Investors rely on proxy adviser recommendations to fulfill their fiduciary duty when voting in corporate elections. However, the proxy advisory industry is highly concentrated, with ISS and Glass Lewis controlling over 90 percent of the market,1 and the advice they offer has been criticized as overly simplistic (“one size fits all”),2 based on very little research,3 or politically slanted.4 Given how many investment funds rely on proxy advice from these two companies, many observers are concerned that votes cast in corporate elections might not adequately represent the interests of the funds’ beneficial owners.

Yet these criticisms are themselves controversial, with limited empirical evidence either way. One important question is how such inefficient practices could survive the rigors of market competition. In a new working paper, we set out to investigate this question theoretically by developing a model of the proxy advisory industry and exploring how the desire to maximize profit and the presence (or absence) of competitors affect the quality of proxy advice.

Our model assumes that proxy advisers are nonideological and free of conflicts of interest that might induce them to slant their advice, which abstracts away two potentially important causes of advice bias to isolate the effect of competition alone. The model also assumes that mutual and pension funds that purchase advice differ on how much importance they place on financial versus nonfinancial returns. For example, a socially responsible investment (SRI) fund may want recommendations that emphasize a proposal’s climate effects over its financial impact, while a traditional fund may want recommendations that focus on financial returns alone.

Proxy advisers compete through their prices and their choice of an advice policy, where an advice policy is a process that boils down detailed information into simple recommendations. Specifically, our model assumes that advisers choose the weight they place on a proposal’s financial versus nonfinancial returns when forming recommendations, and this weight is an attribute known to the market. Because proxy advisers produce only binary recommendations (support or oppose), and do not convey the complete set of information discovered by their research, their recommendations may be slanted toward nonfinancial or financial returns. Although their customers know this, without the underlying raw information they are unable to “unravel”  the slant and may rationally follow this biased advice.

We assume that shareholders have varied preferences about how important they consider financial versus social returns and also differ in the importance they place on voting in corporate elections. As a result, funds differ in how slanted they want the recommendations they receive to be and how much they are willing to pay for advice aligned with their values. An SRI fund, for instance, may place a high value on voting in a way that expresses its support for certain social policies because that is part of its marketing strategy, while a small, passive fund may not care as much how it votes.

Reasoning from simple economic principles, one might expect that the desire to maximize profit would lead advisers to produce as accurate recommendations as possible, and this incentive would become stronger as competition increases. However, our model shows that this is not always the case.

To understand the intuition behind these results, consider first why a profit-maximizing, monopolist proxy adviser might issue recommendations that do not align with the preferences of most investors. Given the distribution of demand in the market, a profit-maximizing adviser chooses a recommendation policy that appeals to customers who place the highest importance on their votes. (We assume that advisers are limited in their ability to customize advice, so they cannot narrowly tailor it to each customer.) If, as suggested above, SRI funds place more importance on voting than passive funds do, then our model implies that the proxy adviser’s recommendations would cater to SRI funds because they are willing to pay the most for aligned recommendations. Such a slant in advice would be profit-maximizing even if SRI funds do not comprise a majority of investors.

Selling slanted advice is desirable if it accurately reflects the underlying preferences of customers because the recommendations would help the customers align their votes with their values. But if some funds purchase advice merely to establish a regulatory safe harbor for fiduciary responsibility, the slant induced by funds that are willing to pay a lot for aligned advice spills over to the rest of the market – even if it does not reflect their preferences. If the advice market were served by a single firm, the biases would essentially drive corporate elections. This seems a real possibility in the U.S. where the market is close to a duopoly, with ISS alone serving about two-thirds of the market.

Having shown that the desire to maximize profit induces a monopolist adviser to produce slanted advice, we then ask whether competition can mitigate this bias. We find that, if investors were solely concerned with the quality of proxy advice, competition would enhance representation (in the sense of allowing votes to accurately reflect fund preferences) by driving out low-quality advisers. However, if misrepresentation in proxy advice stems less from quality issues than from a slant in the nature of advice, competition can hurt as well as help.

The key insight is that competition helps by lowering prices, making advice more accessible to voters, but it leads advisers to differentiate the slant of their advice products. Unlike in the famous median voter (Hotelling) model, competition does not lead firms to converge on “centrist” advice, but rather to move away from the center and offer differentiated advice. This product differentiation occurs because, if firms were to offer identical advice, their price competition would drive profits to zero; they can make more money by adopting contrasting positions and exploiting their local monopoly power. This effect of competition resembles the situation in the news media, where competition drives coverage toward ideological polarization.

Having a choice between multiple “slants” of advice can improve the proxy voting market if it allows investors to find advice better aligned with their preferences. But it can hurt if investors with centrist preferences are unable to find an adviser that is aligned with their values. The net effect depends on which of these two forces is stronger.

One policy implication of our model is that regulatory rules that require funds to vote and prohibit abstention may reduce the representativeness of shareholder democracy. Our model suggests that if uninformed funds, which would otherwise abstain, are compelled to vote, one of these otherwise uninformed funds may end up as the decisive (median) voter. Because that fund’s vote is essentially delegated to a proxy adviser, the election outcome would be determined by the advice policy of the adviser, which, as noted, may be biased toward the preferences of SRI funds that place greater emphasis on voting.

ENDNOTES

1. Chong Shu, “The proxy advisory industry: Influencing and being influenced,” Journal of Financial Economics, April 2024: https://www.sciencedirect.com/science/article/abs/pii/S0304405X24000333.

2. Doron Levit and Anton Tsoy, “A Theory of One-Size-Fits-All Recommendations,” American Economic Journal: Microeconomics, November 2022: https://www.aeaweb.org/articles?id=10.1257/mic.20200138.

3. Bernard Sharfman, “The Risks and Rewards of Shareholder Voting,” SMU Law Review, 2020: https://scholar.smu.edu/smulr/vol73/iss4/5/.

4. Patrick Bolton, Tao Li, Enrichetta Ravia, Howard Rosenthal, “Investor ideology,” Journal of Financial Economics, August 2020: https://www.sciencedirect.com/science/article/abs/pii/S0304405X20300635

This post comes to us from professors Odilon Camara and John Matsusaka at the University of Southern California and Chong Shu at the University of Utah. It is based on their recent working paper, “Shareholder Democracy and the Market for Voting Advice,” available here.