Climate risk is one of the greatest challenges facing the world today and investors are taking notice. But a full understanding of how climate risks get implemented into firm policy remains elusive. In our recent paper, we study the propagation of climate disasters through common ownership networks as a mechanism through which investors may impact corporate ESG policies.
Our study is motivated by the bounty of evidence showing investors not only care about climate risks but also actively engage with firms to affect corporate policy. On the investor side, investors can influence their portfolio firms through their ownership. For example, a June 2022 Wall Street Journal article discusses how Blackrock employs 70 analysts on its investment stewardship team to help represent “$4.6 trillion worth of shares…” across the public stock markets. On the firm side, when a sufficiently large mass of ownership coalesces around a policy, firms’ actions steer toward a new direction. So, to the question of how climate risk impacts firms, we conjecture when climate disasters hit parts of investors’ portfolios, investors change how they engage with their portfolio firms, which propagates to indirectly affect even firms not physically exposed to climate disasters.
To test this hypothesis, we collect extensive data on large climate disasters and institutional investors’ holdings from 2003 to 2019. We then measure investors’ portfolio exposure to climate disasters, accounting for disaster-hit firms’ geographic footprints and investors’ attention to these hit firms via portfolio weights. Importantly, we benchmark investors’ portfolio exposure to climate disasters by overweighting geographic areas that experienced worse disasters than that of the 1990s. Overall, we can intuitively interpret our main measure as the percent of investors’ portfolio value exposed to a worse climate event year than would have been expected from the 1990s.
We next investigate our conjecture by testing if investors’ portfolio exposure to climate disasters predicts how they engage and eventually affect firms’ policies. We find that investors’ exposure to climate disasters via one portfolio firm impacts their climate-related voting behavior and, in turn, the environmental focus of other (non-disaster hit) firms in their portfolios.
Our findings can be analyzed into short- and long-run effects. The first immediate effect we observe is that investors with portfolios hit by disasters in the previous year vote more for climate-related shareholder proposals on the other firms they own. This is in comparison to their own voting behavior at different times and the voting patterns of other investors on the same proposals. A one standard deviation increase in the investor’s past year portfolio exposure is associated with a 3.49 percentage point increase in support for climate-related proposals. Given that climate and ESG-related proposals normally receive very low support from institutional investors – the approval rate has remained below 30% in the past 10 years – our finding indicates that portfolio climate exposure affects institutional investors and changes their voting decisions. As evidence of climate risk salience, this effect strengthens in the six months before most firms tend to vote in their Q2 shareholder meetings, which coincides precisely with the Q3 and Q4 quarters of the prior year when climate disasters tend to hit. These affected climate proposals also tend to target “brown” firms that emit large amounts of CO2, like Exxon Mobil, and relate to policies like the promotion of greenhouse gas (GHG) targets or goals. Surprisingly, we do not find similar effects for non-climate-related environmental proposals or social-oriented proposals. Collectively, our voting evidence supports the contention that investors, by responding with support for more aggressive climate policy targeting brown firms, become more concerned about future climate risks when they experience climate disasters in their portfolio.
In further validation, we find the voting effect of climate disaster exposure is elevated in periods of high climate attention and for high CO2 emitting firms. We find little evidence, however, that the effect varies depending on investor type. Thus, even the most impactful votes from the largest institutions, such as Blackrock or Vanguard, are affected.
While the immediate impact on voting is a particularly powerful set of evidence, it remains the (revealed) tip of the iceberg for all the direct or background avenues of investor engagement in firms. To dig into these routes, we move from the tests at the investor level to the firm level by considering whether firms’ investor base aggregate disaster exposure affects corporate outcomes. Specifically, we construct firm-level exposure via weighting investor-level portfolio exposures by the investor’s percentage ownership of shares in a firm. To show that this is a propagation effect across ownership networks to portfolio-connected firms, we focus on the sample of firms that are not contemporaneously hit by disasters. This is because even if firms do not experience a disaster, as long as one of its investors is exposed, then our firm-level measure would be non-zero.
Armed with this spillover measure, we begin by examining the more immediate effects on climate sentiment in conference call discussions of the climate and find that climate change sentiment is more pessimistic in calls that follow their investors’ climate disaster exposure. A one standard deviation increase is associated with a 0.39 percentage point decrease in conference call climate change sentiment. When we break down this estimate by positive or negative sentiment, we find that 67.5% of the overall effect is due to a more negative discussion of climate change.
We then study the impact of investors’ climate disaster exposure on long-run environmental policy. Our findings indicate that firms adopt specific governance mechanisms, such as linking their executive pay policies to GHG emission reductions and providing their boards with explicit responsibility for climate change in the two-year period following climate shocks to their institutional investors. Concurrently, firm-level GHG emissions and energy usage cumulatively decline, suggesting that changes in these governance mechanisms incentivize firms to shrink their carbon footprints.
Overall, our study is particularly important in the current context of increasing awareness and action against climate change. Our work contributes to the growing literature on climate finance and the role institutional investors may play in addressing the challenges posed by climate change. In particular, our conclusions speak to the debate on whether institutional investors should exit (divestment and boycott) or voice (engagement) when aiming to make impactful, sustainable investments. We show that, when exposed to climate disaster events in their portfolios, institutional shareholders improve corporate sustainability performance through engagement and voting.
Matthew Gustafson is a Professor of Finance at the Smeal College of Business at Penn State.
Ai He is an Assistant Professor of Finance at the Darla Moore School of Business at the University of South Carolina.
Ugur Lel is an Associate Professor of Finance and the Nalley Distinguished Chair at the University of Georgia.
Zhongling Qin is an Assistant Professor of Finance at the Raymond J. Harbert College of Business at Auburn University.
This post was adapted from their paper, “Propagation of Climate Disasters Through Ownership Networks,” available on SSRN.
