Notes from the Vault
Larry D. Wall
For many people, the beginning of the year is an opportunity to set new goals. For many in the banking industry, it represents the time when they learn their incentive compensation for their contributions over the previous year. The media often focuses on the magnitude of the bonuses, with industry participants sometimes expressing concern that the bonuses are not large enough and editorial pages many times denouncing them as far too large.1 However, the media frequently do not address an important topic: how individual bankers' compensation is determined.
Academics and regulators often focus more on the question of "How are bonuses determined?" than "How big are bankers' bonuses?" because the process for determining bonuses arguably has a substantial influence over how much risk banks take. Indeed, a survey of large banks by the Institute for International Finance (2009) found that almost all of the 37 banks in its sample agreed that "compensation structures were one of the factors underlying the current (2007–09) crisis." From an academic perspective, one important issue is whether the banks were correct that compensation practices contributed to the crisis, and, if so, which practices. From a regulatory perspective, an important issue is whether the postcrisis adoption of regulatory rules on compensation is likely to make the banking system safer.
This post discusses some of the highlights of my 2019 paper on bank incentive compensation, "Is Stricter Regulation of Incentive Compensation the Missing Piece?" The reference to the "missing piece" is in recognition that other attempts to reduce the risk of a banking crisis have made clear progress, but these attempts have yet to overcome some difficult barriers.2 The paper asks whether incentive compensation regulation is the missing piece needed to reduce the risk of a future crisis.
Postcrisis changes in regulation
Before addressing the question of what more could be done with regulation, consider what the regulators have done since the crisis. Bank compensation was largely unregulated prior to the crisis. U.S. regulators could (and likely did) act in some cases where they observed extraordinarily poor compensation design under their general power to stop unsafe and unsound practices at individual banks. However, the United States did not have regulations focused on compensation practices, nor did the international community agree on standards. Moreover, a Federal Reserve Board of Governors study states that a survey immediately after the crisis found "no firm had a well-developed strategy to use risk adjustments and many had no effective risk adjustments."
Immediately after the crisis, the G-20 leaders issued a statement calling for "compensation practices to support financial stability." The Financial Stability Forum had already taken steps in this direction by issuing the FSF Principles for Sound Compensation Practices in 2009. The forum's successor, the Financial Stability Board, followed this up later in 2009 with FSB Principles for Sound Compensation Practices: Implementation Standards. These documents provided a set of principles for both the process and substance of sound compensation practices. The substantive requirements were based on the principle that incentive compensation should take into consideration not only the earnings generated by the employee but also the full range of risks taken by that employee. Importantly, the standards also extend the requirements to all bank "employees whose actions have a material impact on the risk exposure of the firm." This includes employees who are not part of the senior management group but who generate material risk exposures, such as traders in security units.
The U.S. federal bank regulatory agencies issued guidance on incentive compensation to implement the guidance in the principles and standards in June 2010 (Federal Reserve press release). Congress then addressed compensation policies in Section 956 of the Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010. Section 956 of that act mandates that the federal financial regulatory agencies write a regulation or guidance on incentive compensation practices that encourage excessive risk taking in financial firms more generally. This broader group of federal regulators—which includes several nonbank regulators such as the Commodities Futures Trading Commission, the Federal Housing Finance Agency, and the Securities and Exchange Commission—has not yet reached agreement on the language of a regulation implementing Section 956.
The Financial Stability Board has subsequently made a series of annual progress reports, which finds the principles and standards have been widely adopted by major countries and implemented by their banks, including in the United States. Additionally, some jurisdictions have gone beyond the principles and standards. Most notably, the European Union has imposed significant limits on the proportion of total compensation accounted for by variable compensation.
Why study CEOs and other employees?
We ordinarily think of a bank's board of directors as exercising delegated authority on behalf of the shareholders over a firm's decision-making process. However, the board does so only indirectly via its decision over who to hire as the bank's chief executive officer (CEO) and the compensation contract it negotiates with that individual. Ultimately, the decisions on how much risk to take are made by the CEO and his or her subordinates. Thus, regulatory limits on CEO and employee compensation could, in principle, have a material impact on a bank's risk exposure by changing their incentives to take risks.
Almost all studies of the effect of compensation practices on bank risk taking focus either on the CEO or on lower-level employees based on both theoretical and practical considerations. Theoretically, the CEO has overall responsibility for all of the decisions made in the bank, whereas lower-level employees have influence only in the areas they work in. Thus, studies focusing on CEOs can reasonably use bank-wide measures such as accounting earnings and stock performance to evaluate the CEO's actions. In contrast, more narrowly defined measures are appropriate for lower-level employees.
