November 02, 2009
Incentive Compensation, Risk Management, and Safety and Soundness
Governor Daniel K. Tarullo
At the University of Maryland's Robert H. Smith School of Business Roundtable: Executive Compensation: Practices and Reforms, Washington, D.C.
It is a pleasure to be here this afternoon to give the closing address at this very timely conference on executive compensation. Executive compensation has for some time been an important part of academic and policy discussions of corporate law and governance. The public's attention to the subject has periodically increased. Such heightened interest is not often part of corporate governance debates, but rather in response to high compensation levels for certain executives, or the contrast between rising executive compensation and sluggish overall wage growth. I think it fair to say, however, that public scrutiny has never been more intense than in the past year. And never has it been more concentrated on a particular industry than in the current focus on large financial services firms.
The interest and role of the Federal Reserve is both broader and more targeted than the issue of executive compensation as such. It is more targeted in that our concern with compensation matters rests squarely on our statutory responsibility to ensure the safety and soundness of the banking organizations we regulate. It is broader in that we are concerned not only with the compensation of executives, but also with compensation arrangements of employees other than executives where those arrangements may adversely affect the safety and soundness of their firms. In the remainder of my remarks, I will first elaborate a bit on how our safety and soundness mandate both defines and limits the scope of our activities on compensation. Then I will turn to the supervisory approach we have taken in the proposed guidance we issued 10 days ago, before closing with an explanation of the specific supervisory processes we intend to follow in putting this guidance into practice.1
Compensation as a Safety and Soundness Issue
Previous speakers have touched on some of the reasons why compensation arrangements of corporate employees can be a public policy matter. As I have already suggested, the compensation packages of employees--particularly senior executives--implicate the principal-agent concerns that inform much of corporate law. Shareholders have interests both in providing employees with incentives to maximize shareholder value and in restraining employees from using their direct control over the firm's resources to take more than is necessary to attract and motivate quality employees. Additionally, as evidenced by the very fact of Ken Feinberg’s job as special master, the public has an interest in appropriate compensation arrangements at firms in which the government has made significant investments.
Broader issues of fairness may also be raised by high compensation levels for some employees. Of course, these issues can quickly involve rather fundamental considerations of political philosophy and political economy. At present, the public debate over the fairness of compensation levels in financial firms has a more specific focus. Many would argue that the bonuses that may be collected this year by employees of some financial firms and, for that matter, the profits accruing to the firms themselves, are possible only because the government rescued the entire financial system through a series of actions beginning last year. Seen in this light, the bonuses and profits appear to some observers as an unfair channeling of gains to private actors, after potentially devastating losses to those same firms and employees were avoided by placing taxpayer funds at risk.
Each of the foregoing perspectives on compensation raises important policy issues. But the regulatory framework established by statute makes them relevant to the supervisory functions of the Federal Reserve only to the extent they affect the safety and soundness of financial institutions. An evaluation of some compensation practices at banking organizations preceding the financial crisis reveals that they did, in fact, contribute to safety and soundness problems. Indeed, financial firms have themselves identified compensation practices as a factor. In this regard, I was struck by the results of a survey conducted earlier this year on behalf of the Institute of International Finance.2 Of the 37 large banking organizations engaged in wholesale banking activities that responded, 36 agreed that compensation practices were a factor underlying the crisis.
Of particular note are incentive compensation arrangements for employees in a position to expose the firm to substantial risk that failed to align the employees' interests with those of the firm. For example, some firms gave loan officers incentives to write a lot of loans, or traders incentives to generate high levels of trading revenues, without sufficient regard for the risks associated with those activities. The revenues that served as the basis for calculating bonuses were generated immediately, while the risks might not have been realized for months or years after the transactions were completed. When these or similarly misaligned incentive compensation arrangements were common in a firm, the very foundation of sound risk management could be undermined by the actions of employees seeking to maximize their own pay.
There is thus significant overlap between the interests of shareholders and of supervisors in the area of employee compensation. That is why, as I will explain in a moment, we include an emphasis on sound corporate governance measures in our proposed guidance. Indeed, supervisory initiatives can reinforce the impetus for change already coming from shareholders and other market participants. By requiring banking organizations to incorporate a thorough, risk-appropriate approach to incentive compensation arrangements, supervisors can help address concerns that these firms face a kind of collective action problem, which may make a firm reluctant to rectify misaligned arrangements for fear that other firms will poach the best employees with promises of substantial near-term bonuses. This concern exists at the international, as well as national, level, which is one of the reasons why the Financial Stability Board has devoted so much attention to the issue in the past year.
