Institutional ownership has increased tremendously over the past few decades. This shift in ownership has led to major concerns as to whether institutional investors have the incentives or the tools to be active shareholders. Assessing how effective institutional investors are in their stewardship activities has therefore become critical. However, such an assessment has proven difficult, given that one does not observe the ex-ante preferences of institutional investors.
In my recent article, I overcome this challenge by constructing measures of mutual funds’ environmental, social, and governance (ESG) preferences from their proxy voting guidelines. In these regulatory documents, mutual funds describe how they generally vote on common ballot items at the shareholder meetings of their portfolio firms. For example, in its 2008 guidelines, BlackRock Advisors states that it will vote against shareholder proposals seeking to separate the roles of CEO and chairman.
I hand-collect over 17,000 of these voting policies from the 2006-2018 proxy voting guidelines of 29 of the largest US mutual fund families, including BlackRock, Vanguard, and State Street. I focus on the 100 most common ESG issues. Among others, my sample includes voting policies regarding proposals to de-stagger the board, to disclose political contributions, or to reduce CO2 emissions. I first examine whether mutual funds vote the way they say they vote. I then study whether portfolio firms adopt the announced ESG preferences of their mutual fund shareholders. If this is the case, through which channels do mutual funds convey their preferences to investee companies?
Mutual funds’ ESG preferences
My analysis of mutual funds’ proxy voting guidelines reveals a high level of heterogeneity among mutual funds’ voting policies. Some mutual funds support shareholder rights enhancement, while others support management independence. Similarly, some advocate that firms should take the environmental and social (E&S) implications of their actions into consideration, while others emphasize their fiduciary duty to maximize shareholder value. Furthermore, I find that voting policies change extensively over time. I especially document a significant increase in the support for E&S issues over the sample period. These results indicate that different ESG preferences coexist among institutional investors. They also provide novel evidence that these preferences are not stable.
Do mutual funds vote as they say they vote?
Many factors, such as proxy advisors’ or management recommendations, may influence mutual funds’ votes. For this reason, the informativeness of proxy voting guidelines about mutual funds’ preferences is unclear. Therefore, I compare mutual funds’ announced voting policies to their actual votes. I find that mutual funds comply with their voting policy in 80% of the cases. In other words, voting policies are a major predictor of mutual funds’ votes, ahead of the largest proxy advisor’s (ISS) recommendations and ahead of management recommendations. I use the staggered changes in voting policies to show that when mutual funds modify their voting policies, they adapt their voting strategies accordingly.
However, my results also highlight many cross-sectional differences. For example, some families rely on their guidelines for over 95% of the votes while others fall below 50%. Moreover, funds follow their policies in less than 45% of the cases when the policies support E&S issues. In comparison, when their policies oppose E&S issues, their compliance level is as high as 96%. These results suggest that mutual funds present themselves as much more friendly towards environmental and social issues than they actually are when it comes to voting.
Do portfolio firms adopt the ESG preferences of their mutual fund shareholders?
Next, I investigate whether portfolio companies adopt the ESG policies that their mutual fund shareholders have stated to favor. One challenge for my analysis is the endogenous nature of mutual funds’ portfolio selection as mutual funds may simply select firms that have already adopted their preferred ESG structure. To tackle this identification challenge, I exploit the staggered changes in voting policies across mutual fund families. This allows me to analyze whether portfolio firms adjust their ESG structure to meet the changes in their mutual fund shareholders’ announced preferences. I find that mutual funds are able to influence the governance policies of their portfolio firms. However, I find no evidence of portfolio firms adopting the announced E&S preferences of the mutual fund shareholders.
Through which channels do mutual funds convey their preferences?
I explore the channels through which mutual funds convey their governance preferences to investee firms. The most straightforward channel is the voting process itself. When applying their voting policies to their votes, mutual funds may increase (decrease) the overall support given to a proposal. The proposal will therefore be more (less) likely to win a majority which would subsequently encourage (discourage) its implementation. I find strong evidence supporting this channel.
Then, I investigate whether mutual funds play a more outspoken activism role by submitting proposals requesting the implementation of their favorite provisions. I find that the mutual funds in my sample hardly sponsor proposals. Therefore, proposal submission by mutual funds is not a credible channel through which mutual funds foster the adoption of their preferred policies. However, I also examine whether proxy voting guidelines stimulate activism by non-mutual fund shareholders. I show that proxy voting guidelines allow non-mutual fund shareholders to identify and strategically sponsor proposals that mutual funds are more likely to support.
Finally, I assess whether mutual funds use private negotiations to obtain the implementation of their preferred governance provisions. As I do not observe private settlements, I analyze cases where firms adopt the preferred governance provisions of their mutual fund shareholders in the absence of a shareholder-initiated proposal on that provision in the previous two years. Changes in governance provisions without shareholder proposals may indeed result from private settlements between management and shareholders. My findings are consistent with the existence of such private negotiations. However, I cannot disentangle the private negotiations hypothesis from the situation where firms adopt the preferred policies of their mutual fund shareholders on their own initiatives, without any behind-the-scenes negotiations.
How can mutual funds be active monitors while dedicating very few resources to their stewardship activities?
One of the elements that have fed the concerns regarding the lack of active monitoring by institutional investors is the limited amount of resources that these investors dedicate to stewardship. As an illustration, Krouse et al. (2016) report that in 2016, BlackRock, Vanguard, and State Street employed less than 50 staff members altogether in their voting and stewardship teams while covering thousands of firms. Contrasting with this anecdotal evidence, there is ample evidence that mutual funds do not all vote in the same manner, blindly following management or proxy advisors’ recommendations (e.g. Morgan et al., 2011 or Iliev and Lowry, 2014).
I examine the voting procedures of mutual fund families in my sample. In these documents, many fund families explain clearly that they use their proxy voting guidelines to outsource their voting strategy to proxy advisory firms. Such a business model explains how mutual funds can be active voters while employing very small teams dedicated to stewardship. My empirical findings support the existence of this business model and show that it allows mutual funds to obtain the implementation of their preferred governance policies. It hence demonstrates that proxy voting guidelines allow institutional investors to do governance at scale.
My results reveal that mutual fund families have developed diverging ideologies regarding ESG matters. By measuring the distance between announced and revealed preferences, they shed light on a growing concern among mutual funds’ investors and policymakers, namely that mutual funds’ public statements and policy positions reflect marketing rather than stewardship intentions. This concern appears to be well-founded for policies that support environmental and social reforms. In consequence, portfolio firms do not adopt the announced E&S preferences of their mutual fund shareholders.
My results also have important implications for the growing debate on the impact of institutional ownership, especially passive ownership, on governance. While it is often argued that passive investors do not have the tools to monitor companies, I provide evidence that proxy-voting guidelines are an important tool that allows mutual funds to do governance at scale. Furthermore, my results demonstrate that voting, a voice mechanism that passive investors use extensively, is sufficiently meaningful to foster the implementation of institutional investors’ preferred governance policies. Finally, proxy voting guidelines ensure clear communication of mutual funds’ preferences to activists. Overall, the evidence I present suggests that proxy voting guidelines are an essential device for mutual funds to perform their stewardship role effectively.
Maxime Couvert is a PhD candidate at the Swiss Finance Institute and the Ecole Polytechnique Fédérale de Lausanne.