Why Are CEOs Rarely Fired? Evidence from Structural Estimation



    Search for more papers by this author
    • Lucian A. Taylor is from The Wharton School, University of Pennsylvania. I thank my dissertation committee at the University of Chicago Booth School of Business—Steven Kaplan, Gregor Matvos, Ľuboš Pástor (chair), Morten Sorensen, and Pietro Veronesi—as well as the anonymous referees, the editors, Mike Burkart, John Campbell, Doug Diamond, Alex Edmans, Xavier Gabaix, Raife Giovinazzo, Lukasz Pomorski, Joshua Rauh, Michael Roberts, Amit Seru, Toni Whited, Motohiro Yogo, and seminar participants at the 2009 EFA meetings, 2009 Winter Econometric Society meeting, 2008 CEPR European Summer Symposium in Financial Markets, Arizona State University, Harvard University, London Business School, New York University, University of California at Berkeley, University of Chicago, University of Illinois Urbana-Champaign, University of Michigan, University of Pennsylvania, University of Rochester, University of Texas at Austin, and University of Wisconsin at Madison for helpful comments. I am very grateful to Robert Parrino for providing data.


I evaluate the forced CEO turnover rate and quantify effects on shareholder value by estimating a dynamic model. The model features learning about CEO ability and costly turnover. To fit the observed forced turnover rate, the model needs the average board of directors to behave as if replacing the CEO costs shareholders at least $200 million. This cost mainly reflects CEO entrenchment rather than a real cost to shareholders. The model predicts that shareholder value would rise 3% if we eliminated this perceived turnover cost, all else equal. The model also helps explain the relation between CEO firings, tenure, and profitability.