Creating Controversy in Proxy Voting Advice

Proxy advisory firms have emerged as major players in corporate governance by helping to address the public goods aspects of information production in corporate governance. These firms provide both a) recommendations on how to cast proxy votes and b) research reports that contain the full rationale for their recommendations, including detailed information on the operating firm’s governance. While proxy advisers’ research reports are only available to their subscribing shareholders, their recommendations are often made public in the media. Through both these public recommendations and private research reports, proxy advisers, such as ISS, have a substantial impact on voting outcomes.

Given the strong influence of proxy advisers, the quality and information content of their reports and recommendations are an important topic of discussion among market participants and policymakers. Motivated by these discussions, our paper (Malenko, Malenko and Spatt, 2021) studies the information design problem of a proxy adviser that aims to maximize its profit from information sales and addresses the following questions.[1] What criteria should the proxy adviser use to formulate its research reports and recommendations?  In effect, what are the objectives of proxy advisers and how “does the lack of a direct pecuniary interest in the effects of their recommendations” (SEC’s 2010 Concept Release on the United States proxy system) affect the recommendations?

While the proxy adviser designs the recommendations and research reports to maximize its profit from information sales, it does not maximize the value of either the asset managers or the operating companies, highlighting a fundamental conflict of interest. In fact, value maximization for the downstream entities (asset managers and operating companies) would suggest that the public recommendations are perfectly aligned with the reports and reveal all the information the proxy adviser has—but then no investors would need to purchase the reports and the proxy adviser would have zero revenue. (Arrow (1962) highlights the underlying challenge of getting paid for information.[2]) This raises the question of how the design of recommendations would influence the fees that can be obtained from the subscribers. What would lead to large fees and profit by the proxy adviser?

The value of the information purchased by the subscriber reflects the extent to which that information helps the shareholder cast a more informed vote (beyond receipt of the recommendation) and improves the outcome through his vote. In other words, it is worth paying the costs of the subscription if the resulting anticipated improvement in the value of the assets at least covers the cost of bringing about that improvement.

Given this, our main result shows that even if all shareholders are unbiased and aligned at maximizing the value of their shares, the profit-maximizing proxy adviser often has incentives to produce public recommendations that are biased against the a priori more likely alternative. By recommending for the unexpected alternative (which is typically the alternative that is opposed by management) too often, the proxy adviser creates controversy around the proposal, which increases the probability that the vote will be close. Since information purchased by a shareholder is valuable to the extent to which it helps the shareholder improve the outcome through his vote, the value of the information is higher if the shareholder’s individual vote is more likely to be pivotal for the outcome. Thus, recommendations that create controversy and make the vote outcome close increase each shareholder’s likelihood of being pivotal and his willingness to subscribe to the proxy adviser’s research, increasing the adviser’s profits.

At the same time, we show that the proxy adviser has incentives to produce informative and unbiased research reports for its subscribers. By making the research report accurate and unbiased, the adviser can obtain the maximum revenue from the fees it charges to the subscribing shareholders. This is a sense in which the interests of the proxy adviser are aligned with those of the shareholders to whom it sells subscriptions. The combination of public recommendations and research reports available through the subscriptions is therefore central to the ability of the proxy adviser to extract revenue from the advising process. In our paper, proxy advisers serve the needs of their clients (subscribers), while limiting how much information they release through recommendations and biasing them in order to obtain maximum revenue from selling the subscriptions.

Arguably, the value proposition for the proxy advisers is the reports they generate rather than their recommendations. Consistent with this, many mutual funds that subscribe to the proxy advisory reports and that also invest in stewardship suggest that their primary interest in the feedback from the proxy advisers is in the detailed data and reports that they generate rather than the specific recommendations.  For example, according to the survey of institutional investors by Bew and Fields (2012), “virtually unanimously, research participants highlighted the value they derive from … proxy advisers … digest[ing] and normaliz[ing] the vast quantities of data present in proxy statements in a short period of time,” but the “value of … voting recommendations is distinctly secondary.”[3]

