It is flattering to know that the always insightful (and knowledgeable) David at Deus ex Macchiato found insightful my observation that capital requirements can result in crowded trades–and hence systemic risk. I think the idea is pretty straightforward. It’s all about incentives.
Risk-based capital requirements are like a regime of price controls, in this instance, risk price controls. If some risks are mispriced, and particular, priced too low, all affected institutions face the same incentives to take on those particular risks. The more institutions that fall under the capital regime, the more institutions that will take on these underpriced risks. That’s why I am very leery of global capital regimes, a la Basel. If they screw up the prices–and screw them up they will, with metaphysical certainty–the effect of the perverse incentives will be global.
David suggests the need for regulators to set up capital arbitrage groups to address this problem. The foregoing suggests that these groups should look for the stuff that everybody is doing–those are the risks that are likely to be mispriced.
The problem is that because everybody is doing it, if the regulators try to crack down, the political pushback will be intense. For evidence of this, see this dispiriting paper about how those who went into the subprime craze succeeded in getting Congress to add fuel to the craze, rather than crack down on it. Moral of the story: subsidies (underpricing) tend to create constituencies that favor the continuation of the subsidies. (Cf. the American sugar program, or farm programs generally.) As a result, I think that cap arb groups inside regulators will be about as effective as the war on drugs.
Not everybody is so impressed with my argument. Synthetic Assets, for example:
Craig Pirrong and David at Deus ex Macchiato are worried that the minimum capital requirements set by Basel III will face the same problems as those created by earlier versions of Basel: As banks seek to minimize their capital positions they are all pushed by the regulations into the same trades, this leads to the growth of unregulated financial sectors and crowded trades.
The question I have for proponents of this view is: Why would a bank seek to minimize its capital position, when the purpose of capital is to protect the firm against unpredicted — or even unpredictable — losses? There’s a reason Jamie Dimon has been feted for his “fortress balance sheet” approach to the crisis that everyone — including Chuck Prince — could see coming.
Whenever minimum capital requirements are the determinants of bank behavior, that’s a good indicator of a deep structural problem with the financial system, because it means that you have a financial system populated by banks that are more concerned with maintaining profitability in the short-run than with ensuring that the bank is a viable entity in the long-run. In short, when banks are maintaining only minimal levels of capital, you have pretty clear evidence that repeated bailouts have resulted in the complete perversion of financial system incentives.
SA–with all due respect, you miss the point. Neither I nor David (if I can speak for him) anywhere claim that banks set capital levels at the absolute minimum. It is evident though, that they do economize on costly capital (their shareholders demand it), and make asset allocation decisions strategically in response to the incentives inherent in capital rules. There is no way to explain what transpired in the lead up to the crisis without understanding the crucial role of capital requirements, most notably the extremely favorable capital treatment accorded to AAA rated CDOs, SIVs, etc. You’d have an easier time explaining Hamlet and leaving out a certain prince from your telling.
Yes, banks held more than the minimum in capital. But that was extremely misleading about the true risks in the system, because the capital requirements were fatally distorted. The capital required against certain instruments (government debt being another example) was too small relative to the true risks of those instruments. So too many banks loaded up on them. (This exacerbated the underpricing problem because the capital requirements didn’t take into account how the correlation in trading activity across entities increased the risks of their portfolios.) JPM was the decided exception that proves the dreary rule.
This is a story about relative prices. In a risk based capital regime, some risks are mispriced relative to others. Banks load up on those mispriced risks. Since all face the same distorted pricing signal, they tend to trade the same way. They held more capital than they were required to, but that provided a false sense of security because the required levels of capital did not accurately reflect the risks.
There is, in fact, dysfunction in the financial system. That dysfunction, in the first instance, is the result of the deadly combination of implicit and explicit guarantees that stoke moral hazard, and woefully inadequate and scarily expansive capital requirements that are intended to make it difficult for banks to exploit that moral hazard, but fail to do so.
That’s why I find it chilling that regulators around the world are touting capital requirements as the panacea that will prevent the next crisis, when in fact universal or near-universal requirements are more likely to be the cause of the next crisis. It is inevitable that any requirements will misprice some risks. Banks, motivated in part by the explicit or implicit safety net that means that they do not internalize risks, will methodically seek out those underpriced risks. With identical, or even highly similar, capital requirements across jurisdictions, their portfolios will be overweighted towards those underpriced risks. Everything will look just fine because the banks will say–“look, we have so much more capital than the minimum, so no need to worry.” But the minimum is way too small. And when those overweighted investments get hit by an adverse shock, it is a correlated shock, meaning that there is a substantial risk that the death spiral will begin, as everybody tries to get out of crowded trades.
And that’s why I think Basel–I, II, III. . . or IV, or XXIV–has been faulty, and will be faulty. The fault is inherent in any global system to set risk prices. The knowledge problem will kill you, every damn time.