Discussion of "Capital Regulation and Tail Risk"
本文讨论Perotti等人关于尾部风险如何削弱银行资本监管有效性的研究，指出尾部风险会扭曲资本与风险承担的关系，导致资本充足率提高时银行反而可能增加风险。
During the recent financial crisis, the notion of “tail risk”— exposure to very unlikely yet massive losses—rapidly became the foremost concern of regulators, banks, and other market participants alike. Perotti, Ratnovski, and Vlahu (this issue) analyze how the presence of such risks affects the relationship between bank capital and bank risk taking. With policymakers looking to revamp capital regulations so as to prevent a similar crisis from occurring, there is no question that this is a very timely paper indeed. Perotti, Ratnovski, and Vlahu argue that tail risk differs from the “normal” risks modeled in the banking literature because it can wipe out any amount of bank capital. As a result, the presence of tail risk weakens the effectiveness of capital in reducing risk-shifting incentives. Moreover, when capital is costly to raise, the combination of tail risk and less-catastrophic (“non-tail”) risk can complicate the relationship between capital and risk shifting, making it non-monotonic: although an increase in bank capital from low levels tends to deter risk shifting, at even higher levels the bank may start taking risks again, only to revert to a safer strategy if capital is sufficiently high. This possible reversal arises from the fact that regulatory capital requirements apply not only ex ante but also to the bank’s condition after risks have materialized. The reversal becomes stronger as the probability of tail risks increases. The upshot is that capital regulation is even less effective in the presence of tail risk. This paper makes a number of innovations, including not only the distinction between tail and non-tail risk but also the impact of risk realizations on ex post capital adequacy and how, in the presence of recapitalization costs, this in turn feeds back into initial risk choices. Nevertheless, I think there are several aspects in which it could be