However, it is important to note that the interests of shareholders and supervisors, representing the public, are not wholly congruent. Because of the federal safety net--as manifested in the deposit insurance system, the availability of discount window lending, and too-big-to-fail perceptions--shareholders may at times be willing to tolerate a degree of risk in a firm’s activities that does not fully reflect the costs of that risk to society as a whole. Thus, even if banking organizations were able to achieve full harmonization of employee and shareholder interests in their incentive compensation arrangements, supervisory oversight would still be warranted.
Proposed Supervisory Guidance
The proposed guidance recently issued by the Federal Reserve is informed by the considerations I have just set forth.3 Our rationale for the guidance is grounded in safety and soundness concerns. Our supervisory approach emphasizes the need to integrate compensation policy into the overall risk-management framework of financial firms. This dual focus on safety and soundness aims and risk-management capacities has been a recurring theme of supervisory efforts in the wake of the crisis. Our compensation guidance is just one piece of a broad program of regulatory and supervisory reforms, which include strengthened capital and liquidity requirements and a more coordinated, "horizontal" approach to large-firm supervision.
The Federal Reserve’s guidance is based on three fundamental principles. First, incentive compensation arrangements at a banking organization should not provide employees with incentives to take risks that are beyond the organization’s ability to effectively identify and manage. The amounts of incentive pay flowing to employees should reflect the risks and potential losses--as well as gains--associated with their activities. Employees are less likely to take imprudent risks if their incentive payments are reduced or eliminated for activities that end up imposing significant losses on the firm.
The guidance identifies four methods that have been most often identified by firms, consultants, and others as useful in aligning incentive compensation with risk. These include adjusting performance awards to reflect the risks of employee activities, deferring payments of awards and adjusting actual payments to reflect risk outcomes, using longer periods for measuring the performance on which awards are based, and reducing the sensitivity of performance measures to short-term revenues or profits. Our expectation is that each banking organization will use these and possibly other methods, either individually or in combination, to achieve risk-appropriate incentive compensation practices throughout the firm.
To be fully effective, risk adjustments to compensation should take account of the full range of risks that the activities of employees may pose for the firm, including credit, market, compliance, reputational, and liquidity risks. Moreover, these adjustments need to be implemented in practice so that actual payments vary based on risks or risk outcomes. If employees are paid substantially all of their potential incentive compensation when risk or risk outcomes are materially worse than expected, employees may have an incentive to take large risks in hopes of substantially increasing their personal compensation, knowing that their downside risks are limited. Simply put, incentive compensation arrangements should not create "heads I win, tails the firm loses" expectations.
The second core principle is that incentive compensation arrangements should be compatible with effective risk management and controls. Banking organizations should ensure that risk-management personnel have an appropriate role in designing incentive compensation arrangements and assessing their effectiveness in restraining excessive risk-taking. Risk-management breakdowns often occurred in firms not because risk-management officers were unaware of problems, but because they were not adequately integrated into the firm’s strategic and operational decisionmaking processes related to compensation. In addition, a banking organization should have strong controls governing its process for designing, implementing, and monitoring incentive compensation arrangements to help prevent employees from circumventing or weakening these processes in ways designed to increase their personal compensation.
The third core principle is that incentive compensation arrangements at banking organizations should be supported by strong corporate governance practice, including active and effective oversight by boards of directors. This oversight should go beyond reviewing and approving the compensation arrangements for senior executives. Boards of directors, or their compensation committees, should regularly review the design and functioning of the firm’s incentive compensation practices to ensure their consistency with safety and soundness.
In developing these principles, we were guided by two key considerations. The first, as I noted earlier, is that problematic compensation practices were not limited to the most senior executives at financial firms. Traders with large position limits and loan officers who, as a group, originate loans accounting for a material amount of an organization’s credit risk, are two examples of non-executive personnel who have the ability to expose a banking organization to material amounts of risk. For this reason, our guidance extends beyond the senior executives at banking organizations to include all other employees who, individually or as a group, may expose the firm to material amounts of risk.
Second, in drafting this guidance we opted for principles, rather than generally applicable rules. The complex and varying nature of financial activities, particularly those of large banking organizations, suggested that uniform requirements for compensation arrangements were not well-advised, at least not on the basis of our present knowledge and experience. Activities and risks, as well as the reliability of related risk measures, may differ significantly across banking organizations and across employee groups within one banking organization. In these circumstances, we concluded that use of a single, formulaic approach to making incentive compensation arrangements risk-sensitive would likely not be effective in adequately containing risky behavior.
