This paper studies how directors' reputational concerns affect board structure, corporate governance, and firm value. In our setting, directors affect their firms' governance, and governance in turn affects firms' demand for new directors. Whether the labor market rewards a shareholder‐friendly or management‐friendly reputation is determined in equilibrium and depends on aggregate governance. We show that directors' desire to be invited to other boards creates strategic complementarity of corporate governance across firms. Directors' reputational concerns amplify the governance system: strong systems become stronger and weak systems become weaker. We derive implications for multiple directorships, board size, transparency, and board independence.

While Section 2.1 compared the most e¢ cient equilibrium under common ownership with the benchmark, Proposition 7 considers all equilibria under common ownership. Proposition 7 (Comparison of equilibria, exit): Suppose L=n v (1 F ()). There is 2 [0; 1) s.t., if , any equilibrium under common ownership is strictly more e¢ cient than any equilibrium under separate ownership. Proof of Proposition 7. Suppose L=n v (1 F ()). From Lemma 3, the unique working threshold in any equilibrium under separate ownership is strictly smaller thanthat is, c so;exit <. Moreover, since c so;exit solves exit (F (c)) = c , lim !1 c so;exit = (1 !). From Lemma 4, a type (i) equilibrium under common ownership exists in which the working threshold is. From the proof of Lemma 4, the only other possible equilibrium is a type (i) equilibrium in which the investor retains bad …rms if = 0. If such an equilibrium exists, the working threshold must satisfy exit (F (c)) = c. Note that, as ! 1, any solution of exit (F (c)) = c converges to. Therefore, if is su¢ ciently close to 1, any equilibrium under common ownership is strictly more e¢ cient than any equilibrium under separate ownership.

Shareholder proposals are a common form of shareholder activism. Voting for shareholder proposals, however, is nonbinding since management has the authority to reject the proposal even if it received majority support from shareholders. We analyze whether nonbinding voting is an effective mechanism for conveying shareholder expectations. We show that, unlike binding voting, nonbinding voting generally fails to convey shareholder views when manager and shareholder interests are not aligned. Surprisingly, the presence of an activist investor who can discipline the manager may enhance the advisory role of nonbinding voting only if conflicts of interest between shareholders and the activist are substantial.

This paper studies communications between investors and firms as a form of corporate governance. The main premise is that activist investors cannot force their ideas on companies; they must persuade the board or other shareholders that implementing these ideas is in the best interest of the firm. In this framework, I show that voice (launching a public campaign) and exit (selling shares) enhance the ability of activists to govern through communication. The analysis identifies the factors that contribute to successful dialogues between investors and firms. It also shows that public communications are likely to be ineffective, justifying the prevalence of behind-the-scenes communications.

We also thank Alon Brav and Carola Schenone for sharing data on activists and airlines, respectively. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.