Tail risk: Higher capital is not enough

Enrico Perotti, Lev Ratnovski, Razvan Vlahu 26 August 2011



The financial part of the global crisis centred on banks and bank-like institutions. Naturally, the regulatory reaction has similarly focused on financial institutions. Much of this debate has rightfully focused on the need for higher bank capital.

  • Bank capital serves as a financial buffer and an incentive mechanism.
  • It forces banks to internalise more losses within the equity tranche, reducing the probability of bank failure, and increases shareholders’ “skin in the game”.
  • Bank capital is essential for both microprudential goals (disciplining shareholders against opportunistically increasing risk) and macroprudential policy (controlling counterparty risk and contagion).

Essential? Yes. Sufficient? No.

It is important to recognise that higher bank capital, while essential, is not a panacea against all risks and agency problems in banks.

The literature has identified important limitations of bank capital. It is now accepted that capital charges based on individual bank risk are ineffective in dealing with:

  • Correlation risks (Acharya 2009),
  • Systemic externalities of unstable funding (Perotti and Suarez 2010), or
  • Lack of shareholders control over bank risk-taking (Dewatripont and Tirole 1993).

In our recent work (Perotti et al 2011), we highlight another important limitation of bank capital. While capital is effective in dealing with “well-distributed” risks of traditional banking, we show that bank capital is potentially ineffective in addressing the tail risk present in modern market-based and wholesale banking and may sometimes be counterproductive.

Tail risk

In traditional banking, where small loans are financed by retail deposits, both asset-side and withdrawal risks’ distributions generally satisfy the law of large numbers, leading to a unimodal, close-to-normal risk distribution for equity value. In contrast, with financial innovation, banks and other financial institutions have the ability to manufacture skewed risk profiles. Investment strategies can generate a high frequency of positive returns, counterbalanced by a significant mass in the extreme left tail of the return distribution. So bankers can claim superior returns at the expense of large infrequent losses. Examples from the recent crisis include excessive reliance on short-term wholesale funding (Gorton 2010), the underwriting of contingent liabilities on tail risk (Acharya and Richardson 2009), and undiversified housing exposures (Shin 2009). A useful review of such strategies is provided by Acharya et al (2009); the IMF (2010) highlights the importance of recognising tail risk in financial stability analysis.

Our work shows that high bank capitalisation may be ineffective in dealing with tail risk because their realisation can wipe out almost any plausible value of initial capital, shifting extreme losses to creditors. Tail risk fully undermines the buffer effect of capital, and weakens its incentive “skin in the game” effects. High initial capital can contain the second moment (variance) in the risk distribution, but it may not discourage tail risk creation at the expense of creditors and, ultimately, financial stability.

To complete the picture, we identify a possible unintended effect of bank capital when financial innovation enables the creation of extremely skewed return profiles. When a bank has a significant capital buffer above the minimum, this may enable the bank to take tail risk. This is because a bank with high capital is less likely to breach the minimal capital ratio in case of smaller shocks. This finding is consistent with the observation that before the crisis, seemingly well-capitalised banks and insurers chose to gamble with contingent liabilities by taking skewed risk strategies (Angora et al 2009; Berger et al 2008). A case in point is represented by UBS, which lost its entire equity buffer on supposedly AAA mortgage-backed securities, which turned out to be subject to tail risk once the real-estate bubble burst.

Policy choices

A general conclusion is that since tail risk emerges in response to risk-shifting incentives, it needs to be discouraged by regulations affecting returns or admissible strategies in good times. In contrast, tail risk cannot be addressed by capital charges based on traditional risk weights. First, tail events are rare and not easily statistically measurable, so risk weights will be imperfect. Second, unlike “well-distributed” risks, tail risk is often generated by correlated investment or funding strategies. Thus economically correct charges for tail risk should depend not just on the risk of individual exposures but on the diffusion of the exposure within the financial system (similar to correlation risk). The existing capital regulation framework is not designed to address such exposures.

