Hamid Mehran discusses proposals for reforming bankers’ compensation to better align incentives and risk-taking. Mehran contends market-based, self-regulating compensation structures would enhance governance and financial stability more effectively than punitive bonus caps or clawbacks imposed by regulators.
Bankers’ pay is again in the news. In Europe, Deutsche Bank is asking that the cap on bonuses be lifted for banks in the European Central Bank (ECB) zone, saying that the elimination would help as it can attract and retain talent when competing with U.S. and U.K banks. The latter lifted bank bonus caps at the end of October of last year.
Why is a focus on bonus compensation for bankers important? To illustrate the magnitude of compensation in the U.S. context, the three largest commercial banks paid on average about $33 billion to their employees in 2022, 90% of which was awarded to about 20% of the bank employees. The bulk of pay in banking is in the form of bonuses tied to various performance benchmarks. Thus, focusing on bonuses to control risk and enhance financial stability is reasonable. But how to establish a link between risk and bonus distribution is not obvious.
The European Union adopted its bonus cap in the aftermath of the global financial crisis effective January 1, 2014, when the ECB regulators speculated that bank bonuses had induced some of the risk-taking that led to systemic problems in the financial sector. However, researchers since that time have questioned the benefit of a bonus cap in restoring financial stability because of its potentially unintended consequences, for example employee retention. Fast forward to today, when the ECB’s current view is that chief risk officers (CROs) in banking firms should have greater control over bonus payments because as its distribution might adversely impact bank capital adequacy for example. The assumption is that it would be more effective to cap the bonus pool of a bank as a punishment if it is in breach of supervisory rules. In practice, bank CROs are already in a naturally hostile position within their workplaces. The ECB proposal would create an even more tense working environment for bank CROs, which could hamper their task of managing the enterprise risk. The U.K. compensation reform also has shortcomings that I will discuss below.
In the U.S., progress on bank compensation reform has stalled since 2016, even though it was mandated in the Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010. In that act, the U.S. government asked bank regulators and supervisors—such as the Federal Reserve and FDIC—to develop a framework for prudent compensation. However, the agencies’ suggested approach to was not workable as it required alignment of bonus payments with risk taking activities, a difficult exercise. Further, in the proposed approach bank supervisors would not be able to examine whether the bank complied with the compensation regulation; see Mehran (2022) for further discussion. Thus, no compensation regulation was adopted beyond in the U.S.
More recently, shortly after the failure of Silicon Valley Bank (SVB) and others in spring 2023, lawmakers considered adopting clawback of compensation of the management of banks that might fail in the future. That proposal was effectively dropped as well, partly because of industry push back, and partly because the idea seemed to fail to be a deterrent factor ex ante. The same idea, however, was proposed by the Swiss Financial Authority at the end of last year.
Why so much interest in compensation reform in the banking industry and why so little progress? A 2022 paper I wrote covers the history of attempts to reform bankers’ compensation over the past three decades. Briefly, the fact is that the labor market for banks’ top management is different from that for nonfinancial firms’ leaders. Turnover is far lower, even when banks perform poorly. For example, the top managers of large U.S. banks during the financial crisis, who imposed so much cost on the U.S. and elsewhere, kept their jobs. Poor performance did not impact the career horizon of those executives, as stability of the top management team is important in a time of crisis. Thus, the distortion that is inflicted in the functioning of the labor market should be addressed through other governance mechanism(s). To be sure, regulatory reforms such as capital and liquidity requirements, as important as they are in financial stability, have little impact on bank governance. For these reasons, among others, some argue that regulators should control bankers’ pay on the grounds that compensation drives incentives and that those incentives induce (or reduce) risk-taking. That being said, the competitive labor market principal—a requirement in the U.S. and ingrained in economics—must be taken into account: Limiting pay for talent and performance is unacceptable. Regulators also need to accept that they cannot monitor the actions of bank executives and find, ex ante, a link between risk taking and pay, particularly if pay is in the form of bonuses. Thus, approaches that incorporate these realities are critical in designing pay packages in banking.
What is the objective of pay in financial firms? It is a reward and retention mechanism, just as in nonfinancial firms. But in banking, pay also should be, from the public stakeholders’ point of view, a mechanism of control. Not being able to observe the bank risk ex ante, the public wants to make sure that the banks have adequate capital and liquidity and that they have made judicious investment, financing, and payout decisions. So, the goal is to design pay schemes to achieve these objectives. A historical note is likely to be instructive on this point.
In August 2000, JPMorgan Chase chairman Sandy Warner brought to William McDonough, then president of the Federal Reserve Bank of New York, a concern that guaranteed bonuses were becoming prevalent in the industry and cautioned that they could have an effect on bank viability in the event of a shock, as banks might not be able to adjust their costs. Regulators, apparently for the first time, learned that bonuses could be destabilizing, and the alert came from a banker.
Warner did not say that banks should limit their bonus payment, rather he was concerned about the effect of bonuses on bank capital. Thus, the focus of compensation reform should be on capital and liquidity.
