At a hearing of the House Financial Services Committee yesterday, Barney Frank nicely summarized where we are with regard to re-regulation of our largest financial institutions: some of them are definitely “too big to fail”, with the potential to present the authorities with what Larry Summers calls the “collapse or bailout” choice, but what exactly should be done about it?
On a five-person panel, I had the middle seat (as usual) and found myself agreeing with points made both to my left and to my right. Alice Rivlin is correct that we need to control leverage as well as increase capital requirements, and the Fed’s tools vis-à-vis leverage need modernization – your grandparents’ margin requirements would not suffice. Peter Wallison, a member of the new financial crisis investigation commission, stresses that capital requirements should be higher for larger banks. Paul Mahoney wants to change the bankruptcy code, to make it easier for courts to handle large financial firms in quick time; recent CIT Group events suggest this is a good idea.
And Mark Zandi was persuasive on the point that households had no idea what they were signing up to with option ARMs – even he has trouble with those spreadsheets. Effective consumer protection – including a new consumer safety commission – would definitely contribute to financial system stability.
What will Barney Frank and his committee do? There will be no “Tier 1 Holding Company” category of firms, if Frank has anything to do with it; this is too much like creating an implicit government guarantee. Frank is clearly drawn towards higher capital requirements or more insurance payments from firms that pose more system risk. I suggested total assets of 1% of GDP as a threshold, but we agree this should be essentially a progressive drag on profits – creating the strong market-based incentive for the biggest firms to downsize.
Other than that, watch this space.
My written testimony submitted to the committee is below.
1) The U.S. economic system has evolved relatively effective ways of handling the insolvency of nonfinancial firms (through bankruptcy) and small or medium-sized financial institutions with retail deposits (through a FDIC-run intervention process). These kinds of corporate failures inflict limited costs on the real economy, and even a string of problems in such firms does not generally jeopardize the entire financial system.
2) We do not yet have a similarly effective way to deal with the insolvency of large financial institutions (e.g., any bank with assets over $500bn, which is roughly 3 percent of GDP). When one of these firms gets into trouble, the authorities face an unpalatable choice of “bailout or collapse.” If the problems spread to more than one firm, the balance of responsible official thinking shifts towards: “bailout, at any cost”.
3) The collapse of a single large bank, insurance company, or other financial intermediary can have serious negative consequences for the U.S. economy. Even worse, it can trigger further bank failures both within the United States and in other countries – and failures elsewhere in the world can quickly create further problems that impact our financial system and those of our major trading partners.
4) As a result, we currently face a high degree of systemic risk, both within the United States and across the global financial system. This risk is high in historical terms for the US, higher than experienced in most countries previously, and probably unprecedented in its global dimensions.
5) Short-term measures taken by the US government since fall 2008 (and particularly under the Obama administration) have helped stabilized financial markets – primarily by providing unprecedented levels of direct and indirect support to large banks. But these same measures have not removed the longer-run causes of systemic instability. In fact, as a result of supporting leading institutions on terms that are generous to top bank executives (few have been fired or faced other adverse consequences), systemic risk has likely been exacerbated.
6) Some of our largest financial firms have actually become bigger relative to the system and stronger politically as a result of the crisis. Executives of the surviving large firms have every reason to believe they are “too big to fail.” They have no incentive to help bring system risk down to acceptable levels.
7) Specifically, the surviving large U.S. financial firms and their foreign competitors have a strong incentive to resume “pay-for-performance” incentive systems – they compete by attracting “talent,” and if any one firm brings its compensation under control, it will lose skilled employees. But these firms – and their regulators – have also demonstrated they cannot prevent such incentives from becoming “pay-for-disguised-risk-taking” on a massive scale.
8) The potential for unacceptable systemic risk remains deeply engrained in the culture and organizational structure of Big Finance. Over the past 30 years, this sector has benefited from a process of “cultural capture,” through which regulators, politicians, and independent analysts became convinced this sector had great and stabilizing technical expertise. This belief system is increasingly disputed, but still remains substantially in place – big banks are, amazingly, still presumed by officials to have the expertise necessary to manage their own risks, to prevent system failure, and to guide public policy.
9) There are four potential ways to reduce system risk going forward
- Change our regulations so as to reduce ex ante risk-taking, e.g., by more effectively controlling the extent of leverage in the financial system or by more tightly regulating derivatives transactions.
- Change the allocation of regulatory authority within the financial system, so that the relative powers of the Federal Reserve, Treasury, FDIC and various other regulators are adjusted.
- Make it easier for the authorities to close down failing large financial companies using a revised “resolution authority.”
- Change the size structure of the financial system, so that there are no financial institutions that are “too big to fail”.
10) All of these approaches have some appeal and it makes sense to proceed on a broad front – because it is hard to know what will gain more traction in practice.
11) The growing complexity of global financial markets means that even sophisticated financial sector executives do not necessarily understand the full nature of the risks they are taking on.
12) There is no ideal – or even proven – regulatory structure that will work inside the U.S. political system. Relative to the alternatives, strengthening the FDIC makes sense. For certain levels of potential bailout (e.g., as with CIT Group recently), the FDIC has an effective veto power over providing some forms of government support. This has proved a helpful check on the discretion of the Federal Reserve and the Treasury recently, but it would be a mistake to assume this will be the case indefinitely.
13) While an extended “resolution authority” could be helpful, it is not a panacea. As markets evolve, new forms of interconnections evolve – and we have learned that not even managers of the best run banks understand how that affects the transmission of shocks. Furthermore, as banks become more global, an effective resolution authority would need to span all major countries in comprehensive detail. We are many years away from such an arrangement.
14) The stakes are very high – the country’s fiscal position has been significantly worsened by the current crisis, and our debt/GDP ratio is on track to roughly double.
15) As a result, it makes sense also to consider measures that will reduce the size of the largest financial institutions. The recent experience of CIT Group suggests that a total asset size under $100bn may provide a rough threshold, at least on an interim basis, below which the government can allow bankruptcy and/or renegotiation with private creditors to proceed.
16) Market-based pressure for size reduction can come through a variety of measures, including higher payments to the FDIC (or equivalent government insurance agency) from institutions that pose greater system risk, higher capital requirements for bigger firms, and differential caps on compensation based on the cost of implied government assistance in the event of a failure – think of this as pre-payment for failure.
17) Breaking up our largest banks is entirely plausible in economic terms. This action would affect less than a dozen entities, could be spread out over a number of years, and would likely increase (rather than reduce) the availability of low-cost financial intermediation services.
18) The political battle to set in place such anti-size measures would be epic. But as in previous financial reform episodes in the United States (e.g., under Teddy Roosevelt at the start of the 20th century or under FDR during the 1930s), over a 3-5 year period even the most powerful financial interests can be brought under control.
19) If we are able to make our largest financial firms smaller, there will still be potential concerns about connected failures or domino effects. Much tougher implementation of “safety and soundness” regulation is the only way to deal with this. In that context, stronger consumer protection – through a new agency focused on the safety of financial products – would definitely help (as well as being a good thing for its own sake).
The remainder of this testimony (see this pdf) provides further background regarding how systemic risk developed to its current high levels in the U.S., and suggests why we need new limits on financial institutions whose management regards them as “too big to fail”.