Institutional Investor Voting Behavior: A Network Theory Perspective
59 Pages Posted: 9 Apr 2018
Date Written: April 2018
This paper shows how network theory can improve our understanding of institutional investors’ voting behavior and, more generally, their role in corporate governance. The standard idea is that institutional investors compete against each other on relative performance and hence might not cast informed votes due to rational apathy and rational reticence. In other words, institutional investors have incentives to free ride, instead of “cooperating” and casting informed votes. We show that connections of various nature among institutional investors, whether from common ownership, geographical proximity or formal networks, and among institutional investors and other agents, such as proxy advisors, contribute to shaping their incentives to vote “actively” and also create multi-level competition dynamics: competition takes place not only among institutional investors (and their asset managers), but also at the level of their employees and among “cliques” of institutional investor. Employees compete for better jobs, and for that purpose obtain more information on portfolio companies than may be strictly justified from their employer institution’s perspective, and circulate it within their network. At the networks level, “cliques” of institutional investors compete against each other. Because there are good reasons to believe that cliques of cooperators outperform cliques of non-cooperators, the network-level competition might increase the incentives of institutional investors to collect information. These dynamics can enhance institutional investors’ engagement in portfolio companies and also shed light on some current policy debates such as the antitrust effects of common ownership and mandatory disclosures of institutional investors’ voting.
Keywords: Institutional Investors, Corporate Governance, Network Theory, Multi-level Competition, Cliques
JEL Classification: G20, G28, G30, G34
Suggested Citation: Suggested Citation