The practical motivation for the separation is data availability. Not only are more relevant data about CEO performance publicly available, more data about CEO compensation are also publicly available in some jurisdictions, including the United States. The actual employment contracts are rarely available, but the amount paid to CEOs and their stock holdings are publicly disclosed in the United States. The available public data on lower-level staff compensation and performance are typically highly aggregated, limiting what can be done unless the researcher is given confidential access to more disaggregated data.
Evidence on CEO compensation and risk
Theoretical studies of CEO compensation at nonfinancial firms provide a good starting point for discussing bank CEO compensation. That literature finds that if the firm is financed entirely by equity and the CEO is risk neutral, then paying the CEO entirely in stock will align the CEO with that of the shareholders. Allowing the CEO to be risk averse provides a rationale for providing part of the compensation in the form of a fixed salary rather than entirely with stock.
The complications come when the firm is partially financed with debt. In this case, shareholders can benefit ex post from the CEO taking more risk than is optimal if the increased risk is not anticipated by the debt holders. The shareholders get to keep the gains if the firm does well, but the debt holders bear part of the costs if the firm fails. However, debt holders are generally not naïve. They anticipate the possibility that the firm will take more risk and demand a higher-risk premium from firms likely to take more risk. Creditors' ability to demand higher-risk premiums gives shareholders an ex ante incentive to commit to not taking excessive risk.
Bolton, Mehran, and Shapiro (2011) apply this intuition to banks. They suggest a way to correct the problem of excessive risk taking is to give the CEO a compensation contract that depends not only on equity returns but also on creditors' returns; for example, based on the performance of the bank's credit default swaps. If the relative influence of debt and equity on CEO compensation is properly set, the CEO will be incented to take the ex-ante optimal amount of risk and that will maximize shareholder value. The authors then bring in government-supplied deposit insurance under which bank shareholders can gain from taking excessive risk if deposit insurance premiums do not fully reflect the actual risk taken by the bank. The result may be that the contracts offered by shareholders to bank CEOs encourage excessive risk taking.
However, CEO compensation typically includes not only salary and stock compensation but also cash bonuses and pension plans. The pension plans are important because they are generally an unsecured obligation of the bank that has equal priority with other forms of nondeposit liabilities. These pension plans along with deferred cash compensation are called "inside debt" and give CEOs an incentive to avoid taking excessive risk.3
These theoretical studies help guide our understanding of bank CEO compensation, but they necessarily simplify potentially important parts of the problem. Thus, a number of empirical studies have looked at the relationship between compensation practices and risk taking observed in the data. Many of the studies found a statistically significant, positive relationship between the sensitivity of CEO compensation to equity returns and the amount of risk taken by the bank.4 However, those studies that also estimated models with inside debt found that their results seem to be driven by inside debt. That is, the coefficients on inside debt were statistically significant in explaining bank risk taking, but the coefficients on sensitivity to equity returns became statistically insignificant.5
The relationship between compensation and risk taking has also been challenged by Fahlenbrach, Prilmeier, and Stulz (2012), who found that the stock performance of banks in the 1998 financial crisis predicted the performance of banks in the June 2007 to December 2008 period. Their analysis could not determine exactly why this link exists, but they suggest it likely is a result of the cross-sectional difference in bank risk-taking culture and that this culture evolves slowly over time. Another, not mutually exclusive hypothesis is that risk decisions are being driven in part by each bank's core competencies and customer base, both of which also evolve slowly over time. Under these hypotheses, the relationship between CEO compensation and risk taking is not causal, but rather both are being driven by longer-run factors such as the bank's risk culture and customer base.
Overall, my review of the papers on bank CEO compensation has a few policy implications. First, regulation should not focus exclusively on CEO equity-linked compensation. The empirical evidence suggests that greater reliance on inside debt may be more effective in limiting bank risk taking. Second, one should have modest expectations about the effect of regulating CEO compensation on bank risk taking. The causality may not run simply from compensation to risk taking. Instead, it may go from deeper factors such as bank risk-taking culture and business models to observed compensation and risk taking.
However, my survey paper perhaps stops a bit short by failing to address how regulators may facilitate changes in these deeper factors at excessively risky banks. A comprehensive answer on how to change a bank's culture and business model is far beyond the scope of that paper or this post. However, it seems likely that part of the answer will be to create financial incentives for the CEO as the bank's leader to work on building a culture and business model that is consistent with prudent limits to risk taking. Thus, even if one has modest expectations about CEO compensation regulation in the short run, it likely has an important role to play in the longer run.
Evidence on employee compensation and risk
The empirical literature on lower-level bank employees is small, but it finds consistent evidence that compensation practices designed to incent greater effort can also lead to greater risk taking if they are improperly designed.6 The more novel insights come from the theoretical papers studying both the motivation for and design of regulation.