It is important to realise that, unlike other low-probability, high-impact events, tail risk cannot be insured within the financial system because tail risk is often realised during systemic events. This leads to serious doubts about the economic usefulness of instruments such as credit default swaps on major sovereigns or large banks. More generally, regulators should realise that financial innovation will be constantly opening new avenues for the generation of tail risk, and focus on uncovering new tail exposures before they become widespread.

Regulators should adopt direct tools for dealing with tail risk, including limits on asset- and liability-side risk exposures. On the asset side, this may include controls of undiversified assets which have extreme correlation with the market in downturns (Acharya et al 2010). On the liability side, this should include prudential limits or surcharges on the use of short-term funding (NSFR or tax-based) and on extremely mismatched strategies (Huang and Ratnovski 2011; Perotti and Suarez 2009), properly measured to reflect not just individual but aggregate risk build-up. Quite useful may be to investigate contingent recapitalisation plans via the issuance of convertible bonds with market-linked triggers, which target risk-shifting incentives (Flannery and Perotti 2011; Haldane 2011).

In conclusion, our research indicates that not recognising limitations of bank capital in controlling financial stability risks may lead to a sense of false comfort. Global financial regulation reform should learn to assess tail risk creation in banking as a necessary step forward after implementing Basel III.

The views expressed in this article are those of the authors and should not be attributed to the Dutch Central Bank or the International Monetary Fund.


Acharya, Viral V (2009), “A Theory of Systemic Risk and Design of Prudential Bank Regulation”, Journal of Financial Stability, 5(3):224-255.

Acharya, Viral V, Thomas Cooley, Matthew Richardson, and Ingo Walter (2009), “Manufacturing Tail Risk: A Perspective on the Financial Crisis of 2007-2009”, Foundations and Trends in Finance, 4(4):247-325.

Acharya, Viral V, Lasse Heje Pedersen, Thomas Philippon, and Matthew Richardson (2010), “Measuring Systemic Risk”, Federal Reserve Bank of Cleveland Working Paper.

Acharya, Viral V and Matthew Richardson (2009), “Restoring Financial Stability: How to Repair a Failed System?”, An Independent View from New York University Stern School of Business, John Wiley & Sons.

Angora, Alain, Isabelle Distinguin, and Clovis Rugemintwari (2009), “Excess Capital of European Banks: Does Bank Heterogeneity Matter?”, Working Paper.

Berger, Allen, Robert DeYoung, Mark Flannery, Ozde Oztekin, and David Lee (2008), “Why Do Large Banking Organizations Hold So Much Capital”, Journal of Financial Services Research, 34:123-150.

Dewatripont, Mathias and Jean Tirole (1993), “Efficient Governance Structure: Implications for Banking Regulation”, in Colin Mayer and Xavier Vives(eds.), Capital Markets and Financial Intermediation, Cambridge University Press.

Flannery, Mark and Enrico Perotti (2011), “CoCo design as a risk preventive tool”, VoxEU.org, 9 February.

Gorton, Gary (2010), Slapped in the Face by the Invisible Hand: Banking and the Panic of 2007, Oxford University Press.

Huang, Rocco and Lev Ratnovski (2011), “The Dark Side of Bank Wholesale Funding”, Journal of Financial Intermediation, 20(2):248-263.

Haldane, Andrew G. (2011), "Capital Discipline", Remarks based on a speech given at the American Economic Association, Denver, Colorado, 9 January 2011.

IMF (2010), “Technical Note on Stress Testing, Publication of Financial Sector Assessment Program Documentation, IMF Country Report No.10/244, July.

Perotti, Enrico, Lev Ratnovski, and Razvan Vlahu (2011), “Capital Regulation and Tail Risk”, IMF Working Paper 11/188.

Perotti, Enrico and Javier Suarez (2009), “Liquidity Risk Charges as a Macro Prudential Tool”, CEPR Policy Insight, 40.

Perotti, Enrico and Javier Suarez (2010), “A Pigovian Approach to Liquidity Regulation”, forthcoming. International Journal of Central Banking

Shin, Hyun Song (2009), “Reflections on Northern Rock: The Bank Run that Heralded the Global Financial Crisis”, Journal of Economic Perspectives, 23(1):101-119.



Topics:  Global crisis International finance

Tags:  tail risk, financial regulation, Macroprudential policies