Fabrizio Ferri is Associate Professor of Accounting at University of Miami Business School; and Robert F. Göx is Professor and Chair of Managerial Accounting at the University of Zurich. This post is based on their monograph, recently published in Foundations and Trends in Accounting. Related research from the Program on Corporate Governance includes The CEO Pay Slice by Lucian Bebchuk, Martijn Cremers and Urs Peyer (discussed on the Forum here) and Paying for Long-Term Performance by Lucian Bebchuk and Jesse Fried (discussed on the Forum here).
In our monograph Executive Compensation, Corporate Governance, and Say on Pay, we provide a comprehensive summary and survey of the theoretical and empirical literature on Say on Pay. In the first part of the monograph, we study theoretically how a poor governance structure affects the level and structure of executive pay and identify conditions under which Say and Pay could help shareholders to improve it. In the second part of this monograph, we explain the origins and the cross-country differences in Say on Pay regulation and provide a detailed summary and evaluation of the empirical evidence on the subject.
The core issue among the proponents of the shareholder value view and the managerial power approach is the question of whether executive pay in public firms represents arm’s-length bargaining between managers and shareholders or rent seeking by powerful CEOs. Yet, formal models of executive pay are typically based on the shareholder value view and only a few of them explicitly study the consequences of the firm’s governance structure on its compensation decisions. In Chapter 2, we propose a conceptual framework that allows us to formalize the consequences of a poor governance structure on the board’s compensation decisions and to compare the properties of the contract proposed by a weak board to the optimal contract designed in the best interest of shareholders. This framework serves as a benchmark for studying the economic consequences of Say on Pay in Chapter 3.
We portray the agency problem between shareholders and managers as a problem of moral hazard. Different from the standard model, we assume that the firm’s compensation decisions are taken by the board of directors and not by the firm’s shareholders. We study the optimal compensation contract under various restrictions faced by the board when setting the agent’s compensation and compare the solutions with the contract that maximizes shareholder value. These restrictions include limited liability, possible forms of an “outrage constraint”, and the “million dollar tax cap” of the Internal Revenue Code, Section 162(m).
Not surprisingly, we find that a management-friendly board always inflates the CEO’s compensation level albeit the form of the optimal compensation contract critically depends on the relevant contractual constraints. Perhaps less obviously, we also present formal arguments showing that shareholders must not necessarily suffer from the presence of a management-friendly board. First, we show that shareholders strictly benefit from a moderately management-friendly board if it has superior information about the agent’s marginal contribution to firm value. Second, we demonstrate that the need to provide the CEO with incentives for the search of profitable investment projects can render a management-friendly board beneficial to shareholders.
In Chapter 3, we extend our model to study the economic consequences of Say on Pay. We begin the analysis with the advisory Say on Pay model used in the Anglo-Saxon countries and show that it can be a powerful instrument for shareholders to interfere with the compensation policy of the board. Its effectiveness critically depends on the consequences of a negative shareholder vote faced by the board of directors. The stricter the regulatory environment, the higher the willingness of the board to limit the agent’s compensation to avoid a negative voting outcome. However, this mechanism is only unambiguously desirable from a shareholder perspective if they possess the relevant information to determine the efficient compensation level. Otherwise, shareholders run the risk to distort erroneously the compensation policy of a board acting in their own best interest.
We also study the consequences of two forms of the binding Say on Pay model as used in some European countries. We first study the case where the binding Say on Pay vote is retroactive and show that it creates a hold-up problem on the part of shareholders that could destroy shareholder value if the contractual obligations from the compensation contract are subject to shareholder approval. We also study the case where the binding Say on Pay vote is prospective and show that the threat of disapproving the agent’s compensation ex ante is only effective if the shareholders do not have full control over the pay level proposed by the board. Overall, our analysis suggests that the effectiveness of Say on Pay and its desirability from a shareholder perspective critically depend on the incentives and the information of the parties involved in the pay-setting process as well as on the organization and the legal and economic consequences of the vote.
In Chapter 4, we provide an overview of the empirical research on the subject. We begin with a brief history of Say on Pay, placing it in the broader context of the trend toward greater shareholder democracy. We then review the empirical evidence on the effect of advisory Say on Pay votes, respectively, on executive pay and firm value, both in the US and in other countries. Finally, we review the corresponding evidence regarding binding Say on Pay regimes and discuss other issues related to Say on Pay votes. Overall, across various countries adopting Say on Pay, a few common findings emerge.
First, failed Say on Pay votes are rare, though cases of significant voting dissent are not uncommon. This may indicate that executive pay problems may not be as widespread or that a large fraction of investors are reluctant to interfere with and micromanage the pay-setting process. Voting dissent appears to be higher at firms with excess CEO pay (i.e. high pay and poor performance) and firms with compensation provisions viewed as reducing pay-for-performance. In many countries, proxy advisors play an important role in shaping shareholders’ votes.
Second, with respect to its effect on executive pay, the adoption of Say on Pay and adverse Say on Pay votes are followed by an increase in pay-for-performance sensitivity, while pay levels do not seem to be much affected. Firms often directly respond to adverse votes by engaging with institutional investors and changing compensation contracts to remove those controversial provisions that caused the adverse vote.
Third, with respect to the effect on firm value, most studies document a positive stock price reaction to events suggesting the future adoption of Say on Pay, though the stock price reaction to Say on Pay-induced actual compensation changes is either negative or insignificant. One possibility for these apparently conflicting findings is that investors’ (positive) expectations of the effects of the Say on Pay regime have not materialized. Another potential explanation is that those expectations were not driven by anticipated improvements to compensation contracts but other anticipated side benefits of Say on Pay (i.e. greater pressure on managers to perform, greater engagement with key investors on issues other than executive pay).
The complete monograph is available here.