Break Bank of America up: Keep only the Retail operation!

Three Questions For The Financial Stability Oversight Council

Posted: 03 Sep 2011 04:16 AM PDT

By Simon Johnson

The Dodd-Frank financial reform legislation of 2010 created a Financial Stability Oversight Council (FSOC), with the task of taking an integrated view of risks in and around the U.S. financial sector. The FSOC is comprised of all leading regulators and other responsible officials, chaired by the Treasury Secretary. So far, it has done little – fitting with the predominant official view being that in the post-crisis recovery phase, financial risks in the U.S. were generally receding rather than building up.

But this summer has established three important and related issues on which FSOC needs rule quickly. These are: impending bank mergers that could create two more “too big to fail” banks; whether to force the break-up of Bank of America; and how to rethink capital requirements for large systemically important banks, particularly as the continuing European sovereign debt problems undermine the credibility of the international Basel Committee approach to bank capital.

On the mergers, Capital One plans to buy the online business of ING and PNC is acquiring the US business of Royal Bank of Canada. Both acquisitions would create banks with assets around $300 billion. (Steve Pearlstein had a very good column in last Sunday’s Washington Post on the background.)

In some official minds, Dodd-Frank has made it impossible for “too big to fail” banks to exist – meaning that if any such bank got into trouble, it would be shut down without any significant costs being incurred by taxpayers. Most independent analysts and many people active in financial markets regard this proposition as unproven at best and, most likely, simply incorrect.

For example, in a new NBER working paper, “Too-Systemic-To-Fail: What Option Markets Imply About Sector-Wide Government Guarantees”, Bryan T. Kelly, Hanno Lustig, and Stijn Van Nieuwerburgh compare the price of put options (i.e., the option to sell and therefore lock-in a price) for the financial sector stock index relative to put options on individual banks’ stocks. Put options are cheaper if they are less valuable to investors as protection against price collapses, and the index puts are a lot cheaper than the appropriately weighted sum of put options on individual bank stocks, particularly during the recent financial crisis.

The authors infer that “investors price in substantial government bailout guarantees for the financial sector as a whole” – thus the index puts are cheap, because you don’t need to privately insure against overall collapse – with around half of the market value of the financial sector accounted for by collective bailout guarantees during 2003-09. No other sector in the US economy gets anything like this kind of insurance.

At the same time, the researchers point out that the government does not eliminate all idiosyncratic firm-specific risk – this is why put options on individual firms’ stocks are relatively more valuable. In a sense this message is encouraging, because it suggests some specific firms can fail or otherwise go out of business.

But presumably at critical moments specific megabanks have a particular and complete kind of downside protection – it’s hard to envisage the potential failure of a $2 trillion bank like Citigroup or JPM Morgan Chase or Bank of America without this producing system-wide adverse consequences.

The first question for the FSOC is therefore: Wouldn’t allowing the merger Capital One-ING and PNC-RBC mergers create financial firms that are more likely to be systemic?

The largest financial institution allowed to fail without a bailout since the collapse of Lehman Brothers was CIT Group, which had a balance sheet of around $80 billion. Perhaps Capital One and PNC are already too big to fail; PNC is number 12 and Capital One is number 13 on the official list of bank holding companies, ranked by assets as of June 30, 2011. We don’t know where the critical cut-off is – and perhaps more studies along the lines of Kelly, Lustig and Nieuwerburgh would be helpful or the FSOC could find another way to make a reasonable and fact-based determination.

But what really matters is what could happen in future systemic crises – and this is very hard to predict. So why not err on the side of caution and keep large banks from becoming bigger through merging? Or the FSOC could require these merging banks to demonstrate they will generate social value commensurate with or in excess of the extra social risks that they are creating.

The second question is closely related: Why not break up Bank of America? The Dodd-Frank legislative process ended up rejecting the idea that existing banks, as of 2010, should be broken up – as long as they continue to operate in a reasonable and sustainable fashion. But the legislative intent was also clear with regard to big banks that are in trouble – there should be preemptive action, either through pressing bank management or, if that doesn’t work, through regulator-imposed requirements.

These requirements can include making the bank smaller, simpler, and less systemic – in other words making sure that any kind of future “resolution” or “intervention” for that bank (both euphemisms for a form of bankruptcy) would not be a systemic event.

If Bank of America were to fail today, that would create a systemic problem and presumably trigger some sort of desperate policy reaction. B of A, as it is known, is the largest bank holding company in the US today – with assets at the end of June over $2.26 trillion.

If Bank of America is forced to divest various activities, such as those it bought from Merrill Lynch, that would not eliminate systemic risk. But it would make this one troubled institution less central to the economy – and, if handled properly, less of a brake on economic recovery.

This raises the third and arguably most important question: Why not increase capital requirements further for so-called Systemically Important Financial Institutions (SIFIs)?

Warren Buffett’s agreement last week to invest in Bank of America has highlighted the lack of capital, at least for that one SIFI – it has too little equity funding relative to its debts, hence the need for Mr. Buffett’s involvement.

But Mr. Buffett, it now appears, is getting cumulative preferred stock – so he gets paid a guaranteed dividend before any common stock holders get a return. This makes sense for him, without question. And his holding is loss-absorbing, in the sense that his equity would be wiped out before there was any question of defaulting on money owed to any creditors.

This is presumably the best that Bank of America could do in terms of raising new capital through the market. But it should not be enough from the perspective of FSOC, which is charged with overall responsibility for systemic risks.

The Basel Committee on Banking Supervision has proposed a methodology for SIFI capital but it really rests on a very weak analytical basis – as Americans for Financial Reform pointed out in a recent letter. The FSOC would be making a very bad mistake if it continues to follow the European lead which set the lowest common denominator at Basel.

The evident capital problems of European banks – and the way this will slow growth – were flagged at Jackson Hole last week by Christine Lagarde, the new Managing Director of the IMF and, until recently, French Finance Minister (see this assessment by Felix Salmon). The FSOC should listen to her warnings and think about what this means for US banks.

If the Dodd-Frank legislation is to have lasting impact, the FSOC needs to establish itself as a meaningful overseer of systemic financial risks. It needs to meet and deliberate in an open and transparent manner. It should confront pressing questions of systemic risk head on, being clear about the analytical basis for its decisions. Business as usual is a recipe for disaster – in the US, as in Europe.

An edited version of this post appeared on the NYT.com’s Economix blog; it is used here with permission. If you would like to reproduce the entire post, please contact the New York Times.