As discussed in the guidance, incentive compensation arrangements for senior executives at a large, complex organization are likely to be better balanced if they involve deferral of a substantial portion of the executives’ incentive compensation over a multiyear period, with payment made in some form of equity that is ultimately calculated with reference to the performance of the firm during the deferral period. However, the use of equity-based incentive compensation may not be effective in aligning the incentives of mid- and lower-level employees with the interests of the firm, because these employees may view the outcomes of their decisions as unlikely to have much effect on the firm or its stock price. Moreover, deferral may not be effective in constraining the incentives of employees who have the ability to expose the firm to long-term or "bad tail" risks, as these risks are unlikely to be realized during a reasonable deferral period. Consider, for example, how different the incentive compensation packages might need to be for a trader who deals in only one kind of highly structured, long-term instrument, as compared to the arrangements for the firm’s chief executive officers.
In issuing the proposed guidance, the Federal Reserve also announced two separate supervisory initiatives to ensure that the principles are incorporated by banking organizations in their compensation arrangements. These initiatives are also intended to help identify and promote emerging best practices in the industry. Like the principles themselves, the separate supervisory programs recognize the differences among the many banking organizations we supervise.
The first initiative applies to the institutions we classify for general supervisory purposes as large, complex banking organizations (LCBOs). LCBOs warrant special supervisory attention because they are significant users of incentive compensation arrangements and because flawed practices at these institutions are more likely to have adverse effects on the broader financial system. The challenge with respect to these institutions was that while we did not want to impose inflexible, and possibly ineffective, practices across the wide range of compensation arrangements, we also did not want there to be insufficient attention paid to the very particulars that led us to adopt this approach in the first place. Accordingly, we decided to place the burden on each of these LCBOs to develop a plan that would, with appropriate specificity, implement the principles we have identified into the firm’s compensation policies and practices. A rigorous process of planning and implementation, with accompanying supervisory oversight, makes it possible for these firms to customize their compensation arrangements.
Our supervisory approach to compensation practices at LCBOs will include an extensive horizontal review process, in keeping with our increased emphasis on this approach for the supervision of the largest institutions. This horizontal review, which we expect to commence shortly, will be led by a multidisciplinary coordinating group of Federal Reserve staff. We anticipate that it will include staff from the Board, from Reserve Banks that supervise LCBOs, and from other financial regulators. This group also will serve as a resource for supervisory staff across the Federal Reserve System on incentive compensation matters.
This review process will enhance our understanding of existing incentive compensation practices, identify existing or emerging best practices across the organizations, and provide an informed framework for evaluating firms’ plans for improving their practices. In this regard, each LCBO will be expected to develop and present to the Federal Reserve a plan that identifies where improvements to the organization’s compensation arrangements and related practices are needed to bring them in line with the principles embodied in the guidance. These submissions must include the firm’s plans, including relevant timetables, for achieving meaningful and effective improvements to the risk-sensitivity of its incentive compensation arrangements and the firm’s related risk management, controls, and corporate governance processes. We expect to work with each organization through the supervisory process in refining its plan. Once finalized, implementation of the plan will become part of the supervisory expectations for the banking organization. As with other supervisory expectations, failure to submit or implement a satisfactory plan could be the basis for enforcement action.
We issued the compensation guidance in proposed form and will be accepting comments on the guidance through November 27. While we may well make changes in some aspects of the guidance based on comments we receive, it is critical that the incentives created by compensation arrangements be reviewed and adjusted to ensure the safe and sound operation of financial firms. Indeed, in light of the experience of recent years, the safety and soundness implications of incentive compensation arrangements would be an appropriate matter for supervisory oversight even in the absence of any specific guidance. For these reasons, we are not waiting to begin the process of gathering information about compensation practices from LCBOs. Today, in discussions across the country, we are communicating our plans and expectations to these firms, with particular attention to beginning this information gathering.
A separate program will apply to the thousands of other organizations supervised by the Federal Reserve, including community and regional banking organizations. Supervisory staff will review incentive compensation arrangements at these organizations as part of the regular risk-focused examination process, in connection with reviews of the organizations’ risk management, internal controls, and corporate governance. These reviews, as well as our supervisory expectations for these organizations, will be tailored to reflect the more limited scope and complexity of these organizations’ activities--a fact also recognized in various aspects of our guidance. For example, we certainly do not expect smaller banking organizations that use incentive compensation arrangements on a limited basis to have the same type of formalized and detailed program for managing the risks of such arrangements as we expect to see at LCBOs. We will also have an ongoing coordination process for developing and clarifying guidance for the non-LCBO institutions.
We expect that our supervisory practice in this area will evolve. We will have to build on the principles we have set forth and we may, over time, identify standards that should be universally applied to certain classes of employees. We will surely face questions from firms that we have not anticipated. But even as we can expect a rapid accretion of knowledge and significant elaboration of desirable policies and practices, the constant in our compensation oversight will be the emphasis on safety and soundness and supporting risk-management practices, which have been the foundation for our supervisory approach.
3. See Board of Governors of the Federal Reserve System (2009), "Federal Reserve Issues Proposed Guidance on Incentive Compensation," press release, October 22. Return to text