Shareholder Votes and Proxy Advisors: Evidence from Say on Pay


  • Accepted by Douglas Skinner. We thank Glass Lewis & Co. for providing its 2011 proxy season reports, Carol Bowie (at Institutional Shareholder Services), and Robert McCormick (at Glass Lewis & Co.) for insightful conversations as well as an anonymous referee, Ana Albuquerque, Conrad Ciccotello, Yaniv Grinstein, Wayne Guay, Camelia Kuhnen, David Larcker, Eddie Riedl, and seminar/conference participants and discussants at the University of Pennsylvania (Wharton), Boston University, Burton Workshop at Columbia Business School, London Business School, Oklahoma State University, the 2012 NBER Summer Institute, the 2012 Annual Meeting of the European Association of Law and Economics, the 2012 Annual Corporate Governance at the University of Delaware, the 2012 Financial Economics and Accounting Conference, the 2012 Conference for Empirical Legal Studies, and the 2012 German Finance Association meeting for helpful comments. We also thank Zhenhua Chen, Miguel Duras, Thomas Steffan, and Forrester Wong for research assistance. Fabrizio Ferri acknowledges a grant from the Eugene M. Lang Support Fund of Columbia Business School. All errors remain our own. An online appendix is available for download here:


We investigate the economic role of proxy advisors (PAs) in the context of mandatory “say on pay” votes, a novel and complex item requiring significant firm-specific analysis. PAs are more likely to issue an Against recommendation at firms with poor performance and higher levels of CEO pay and do not appear to follow a “one-size-fits-all” approach. PAs’ recommendations are the key determinant of voting outcome but the sensitivity of shareholder votes to these recommendations varies with the institutional ownership structure, and the rationale behind the recommendation, suggesting that at least some shareholders do not blindly follow these recommendations. More than half of the firms respond to the adverse shareholder vote triggered by a negative recommendation by engaging with investors and making changes to their compensation plan. However, we find no market reaction to the announcement of such changes, even when material enough to result in a favorable recommendation and vote the following year. Our findings suggest that, rather than identifying and promoting superior compensation practices, PAs' key economic role is processing a substantial amount of executive pay information on behalf of institutional investors, hence reducing their cost of making informed voting decisions. Our findings contribute to the literature on shareholder voting and the related policy debate.