In the aftermath of the Global Crisis that started in 2007, bankers' compensation has become a major issue for banks' corporate governance and regulation. The main question is whether large short-term bonuses spurred too much risk-taking that partly caused the Crisis.
For instance, Rajan (2005), who foresaw some of the key developments that eventually led to the Crisis, emphasised the role of short-term compensation. Post-crisis empirical research on the question is somewhat mixed. Fahlenbrach and Stulz (2011) do not find a link between cash bonuses and US banks’ stock price performance in the Crisis of 2007-8. However, DeYoung et al. (2013), for example, find that a US bank CEO's contractual risk-taking incentives in a given year explain the bank's risk-taking in the following year, measured from the bank's daily stock returns. The effect was strongest and most persistent in the largest banks.1
In response to the compensation concerns, the EU has legislated a cap on bankers’ variable pay (henceforth simply ‘bonus’) of 100% relative to fixed salary, or at most 200%, subject to shareholder approval, along with rules regarding their payment deferrals.2 The US Dodd-Frank Act includes an option for bonus clawbacks which may be paralleled with bonus deferrals.3 Recently, the UK has announced plans to extend the potential bonus clawback period for 7-10 years.
Caps or deferrals?
Which of the two types of bonus restrictions – caps or deferrals – is more effective in containing risk-taking in banks?
We have addressed this question by developing a theoretical model which has been calibrated to data on CEO pay of the 78 largest US banks during 2004-2006, reflecting the time just before the financial crisis started (Jokivuolle et al 2015).4
We first modelled the present value of a banker's future cash bonuses and, based on that, her bonus-induced risk-taking incentives. Both depend on the size of the bonuses and the bonus payment frequency. In the model, the bonus size and the bonus frequency can be constrained by a regulatory bonus cap and deferral.
As expected, we verify that bonuses spur risk-taking since they are a series of call options on future profits. Like regular options, bonuses are more valuable, the higher the volatility of the underlying ‘asset’ is (i.e., the bank’s profits). We show that the risk-taking effect is stronger the less bonuses are deferred. A cap on bonuses clearly reduces the risk-taking incentive by limiting the upside potential of the bonus.
To calibrate the model, we took into account the cost of risk-taking, which can stem from various sources. One obvious source is that by taking too much risk the banker runs the risk of poor performance and hence losing her job. Other examples are costs that stem from corporate culture and regulation.
In the calibration we have made a simple assumption that the costs and benefits of risk-taking in any given sample bank have been in balance, optimised by the banker. Introducing bonus restrictions would change this balance by changing the risk-taking incentives but, by assumption, not the costs of risk-taking. Hence, the optimal amount of risk-taking from the banker’s point of view would have to change. We simulate this effect by considering both bonus deferrals and bonus caps.
Pre-Crisis bonus restrictions
More specifically, we ask the question: "How much would US bank CEOs have reduced their risk-taking, had bonus restrictions been implemented prior to the Global Crisis?"
We find that the effect of bonus deferral on the bankers' risk-taking is immaterial unless bonuses are originally paid more frequently than once a year. That might be the case in hedge funds and private equity companies, but not in banks. In contrast, a cap on bonuses can substantially reduce risk-taking. For an average bank, the risk-reduction effect would translate into reducing the bank’s leverage from 25 to 20 if we assume that the original leverage were 25 and that the entire risk adjustment were done by reducing leverage. Incorporating other forms of variable compensation, such as stock and option grants, although important in size, would not change our results qualitatively.
Intuition suggests that bonus deferrals might be more effective if banks follow a ‘collecting nickels in front of a steamroller' style of investment strategy; i.e. enjoying relatively steady spread returns which are, however, subject to a rare risk of a catastrophic loss. Investments in the securitised credits prior to the crisis were examples of this type of negative tail risks. We find that when re-doing our analysis by assuming that banks’ risk profiles have followed such ‘jump processes’, bonus deferrals indeed start to bite. Thus, according to our analysis, bonus deferrals reduce the large and rare risks, but not the regular volatility of profits. Nonetheless, bonus deferrals are clearly still not as effective in reducing risk-taking incentives as bonus caps are.
The bank-specific effect of the bonus cap varies widely and we find some evidence that it is larger in bigger banks. It is important to note that our analysis is mute on how much risk-taking is desirable and what should be considered excessive. Our aim has been simply to assess the effectiveness of the various bonus restrictions in limiting risk-taking.
On balance, our results suggest that the EU's bonus cap is effective in reining in risk-taking whereas bonus clawbacks included in the US Dodd-Frank Act, or as planned by UK regulators, is less effective.
Balachandran, S, B Kogut, and H Harnal (2010), “The probability of default, excessive risk, and executive compensation: A study of financial services firms from 1995 to 2008”, Columbia Business School Research Paper.
Bhagat, S and B J Bolton (2013), “Misaligned bank executive incentive compensation”.
Bhattacharya, S and A K Purnanandam (2011), “Risk-taking by banks: What did we know and when did we know it?”, AFA 2012 Chicago Meetings Paper.
DeYoung, R, E Y Peng and M Yan (2013), “Executive compensation and business policy choices at US commercial banks”, Journal of Financial and Quantitative Analysis, 48(1): 165-196.
Fahlenbrach, R and R M Stulz (2011), “Bank CEO incentives and the credit crisis”, Journal of Financial Economies, 99: 11-26.
Jokivuolle, E, J Keppo and X Yuan (2015), “Bonus caps, deferrals, and bankers’ risk-taking”, Bank of Finland Discussion Paper, 5/2015.
Rajan, R G (2005), “Has financial development made the world riskier?”, NBER Working Paper, No. 11728.
1 See also Balachandran et al (2010), Bhagat and Bolton (2013), and Bhattacharya and Purnanandam (2011).
2 The EU rules regarding deferrals could be approximately summarised as implying that, instead of the standard one-year horizon, bonuses should be paid over a two-year performance horizon.
3 The primary motivation of clawbacks within the Dodd-Frank Act seem to be to discourage malpractice, not directly to limit risk-taking.
4 The sample follows the one in Fahlenbrach and Stulz (2011) but is somewhat smaller because we include only banks with nonzero average cash bonus over 2004-2006, and because some banks in the original sample were missing CEO tenure information or data points that we use to estimate earnings volatility as a measure of risk-taking.