Theoretical arguments for regulating bank employee compensation are provided by two papers. Thanassoulis (2012) observes that investment bankers' compensation levels had reached rather high levels prior to the crisis. He develops a model in which competition for bankers could lead to compensation that exceeds the levels needed to incent optimal effort. Thanassoulis viewed such excessive competition as a "market failure," though not everyone would necessarily agree that more competition leading to higher compensation is a market failure. Bannier, Feess, and Packham (2013) developed a model in which the better bankers are prepared to signal their quality by accepting compensation packages with performance-based bonuses. Competition between banks for the best employees would lead to larger performance-based bonuses that have the unintended consequence of incenting those hired to take more risk than would be optimal for the bank.
The theoretical literature, however, also provides some cautionary notes about regulating bank employee compensation. Thanassoulis (2012) warns against focusing solely on limiting the amount of compensation provided by bonuses. He notes that banks will still compete for the best employees. All that limiting bonuses does is force the banks to pay higher salaries, which cannot be so easily cut when the overall bank is performing poorly. Kupiec (2013) and Jarque and Prescott (2013) focus on specific, seemingly commonsense aspects of the Financial Stability Forum (2019) principles and Financial Stability Board (2019) standards that can have the perverse effect of increasing the riskiness of banks in certain situations.
Overall, the studies of employee risk taking suggest a potentially important contribution was made by bringing material risk takers within the scope of the international principles and standards. However, these studies also issue an important qualification. Poorly designed regulations can aggravate risk-taking incentives rather than reduce them. The message that I take from the Kupiec (2013) and Jarque and Prescott (2013) papers is that in order to create appropriate incentives, it is critical to understand the processes by which risk is created at individual banks as well as how different compensation structures affect those risks. That implies that truly effective regulation of bank employee compensation cannot be done by merely setting up a checklist of best practices, but requires careful thought by both bank compensation committees and bank supervisors.
My earlier review of bank compensation practices ended by noting one important limitation of its analysis. The existing literature on bank compensation focused on banking as it has been traditionally practiced. However, ongoing developments in technology are blurring the lines between traditional banks, nonbank financial firms, and information technology firms. Indeed, some senior bankers are starting to refer to their firms as being more technology firms. If banking really is becoming a technology business, that has many implications for the financial system and bank regulation. My review highlighted two important points relevant to bankers' compensation. First, if banks are competing with tech firms for business, they are also competing with tech firms for employees, which may have implications for compensation structure. Second, the saying "move fast and break things" that sometimes applies to tech firms would be a frightening prospect if applied to systemically important financial firms. The growing overlap of banking and tech is not going to diminish the need for well-designed prudential regulation and supervision of banks.
Larry D. Wall is executive director of the Center for Financial Innovation and Stability at the Atlanta Fed. The author thanks Carl Hudson for helpful comments. The view expressed here are the author's and not necessarily those of the Federal Reserve Bank of Atlanta or the Federal Reserve System. If you wish to comment on this post, please email email@example.com.
Acharya, Viral V., Lubomir P. Litov, and Simone M. Sepe (2013). "Non-Executive Incentives and Bank Risk-Taking," Working Paper, New York University. https://pdfs.semanticscholar.org/e39c/e20e4e90e514937e70a1708944320648deb6.pdf.
Agarwal, Sumit and Itzhak Ben-David (2017). "Loan Prospecting and the Loss of Soft Information." Journal of Financial Economics 129, no. 3: 608–628. Available behind a paywall at https://doi.org/10.1016/j.jfineco.2018.05.003.
Bai, Gang and Elyas Elyasiani (2013). "Bank Stability and Managerial Compensation." Journal of Banking & Finance 37, no. 3: 799–813. Available behind a paywall at https://doi.org/10.1016/j.jbankfin.2012.10.026.
Bannier, Christina E., Eberhard Feess, and Natalie Packham (2012). "Competition, Bonuses, and Risk-Taking in the Banking Industry." Review of Finance 17, no. 2: 653–690. Available behind a paywall at https://doi.org/10.1093/rof/rfs002.
Bennett, Rosalind L., Levent Güntay, and Haluk Unal (2015). "Inside Debt, Bank Default Risk, and Performance during the Crisis." Journal of Financial Intermediation 24, no. 4: 487–513. Available behind a paywall at https://doi.org/10.1016/j.jfi.2014.11.006.
Bolton, Patrick, Hamid Mehran, and Joel Shapiro (2015). "Executive Compensation and Risk Taking." Review of Finance 19, no. 6: 2,139–2,181. Available behind a paywall at https://doi.org/10.1093/rof/rfu049.