In a paper I authored with Bolton and Shapiro (2010), we offered a solution to this challenging issue. Our aim is not to determine how much bankers should be paid, but how bankers should be compensated. We argue that in order to ensure that bankers’ decisions do not adversely impact the public (or creditors), compensation should be tied to bank risk exposure. Here’s how it works: Assume that the market assesses bank risk by observable sources of information. In large banks, the risk is captured by credit default swap (CDS) spreads. The bank board at the beginning of the year decides on the risk, proxied by the CDS spread. Some or all of the bonuses could be retained if the CDS spread is above the determined spread at the end of the year. This is a transparent approach and does not require knowledge about who took risks and why. Another important point is that while the market assesses that bank risk has gone up, the resources available to the bank to overcome capital shortfalls in the event of a problem expand as the bank withholds bonuses. The clawback of bonuses is likely to contribute to the health of the bank even in a crisis as the CDS spread widens not because of bank employee risk taking but because of changes in risk due to a new operating environment. The American Insurance Group (AIG) has already adopted a version of this idea in its senior management compensation packages.
What if there is no CDS spread on the bank and regulators’ private information is not captured in the bank security prices by even sophisticated analysts and investors or depositors? This is, in effect, the story of SVB. For the CDS spread scheme to work, the information requirement has to be considered. This is particularly relevant to its application to regional banks. An alternative would overcome the information requirement, which is the approach outlined by my paper with Tracy (2016), which advocates funded deferral of compensation for senior management. A determination must be made of how many employees would be covered by the scheme, what fraction of their bonuses should be deferred, and what the length of the deferral and the vesting requirements should be. The size of the deferral, its length, and vesting features could vary with rank in the organization. Application of this approach produces a pool of funds that could be utilized in a crisis. Under conservative assumptions, I show that had SVB adopted the scheme before its collapse, it would have accumulated enough to cover the shortfall that it tried to deal with via its failed equity offering in early 2023.
Neither the funded deferral approach nor the CDS spread scheme restricts the level of pay. Banks could consider both approaches, without regulatory interventions being used as a risk-management tool. Neither approach requires monitoring by regulators or by the market to control risk-taking activities by the bank and its employees.
In both schemes, monitoring is shifted to the employees who are the most effective monitors. If the firm suffers due to mismanagement, risky decisions or fraud, then the bonuses of all of the employees in the entire firm might be at risk. Losing a chunk of wealth by aggressive risk taking is surely a big deterrent for risk taking as well as for inducing internal monitoring. There is no impact on labor mobility, employees would not lose their claims by departing from the firm, but they would not be able to access their funds until the deferral and vesting requirements were fulfilled. The incentive for risk taking thus would be reduced before termination of employment. Deferral would not adversely impact the incentive for risk-taking as bonuses are directly related to performance.
The deferral is essentially a source of funds and would impact the bank’s credit rating. The approach has many similarities to the capital conservation buffer, but it would be built up with the compensation of actual risk takers, not bank owners. The annual stress test and the approaches outlined here have different features. The stress test is always ex post. The bank could be asked to limit payout to its owners and limit its bonus payments. The schemes introduced here are ex ante. The remedy to future problems would already be in motion.
The deferral approach has relevance to the recent litigation by the former chief executive officer of Wells Fargo, who claimed that the firm illegally withheld compensation that was owed to him. Had the firm adopted the deferral plan for its executives, the litigation would be meritless. The firm also would have experienced fewer misconduct cases; see Mehran and Tracey (2016).
It is important to note while most large banks in the U.S. and in Europe already have deferred compensation plans, funded deferred compensation plans discussed earlier are rare. In the current environment, banks do not retain cash against the compensation that is deferred. The bank uses the fund available or earnings in the year that deferral is lifted. If the bank is doing poorly in that year, most likely it will write off the employees’ deferral claim. But the bank would not have a cash buffer to use in its operation.
The lack of a pool of funds to cover the bonuses when due is a serious issue for the reform of compensation in the U.K. The U.K. regulators failed to recognize that their design is likely to induce undertaking risky decisions by bank employees when the bank is vulnerable. Consider a situation when thousands of employees who have a significant claim in deferred compensation, in bonuses or equity, perceive that their claim is at risk. Many of them might take a bigger risk to turnaround the prospect of the firm, and in the process, they could impose a larger damage on the bank’s health. In addition to their incentive to save their claims, they have a much bigger incentive to gamble if they think that there will be large layoffs at the bank. Taking a big risk is justified if it pays off and they will be considered as high-performers, thus they might be retained, they think. This kind of behavior could be mitigated by having a buffer as suggested by the two schemes since it is a form of insurance against poor outcomes, and thus protects employees.
Strategic issues in banking firms (and non-banking firms) are people issues, and people issues are compensation issues. The way to impact strategic issues, including risk-taking incentives, is to design the right compensation structure, ex ante. Imposing pecuniary punishment after the fact, by limiting pay levels or bonuses, would not contribute to establishing a healthier banking institution. A CDS spread scheme tied to bonus distributions and funded deferral pay could help regulators and supervisors with their goal of enhancing financial stability. The approaches described here directly impact bank governance.
Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.