By Simon Johnson
The international discussion among government officials regarding bank reform is, at an informal level, going better than you might think. Top people in the “official sector” are increasingly willing to confront the banking lobby and even refute its more egregious claims, particularly the completely erroneous notion that making banks safer – by requiring them to hold more capital – would actually hurt the broader economy and undermine growth.
Unfortunately, the structured intergovernmental process that actually changes the rules around banks – known as Basel III (or “Basel 3”) – was rushed to an unsatisfactory conclusion last weekend. The US and other countries with major financial centers will need to add substantial additional capital requirements at national levels if these new rules are to be at all effective.
The heart of the substantive discussion regards whether tightening “capital requirements” – the buffers against losses that banks are required to hold – will have a negative impact on the economy. The banks insist that requiring them to hold more capital would slow lending and therefore slow the real economy. The global banks’ Institute for International Finance issued a paper in June that insisted on this point, but there is really no substance to their claims.
The most readable counter arguments come from a paper by Sam Hanson, Anil Kashyap, and Jeremy Stein (reviewed here recently). The fact that three leading academics would take on the banks’ arguments is not news, but in recent months it has become clear that – behind the scenes and off-the-record – a growing number of important officials agree broadly with this view. The Bank for International Settlements itself has produced two serious assessments which, if you look at them carefully (and this is not light reading) argue strongly that longer-term growth would benefit from higher capital requirements because we would experience fewer mega-crises, and that the transition to such arrangements would be much smoother than the industry claims.
Running parallel with this nascent intellectual renaissance among officials, there has been pressure on the Basel Committee on Banking Supervision – comprising 27 countries with substantial financial sectors – to produce an agreement on capital requirements ahead of the Seoul G20 summit in November. The rush to reach agreement in this time frame undermined those arguing for a broader rethink – as the consensus among officials only shifts slowly. The result – the framework agreement announced this weekend – increases capital requirements only modestly and phases those increases in only gradually.
The key conceptual breakdown is this: While pushing back against the banks’ claims that higher capital requirements would be harmful, the officials have still allowed the banks to frame the discussion – implicitly conceding that moving to higher requirements quickly could reduce lending and damage growth.
But again the best thinking from independent analysts demonstrates that this view is completely at odds with the reality. David Scharfstein and Jeremy Stein, summarizing experience on this issue, point out that there is a world of difference between requiring banks to reach a certain capital-asset ratio (which they are likely to do by shrinking assets, i.e., making fewer loans among other things) and requiring them to raise specific dollar amounts of capital.
You can say what you want about the bank stress tests implemented in the US in spring 2009 (e.g., that they were not tough enough, in terms of considering potential future losses – and therefore how much capital the banks needed), but in terms forcing banks to raise specific dollar amounts of capital subsequently, this is the approach to follow.
Banks do not like to raise capital; they will generally only do it when forced. The fair way to do this is through tough and transparent stress tests; these should be repeated in the US and on a comparable basis in other financial center countries every year. Agreeing to move in this direction should be a major goal of the G20, but alas it is currently nowhere near being on the agenda.
The Hanson-Kashyap-Stein view, which is completely mainstream financial economics (not any kind of radical or political view) is that banks should be required to hold enough capital at the peak of the cycle so that when they suffer losses (and, 2007-2010, US banks lost about 7 percent of their “risk-weighted” assets, which is the denominator here), they still have enough capital so that the markets do not think they will fail – and therefore there is no need to dump assets in a desperate bid to survive. (It’s the forced asset sales of this nature that turn financial distress at particular institutions into broader asset price declines and that can trigger panics.)
The logic here points towards at least 15 percent Tier 1 capital being required in good times; the most forward looking officials in G20 countries start to mention aiming for closer to 20 percent (Tier 1 is a good headline measure of loss-absorbing capital, i.e., what stands between the bank and insolvency; the discussion sometimes slips into other related measures). This is what would really help make banks much safer (i.e., thus making banks’ stock a much less risky investment and reducing the required rate of return for all involved).
Treasury Secretary Tim Geithner is fond of saying that the appropriate response to the crisis – and the way to prevent any kind of recurrence – is with “capital, capital, capital.” The Dodd-Frank financial reform act did not raise capital requirements – as Treasury insisted on deferring to the Basel III process. Basel III, we learned this weekend, is on course to raise capital requirements – but to a level below what US banks have held on average in recent decades (for Tier 1 capital, Basel III posits 8.5 percent at the end of the day; US banks fluctuate roughly around 10 percent).
The best – and perhaps only remaining chance – is for the US to insist on stronger capital requirements for domestic financial institutions, as permitted or even encouraged in Basel III under the heading of “countercyclical buffer”. And systemically important financial institutions, for which the Basel process appears to have completely dropped the ball, should be subject to even higher requirements. This should be coordinated with the UK (where there is already thinking in the right direction), Switzerland (again, forward thinking officials hold sway), and anyone else who can be brought on board.
An edited version of this post appeared this morning 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.