Voting Choice for Fund Investors

The tremendous growth of institutional investors, particularly large passive funds, has drawn attention to their role in corporate governance. Major investment advisers such as BlackRock, Vanguard, and State Street are among the largest shareholders in many publicly traded firms and cast votes on behalf of millions of investors. At the same time, shareholder disagreements over voting issues are increasingly prevalent given the growth in environmental and social (E&S) issues that appear on companies’ agendas.

These trends have generated an intense debate about whether asset managers are in the best position to vote their investors’ shares. Do funds’ votes represent their investors’ preferences? Are their votes sufficiently informed? This debate has become especially heated in the context of E&S issues and has led to several important institutional and regulatory developments. In October 2021, BlackRock announced the “Voting Choice” program, which allows investors to either delegate their votes to BlackRock, as had been the default, or exercise their voting rights themselves (so-called “pass-through voting”). As of September 2022, investors representing 25 percent of BlackRock’s eligible assets had chosen to cast their own votes. While BlackRock initially offered this choice only to institutional clients, it soon expanded the program to some retail investors, and Vanguard and State Street followed suit. Lawmakers have been considering a more drastic change: The INvestor Democracy is EXpected (INDEX) Act, introduced in the U.S. Senate in May 2022, aims to require passively managed funds to collect and vote according to instructions from all their investors.

When is delegating voting to the fund manager beneficial for a fund’s investor, and when will the investor prefer to cast his own vote? Do investors benefit from having the choice between delegating and voting themselves? Does such “voting choice” dominate the two other extremes – the fund voting all its investors’ shares and all investors voting themselves? These are the questions we explore in our new paper.

We analyze a model in which the fund manager owns the firm on behalf of fund investors. There is a proposal up for a vote, whose value is uncertain. The fund manager gets a signal about the value of the proposal and casts the votes that are delegated to her. Without voting choice, all fund investors delegate their votes to the fund manager, whereas under voting choice, all investors independently decide whether to delegate their votes to the fund or to vote themselves.[1] Investors may choose to retain their votes for two reasons: to vote according to their own preferences or to use their private information about the proposal.

If the fund manager has only one investor, voting choice always benefits the investor, and investor welfare is maximized. Intuitively, the investor chooses to retain his votes and not delegate only when it is in his interest to do so. With multiple investors, however, the question of whether voting choice improves investor welfare is more nuanced because of a collective action problem: When an investor decides whether to retain his voting rights or delegate voting to the fund, he trades off the costs and benefits of delegation for himself but ignores the externalities imposed by his choice on other investors.

Our key conclusion is that whether voting choice improves investor welfare crucially depends on whether investors choose to retain their votes because of their private preferences or because of their private information.

We first consider the scenario in which fund investors are uninformed about the proposal but have heterogeneous preferences and may choose to retain their votes because of their private preferences. We show that in this case, voting choice can improve investor welfare if investors’ preferences are sufficiently dispersed and the number of fund investors is relatively small, but that it hurts investor welfare if the number of investors is large enough.

Intuitively, an investor’s decision to delegate his vote to the fund manager affects other investors in two ways. First, the decision is made according to the fund manager’s preferences rather than the delegating investor’s preferences (“preference effect”). Second, the decision is based on the fund manager’s information, whereas the investor’s vote would be uninformed (“information effect”). The preference effect on average hurts other investors. This is because the fund votes a block of shares, so when the vote is split, more investors oppose the fund’s vote than support it. On the other hand, the information effect benefits other investors. Whether voting choice dominates delegation of votes to the fund depends on the interplay between these two effects. We show that as the number of fund investors grows, the information effect eventually dominates. This is because the information effect applies to all investors, whereas the preference effect only applies to a subset of investors who retain their votes. Because each individual fund investor does not internalize this overall positive effect he imposes on other investors through delegation, there is too little delegation relative to the level that would maximize aggregate investor welfare. Moreover, the mere option of having voting choice is then detrimental: Investors would be on average better off if the fund manager voted all the shares and did not offer investors the choice to retain their votes.

The conclusions are very different if investors have heterogeneous information but are aligned in their preferences. In this case, equilibrium with voting choice achieves the efficient level of delegation, which maximizes aggregate investor welfare. As a result, voting choice dominates both the system in which the fund casts all the votes and the system in which all investors vote themselves.

These results have several policy implications. The case of multiple uninformed investors with heterogeneous preferences can capture the scenario in which the fund’s clients are small institutional investors or retail investors. Our model predicts that in this case, investors would be better off if the fund voted all their shares, as has been the default until recently. It also suggests that the INDEX Act, which proposes pass-through voting to all retail investors, may not achieve the goal of maximizing investor welfare. In contrast, the case of privately informed investors with aligned preferences can describe the scenario in which the fund’s clients are relatively large institutional investors focused on profit maximization. Voting choice in this case dominates both the status quo system and the system proposed by the INDEX Act. Finally, the scenario in which the fund’s clients are institutional investors with different ideologies, e.g., as in voting on E&S proposals, is likely to combine both cases, and whether voting choice improves investor welfare depends on their relative importance. In particular, voting choice can make investors worse off if they are not very informed about the financial benefits of the proposals.

We also analyze the implications of voting choice for information acquisition decisions. As we show, voting choice can lead to coordination failure: Even if it is collectively in the interest of investors to delegate voting and information acquisition to the fund, this may not happen in equilibrium. Intuitively, when fewer votes are delegated to the fund manager, she has less incentive to become informed about the proposals. As a consequence, there is a negative feedback loop: When investors choose to not delegate their votes to the fund, the fund manager acquires less precise information, which, in turn, leads investors to delegate even fewer votes to her and may result in less informed voting outcomes.

Our paper highlights that giving fund investors voting choice allows them to vote based on their preferences and to make use of their private information about the proposals. However, because fund investors do not consider the effects of their voting and delegation decisions on other investors, voting choice is not always beneficial to aggregate investor welfare: It may lead investors to withdraw their votes from the fund too often and may result in less informative voting outcomes. The optimal design of voting choice policies should take into account such unintended consequences.


[1] More specifically, under the “Voting Choice” program, the fund’s investors have a choice among delegating their vote to the fund manager, casting their own vote, and voting according to a custom policy offered by a proxy adviser that best aligns with their preferences. In our paper, for simplicity, we analyze the last two options together and treat both as the investor casting his own vote (since for both options, the investor’s vote closely represents the investor’s preferences).

This post comes to us from professors Andrey Malenko and Nadya Malenko at the University of Michigan’s Stephen M. Ross School of Business. It is based on their recent paper, “Voting Choice,” available here.