Overall, our conclusion is that, while the reports are aligned with the needs of subscribing shareholders, the public recommendations reflect bias and an incentive to promote controversy. Thus, our analysis predicts that deviations of shareholder votes from proxy advisers’ recommendations would take a specific form: Compared with the adviser, shareholders would be more predisposed towards the a priori more likely alternative, essentially counteracting the bias in recommendations. In contrast, shareholders would rubber stamp recommendations that are consistent with the priors. This prediction is broadly consistent with the observed empirical evidence on shareholders’ voting behavior. Management proposals that receive a positive recommendation typically pass with very high voting support, i.e., are rubberstamped, whereas those that receive a negative recommendation often generate a lot of disagreement among shareholders (see, e.g., Malenko and Shen (2016) for say-on-pay votes and Ertimur et al. (2018) for director elections).[4] Both patterns are consistent with our predictions, assuming that management proposals are a priori likely to be approved.

One way to explore our conclusions is to compare the recommendations of the proxy advisers with the votes cast by large asset managers, whose interests are potentially more directly aligned with value maximization. Empirical evidence highlights that proxy advisers often make recommendations that are more anti-management than the votes of the major index funds (e.g., Bolton et al. (2020) and Bubb and Catan (2019) for regular proposals and Brav et al. (2020) for voting on proxy fights).[5] While index funds are often criticized because their votes do not align with the proxy advisory recommendations, our analysis suggests that proxy advisory recommendations may not be the right benchmark. Shareholders who support management and deviate from the negative recommendations could be simply correcting the bias in these recommendations rather than being passive voters.

In fact, as we discuss in the paper, the one-size-fits-all approach, which proxy advisers are frequently criticized for, could be one way for them to implement recommendations that create controversy. By giving recommendations according to prescriptive guidelines and without taking into account firm-specific circumstances, the proxy adviser can increase the probability of a close vote, thereby increasing shareholders’ incentives to subscribe to its research reports. In contrast, asset managers that have large stewardship teams and make substantial investments in due diligence could be accounting for firm-specific factors and often coming to different conclusions from proxy advisers on these issues (e.g., Iliev and Lowry, 2015).[6] Therefore, the votes of large asset managers, by virtue of their substantial investments in stewardship, could potentially reflect a more suitable benchmark than either the votes of small asset managers or the proxy advisers’ recommendations.


[1] Malenko, Andrey, Malenko, Nadya and Chester S. Spatt, 2021, Creating controversy in proxy voting advice, Working Paper, University of Michigan at

[2] Arrow, Kenneth J. “Economic welfare and the allocation of resource for inventions, in the rate and direction of inventive activity: economic and social factors.” RR Nelson (red.), Princeton University, Princeton (1962).

[3] Bew, Robyn, and Richard Fields, 2012, Voting decisions at US mutual funds: How investors really use proxy advisers. IRRC Institute.

[4] Malenko, Nadya, and Yao Shen, 2016, The role of proxy advisory firms: Evidence from a regression-discontinuity design, Review of Financial Studies 29, 3394-3427; and Ertimur, Yonca, Fabrizio Ferri, and David Oesch, 2018, Understanding uncontested director elections, Management Science 64, 3400-3420.

[5] Bolton, Patrick, Tao Li, Enrichetta Ravina, and Howard Rosenthal, 2020, Investor ideology, Journal of Financial Economics 137, 320-352; Bubb, Ryan, and Emiliano Catan, 2019, The party structure of mutual funds, Working Paper, New York University; and Brav, Alon, Wei Jiang, Tao Li, and James Pinnington, 2020, Picking friends before picking (proxy) fights: How mutual fund voting shapes proxy contests, ECGI Working Paper No. 601/2019.

[6] Iliev, Peter, and Michelle Lowry, 2015, Are mutual funds active voters? Review of Financial Studies 28, 446-485.

This post comes to us from professors Andrey Malenko and Nadya Malenko at the University of Michigan’s Stephen M. Ross School of Business and Chester S. Spatt at Carnegie Mellon University’s David A. Tepper School of Business. It is based on their recent paper, “Creating Controversy in Proxy Voting Advice,” available here.

Leave a Reply

Your email address will not be published. Required fields are marked *