Whose interests do public pension funds serve? On the one hand, they have a fiduciary duty to maximize value for the pension fund participants whose retirement savings they invest. On the other, they can use their considerable ownership stakes to pressure management at portfolio firms to change governance practices or become more or less favorable to environmental, social, and governance (ESG) issues. The tension between these objectives, and the best way for public pension funds to resolve them, has been the object of considerable scholarly debate. In a recent paper, we use new data to shed light on public pension fund involvement in corporate governance.
We exploit the first set of filings pursuant to a recent Securities and Exchange Commission (SEC) rule change obliging public pension funds to disclose how they voted on say-on-pay (SOP) proposals on executive compensation. Executive compensation is one of the most controversial topics in corporate governance, and “excessive” compensation has drawn the ire of public pension funds. As You Sow, a nonprofit group focusing on corporate social responsibility, reported that public pension funds voted against 100 “overpaid” chief executive officers (CEOs). In one high-profile example, the California pension fund CalPERS opposed Tesla CEO Elon Musk’s pay package, calling it “excessive” and “at odds with CalPERS’ longstanding views on executive pay.”
Voting has traditionally been seen as the primary mechanism for shareholders to hold management accountable, but intermediaries such as public pension funds are obligated to exercise their voting rights to maximize value for their clients—the pension plan participants. However, we document that politics may influence funds’ ability to maximize value. Democratic-run public pension funds are significantly more likely to vote against management on SOP proposals. Their opposition does not seem to be dictated by the financial performance of portfolio companies. Instead, it seems to be influenced by proxy adviser recommendations and the size of the CEO pay package in voting against management. By contrast, the data suggest that pension funds from Florida and Texas—the two Republican states most vocal about issues in corporate governance—base their SOP votes on the financial performance of portfolio firms and are less concerned about proxy adviser recommendations or the size of the CEO pay package. More broadly, states that have enacted anti-ESG legislation (which mostly lean Republican) seem to ground their SOP votes in portfolio firms’ total shareholder return. Our empirical results persist when we restrict the analysis to states with Democratic or Republican control of the governorship and two houses of the legislature (i.e., states where the direction of political influence on pension funds is clearest) and when we account for potential “home bias” in pension-fund decision-making.
We also analyze the new disclosures’ data on share lending by public pension funds. We find that funds with worse funding ratios are markedly more likely to lend out their shares. This is unsurprising, because these funds presumably need to raise money through activities such as share lending. However, high levels of share lending may dilute the influence of public pension funds and reduce the scope of their activism in corporate governance. We do not find any noticeable political valence to the prevalence of share lending.
Our paper is primarily concerned with providing a novel descriptive account of the interplay between political and financial factors in explaining how public pension funds vote on SOP proposals. We do not claim to ascertain which pension fund is “better” at deciding how to vote on these issues, because there is no answer to what the correct vote on any given proposal will be. However, we think our finding that pension funds may be influenced by political factors, rather than shareholder return, when voting on executive pay implicates legal issues. While the details of the legal regime governing public pension funds are complex, state and local pension funds generally have a strict fiduciary duty to maximize the economic interests of the pension plan and not consider non-economic factors.
One may think that funds’ potential reliance on political views on executive compensation is incompatible with their fiduciary duties. Our results show that Democratic-leaning funds appear to prioritize proxy adviser recommendations and the absolute size of CEO pay packages rather than shareholder return. The lack of correlation between Democratic-leaning funds’ voting choices and shareholder return, when compared with the seemingly return-driven decisions of Republican-leaning funds, raises questions about Democratic-run funds’ use of non-economic factors in their decision-making. We do not claim to prove that any funds have violated legal duties. However, we believe that the absence of any correlation between public pension fund voting and shareholder return at least raises questions related to whether the fund abides by the exclusive purpose rule.
Our finding on the partisan asymmetry in responsiveness to proxy adviser recommendations is consistent with the political tenor of recent controversies about the proxy-adviser industry. On September 19, 2024, the Republican-controlled House of Representatives passed the Protecting Americans’ Retirement Savings from Politics Act, which had several provisions specifically attacking proxy advisers. The bill requires proxy advisers to register with the SEC and make disclosures, which would significantly increase compliance costs for ISS and Glass Lewis. Moreover, the legislation asked the SEC to study “whether regulations and financial incentives have created and protected the outsized influence of proxy advisors or a duopoly in proxy advice, and if so, what are the benefits and costs of that outsized influence or duopoly.”
While the bill has not yet passed the Senate or been signed into law, the Republican victory in the 2024 presidential election has increased the likelihood that proxy advisers will face heightened legal scrutiny with Republicans controlling both the executive and legislative branches of the federal government, as well as administrative agencies such as the SEC. Our results, showing that proxy advisers appear to wield significant influence over public pension funds controlled by Democrats but not those run by Republicans, reinforce the notion that there is a partisan divide in receptivity to these advisers’ recommendations. These findings should be seen in the light of these debates about whether proxy advisory recommendations stifle pension funds’ ability to maximize value for their pension plan participants.
Finally, our analysis of the new share-lending data shows that pension funds seem to loan shares because of deeper structural issues with the adequacy of funding for the associated pension plans. While some policymakers were worried that these disclosures would “shame” plans into reducing share loaning to seem more responsible, we believe the long-running and severe nature of state and local pension deficits reduces the prospect of such shaming. We do not have a strong view on the normative desirability of share lending by pension funds and the associated empty voting by the borrowers of these shares.
However, one interpretation of our results on share lending could be that the political conflicts influencing public pension-fund judgment increase the desirability of these funds loaning shares to entities like hedge funds. If public pension funds have political rather than value-enhancing motives, voting outcomes could be improved if these conflicted institutions loaned their shares to unconflicted actors such as hedge funds, which are relatively focused on shareholder return. While our data do not allow us to analyze the costs or benefits of empty voting, we believe that our results could be read as consistent with this optimistic interpretation of share lending and empty voting.
The regulatory mandate we study does not compel public pension funds to disclose votes on issues other than SOP proposals. Therefore, we cannot analyze whether these funds display partisan divides on other, potentially more divisive environmental and social issues such as climate change or board diversity. However, our findings establish important facts about the political economy of public pension-fund voting in corporate governance. The partisan patterns in our data raise questions about whether some public pension funds rely on non-pecuniary factors in making voting decisions. The findings also underline the increasing politicization of market participants’ views of the proxy advisory industry and the structural public finance origins of share-lending by pension funds.
This post comes to us professors Dhruv Aggarwal at Northwestern University’s Pritzker School of Law, Lubomir Litov at the University of Oklahoma, and Shivaram Rajgopal at Columbia Business School. It is based on their recent article, “Value over Values: Evidence from State Pension Funds,” available here.