaction, it searches for offers and take the best deal. It would be unwise to rely exclusively on one bank – they will naturally hit hard you in terms of higher fees.
One area where banks benefit from size is in terms of being able to put their balance sheet behind a transaction – e.g., to get a merger done they may offer a bridge loan, with the real goal being to get merger fees. Bigger banks with a large balance sheet have an advantage in this regard. However, this kind of risk taking is also what gets banks in trouble (e.g., in the 1997-98 Asian Financial Crisis). In the past, both Morgan Stanley and Goldman Sachs did not have large balance sheets but still did well in mergers and acquisition.
Goldman is an interesting case because it had $217 billion in assets in 1998 (that’s $270 billion in today’s dollars); it now has around $1 trillion. Goldman was considered a strong global bank in the late 1990s. Can it really be the case that the idea size for banks has risen so dramatically over the past decade? (Lehman had $154 billion assets in 1998 and above $600bn when it failed).
For derivatives (and other instruments) it’s important to have deep markets, but not necessarily big banks. If you want to buy and sell stock you want a liquid market, and the same is true for derivatives.
If you are a large oil company, and you want to hedge future risk, your choices are:
1. Hedge with a “too big to fail” bank, because you know taxpayers will bail you out and these banks are subsidized by their government support, so they can give you a better price.
2. Or you can hedge with several banks to minimize counter party risk. They then sell of some of the risk – taking take less risk themselves as they are small enough to fail.
If you were hedging you’d prefer the “too big to fail” system because it comes with a nifty subsidy. But this is not what the Federal Reserve should be supporting – Mr. Bernanke may still come out in favor of markets-without-subsidies.
By Peter Boone and Simon Johnson
An edited version of this post appeared previously on the NYT.com’s Economix; it is used here with permission. If you would like to reproduce the entire post, please contact the New York Times.