There has been a recent surge in scholarship on the issue of concentration of power in the CEO, and the subsequent consequences for shareholder wealth maximization and board primacy. There is a general consensus among scholars that, in general, more powerful CEOs (relative to the board as representative of the corporate shareholders) can exacerbate agency conflict, resulting in suboptimal corporate strategies that are detrimental to corporate performance, and as a result, damaging to shareholder interests as well. The basic cause of this excessive power with CEOs lies in the outside board members being dependent on the CEO for their selection, continuation, and remuneration. More problematic is the fact that if a board is populated with a large number of firm executives (inside directors) directly reporting to the CEO, the board power is further compromised. These inside directors, as executives reporting to the CEO, automatically become deferential to the CEO in the hierarchical corporate structure. The proposed solution is to have more outside directors who are not firm employees subordinate to the CEO, and hence, independent of the influence of the CEO.

This raises a very critical question, the answer to which has dodged researchers for decades. Assuming that board independence can be achieved—and some have argued convincingly that it is a very strong assumption—does it really lead to containing CEO power? Answering this question has been a challenge at two different levels. The first challenge lies in defining and measuring CEO power in a manner that reflects actual CEO influence in materially shaping the strategic choices that a firm makes rather than just a manifestation of perceived influence. The second challenge lies in robustly establishing if board independence leads to limiting the CEO power relative to that of the board and the rest of the management team. One must convincingly rule out the possibility that the two are merely correlated with no causality running from board independence to less concentrated CEO power. More difficult is the task of tackling the possibility that the causation runs in reverse direction. That is to say that more powerful CEOs construct more independent boards by placing more outsiders on the board but including overall less informed members (Brown, Jr. 2012), and thereby, usurping extra power.

This paper tries to overcome these challenges while seeking the answer to the question of whether or not board independence leads to limiting the power of a CEO.

In response to the first hurdle of defining and measuring CEO power, the paper uses a robust, objective, and increasingly validated measure of CEO power—CEO Pay Slice (CPS). Developed by Bebchuk, Cremers and Peyer (2011), CPS is measured in terms of the CEO’s total compensation as a fraction of the combined total compensation of the top five executives in a given company. CPS provides a useful proxy for the relative centrality of the CEO in the top management team. This measure has been shown to have strong explanatory power for a diverse set of critical corporate strategic actions and the consequent performance outcomes, including firm profitability and corporate value.

With respect to the challenge related to distinguishing actual causation, the question of correlation and causation, the paper analyzes two external triggers. The first is known as the Sarbanes-Oxley Act (SOX), which mandated increased board accountability, which in turn impelled the non-compliant firms to appoint more independent board members. The second trigger was changes in board composition mandated by the new NYSE and NASDAQ listing rules, which contributed to greater board independence at some firms. The main provisions of these rules mandate that the board of each public company have a majority of independent directors.

The research setting of the study is a sample of 5,912 firm-year observations for the U.S. public listed companies from 1996 to 2010. The basic question that the quantitative research design is able to answer is whether or not the increased board independence, as mandated by new legal requirements, has reduced CEO power as measured by CPS. In order to ensure that the effect of board independence is clearly delineated from other firm characteristics influencing CEO power, the study controls for a large number of firm attributes. This study is the first to investigate the causal effect of board independence on CEO power using this quasi-experimental approach wherein external triggers initiate changes in corporate governance structure.

The fundamental results of the study suggest that the firms that were non-compliant with board independence requirements prior to the passing of the SOX Act in 2002, and raised the representation of independent members on their board in response to the act, experienced a reduction in CEO power. An interesting finding is that while in the Post-SOX era, the CEO pay slice is higher for the entire sample of firms, the increase in the CEO pay slice is significantly lower for the firms that raised the level of their board independence. The results hold and the same conclusion is reached even after controlling for the variations across firms based on their industry affiliation. Utilizing a battery of robustness checks, the paper establishes not only the statistical significance of the results, but also, the economic implications of the conclusions made. The firms that improved their board independence following the SOX Act experienced a significant reduction in relative CEO power by 25 percent.

The preceding post comes to us from Pornsit Jiraporn, Associate Professor of Finance at the Pennsylvania State University – School of Graduate Professional Studies, Seksak Jumreonwong, Associate Professor in the Department of Finance at Thammasat University, Napatsorn Jiraporn, assistant professor of marketing at the State University of New York at Oswego, and Simran Singh, Lang Center Associate for Political Engagement & Public Policy at Swarthmore College. The post is based on their recent paper, which is entitled “How Do Independent Directors View Powerful CEOs? Evidence from a Quasi-Natural Experiment” and available here.