Chesney, Marc, Jacob Stromberg, and Alexander F. Wagner (2018). "Managerial Incentives to Take Asset Risk." University of Zurich NCCR FINRISK Working Paper no 607, https://www.nccr-finrisk.uzh.ch/media/pdf/wp/WP607_B1.pdf.
DeYoung, Robert, Emma Y. Peng, and Meng Yan (2013). "Executive Compensation and Business Policy Choices at U.S. Commercial Banks." Journal of Financial and Quantitative Analysis 48, no. 1: 165–196. Available behind a paywall at https://doi.org/10.1017/S0022109012000646.
Edmans, Alex and Qi Liu (2010). "Inside Debt." Review of Finance 15, no. 1: 75–102. Available behind a paywall at https://doi.org/10.1093/rof/rfq008.
Efing, Matthias, Harald Hau, Patrick Kampkötter, and Johannes Steinbrecher (2015). "Incentive Pay and Bank Risk-Taking: Evidence from Austrian, German, and Swiss Banks." Journal of International Economics 96: S123-S140. Available behind a paywall at https://doi.org/10.1016/j.jinteco.2014.12.006.
Fahlenbrach, Rüdiger and René M. Stulz (2011). "Bank CEO Incentives and the Credit Crisis." Journal of Financial Economics 99, no. 1: 11–26. Available behind a paywall at https://doi.org/10.1016/j.jfineco.2010.08.010.
Fahlenbrach, Rüdiger, Robert Prilmeier, and René M. Stulz (2012). "This Time Is the Same: Using Bank Performance in 1998 to Explain Bank Performance during the Recent Financial Crisis." Journal of Finance 67, no. 6: 2139–2185. Available behind a paywall at https://doi.org/10.1111/j.1540-6261.2012.01783.x.
Institute of International Finance (2009). "Compensation in Financial Services Industry Progress and the Agenda for Change."
Jarque, Arantxa and Edward S. Prescott (2013). "Banker Compensation and Bank Risk Taking: The Organizational Economics View." Federal Reserve Bank of Richmond Working Paper no. 13-03, https://ssrn.com/abstract=2228910.
John, Kose, Anthony Saunders, and Lemma W. Senbet (2000). "A Theory of Bank Regulation and Management Compensation." Review of Financial Studies 13, no. 1: 95–125. Available behind a paywall at https://doi.org/10.1093/rfs/13.1.95.
Kupiec, Paul (2013). "Incentive Compensation for Risk Managers When Effort Is Unobservable." AEI Economic Policy Working Paper 2013-07, https://ssrn.com/abstract=2344907.
Noe, Thomas H., Michael J. Rebello, and Larry D. Wall (1996). "Managerial Rents and Regulatory Intervention in Troubled Banks." Journal of Banking & Finance 20, no. 2: 331–350. Available behind a paywall at https://doi.org/10.1016/0378-4266(94)00127-8.
Thanassoulis, John (2012). "The Case for Intervening in Bankers' Pay." The Journal of Finance 67, no. 3: 849–895. Available behind a paywall at https://doi.org/10.1111/j.1540-6261.2012.01736.x.
Tzioumis, Konstantinos and Matthew Gee (2013). "Nonlinear Incentives and Mortgage Officers' Decisions." Journal of Financial Economics 107, no. 2: 436–453. Available behind a paywall at https://doi.org/10.1016/j.jfineco.2012.08.014.
Van Bekkum, Sjoerd (2016). "Inside Debt and Bank Risk." Journal of Financial and Quantitative Analysis 51, no. 2: 359–385. Available behind a paywall at https://doi.org/10.1017/S0022109016000168.
Wall, Larry D. (2019). "Is Stricter Regulation of Incentive Compensation the Missing Piece?" Journal of Banking Regulation. Available behind a paywall at https://doi.org/10.1057/s41261-019-00119-8.
1 For recent examples of this sort of discussion, see here and here.
2 For example, see my earlier posts "Simple Concept, Complex Regulation" and "Bail-in Debt: Will the Supervisors Pull the Trigger in Time?"
3 See Edmans and Liu (2011) for a model with equity-based compensation, cash bonuses, and inside debt. John, Saunders, and Senbet (2000) and Noe, Rebello, and Wall's (1996) papers examine the interaction of compensation and other regulatory policies.
4 See DeYoung, Peng, and Yan (2013), Bai and Elyasiani (2013), Fahlenbrach and Stulz (2011), Bennett, Güntay, and Unal (2015), and Chesney, Stromberg, and Wagner (2018).
5 For example, see Van Bekkum (2016) and Bennett, Güntay, and Unal (2015).
6 See Acharya, Litov, and Sepe (2013), Efing, Hau, Kampkötter, and Steinbrecher (2015), Tzioumis and Gee (2013), and Agarwal and Ben-David (2017).