Chong Shu
Publications
1. "The Proxy Advisory Industry: Influencing and Being Influenced", Journal of Financial Economics, April 2024 (Editor's Choice)
 |    |    |    |  

This paper develops two new methods to infer a mutual fund’s proxy advisors from SEC filings. It then applies these methods to characterize features of the proxy advice industry from 2007 to 2021: (i) As of 2021, ISS and Glass Lewis collectively control approximately 90 percent of the market. During this period, the market share of ISS remains stable, while that of Glass Lewis has increased. (ii) When a proxy advisor issues a recommendation opposing management, its customers are approximately 20 percentage points more likely to also oppose management compared to other investors. (iii) Funds that subscribe to both proxy advisors tend to vote more similarly to the recommendations of the advisor whose voting platform they use. (iv) Proxy advisors often change their advisory stance when investors disagree with their previous advice. I offer suggestive evidence that this adaptation reflects both learning from informed investors and a desire by proxy advisors to align with the preferences of their customers.
2. "Endogenous Risk-Exposure and Systemic Instability", Management Science, Accetped
 |    |    |  

Most research on systemic stability assumes an economy where banks are subject to exogenous shocks. However, in practice, banks choose their exposure to risk. I show that there exists a network risk-taking externality: the risk exposure choices made by connected banks are strategically complementary. Banks within financial networks, especially densely connected ones, become endogenously exposed to excessive risks. The theory offers several novel perspectives on policy debates. For instance, it suggests that limiting government bailouts to interbank exposures can effectively reduce endogenous systemic risk.

Working Paper
1. "Divestment and Engagement: The Effect of Green Investors on Corporate Carbon Emissions" with Matthew E. Kahn and John G. Matsusaka, 2024
 |    |    |    |    |    |  

This paper investigates if green investors can influence corporate greenhouse gas emissions through capital markets, and if so, whether they have a bigger effect by divesting their stock and limiting polluters’ access to capital, or by acquiring polluters’ stock and engaging with management. We focus on public pension funds, classifying them as green or nongreen based on which political party controlled the fund. To isolate the causal effects of green ownership, we use exogenous variation caused by state-level politics that shifted control of the funds, and portfolio rebalancing in response to returns on non-equity investment. Our main finding is that companies reduced their greenhouse gas emissions when stock ownership by green funds increased and did not alter their emissions when ownership by nongreen funds changed. Other evidence based on activist funds, voting, and shareholder proposals suggests that ownership mattered because of active engagement by green investors and not simply because management adapted proactively to changing shareholder preferences. We do not find that companies with green investors were more likely to sell off their high-emission facilities (greenwashing). Overall, our findings suggest that (a) corporate managers respond to the environmental preferences of their investors; (b) divestment of polluting companies may lead to greater emissions; and (c) private markets may be able to address environmental challenges independent of government regulation.
2. "The Politics of Academic Research" with Matthew C. Ringgenberg and Ingrid M. Werner, 2023
 |    |  

We develop a novel measure of political slant in research to examine whether political ideology influences the content and use of academic research. Our measure examines the frequency of citations from think tanks with different political ideologies and allows us to examine both the supply and demand for research. We find that research in Economics and Political Science displays a liberal slant, while Finance and Accounting research exhibits a conservative slant, and these differences cannot be accounted for by variations in research topics. We also find that the ideological slant of researchers is positively correlated with that of their Ph.D. institution and research conducted outside universities appears to cater more to the political party of the current President. Finally, political donations data confirms that the ideological slant we measure based on think tank citations aligns with the political values of researchers. Our findings have important implications for the structure of research funding.
3. "Does Proxy Advice Allow Funds to Cast Informed Votes?" with John G. Matsusaka, 2022
 |  

This paper estimates to what extent proxy advice allows funds to vote as if they were informed. A fund’s vote is classified as “informed“ if the fund accessed the company’s proxy statement from the SEC’s Edgar website prior to voting. A fund’s proxy advisor, if any, is identified from the format of its Form N-PX filing. Our main finding, for the period 2004-2017, is that proxy advice did not result in funds voting as if they were informed – more often than not it pushed them in the opposite direction – and this distorting effect was particularly noticeable for ISS. The finding is robust to several strategies designed to control for endogeneity of acquiring information and seeking proxy advice, including fixed effects and instrumental variables. We also show that advice distorted votes toward policies favored by socially responsible investment (SRI) funds, and provide suggestive evidence consistent with the idea that proxy advisors slanted their recommendations toward the preferences of SRI funds because of pressure from activists.
4. "Shareholder Democracy and the Market for Voting Advice" with Odilon Camara and John G. Matsusaka, 2024  
 |  

In corporate elections, most votes are cast based on recommendations provided by for-profit proxy advisory companies. We develop a model of the voting advice market to explore how competition and demand for advice shape the slant of the advice offered. In the model, advisory firms compete through prices and their advice policies. Shareholders have heterogeneous goals, differing in the weight they place on financial returns versus nonfinancial ("social") returns, such as reductions in carbon emissions. We assume that investors vote for expressive reasons and may differ in how much they care about voting correctly. In equilibrium, advising firms tailor their advice to reflect the preference of their average investor, which can result in election outcomes being skewed away from those that would prevail if investors had full information. We derive conditions under which advisory firms skew their advice and therefore the election outcome in favor of a minority of investors who have a strong preference for nonfinancial returns. We also study how increasing competition affects equilibrium outcomes.