By Simon Glossner, University of Virginia Darden School of Business
Corporations increasingly integrate environmental, social, and governance (ESG) issues into their business practices and signal this by committing to sustainability initiatives, such as the UN Global Compact. This development has spurred a debate on whether companies can become more profitable by creating societal value (e.g., Edmans 2020), but there is relatively little discussion of companies that have been involved repeatedly in ESG-related incidents.
For example, before its Deepwater Horizon oil spill in 2010, BP had a long history of environmental and safety incidents. Between 2007 and 2009, these included a Texas City refinery explosion, pipeline leaks in Alaska, attempts to manipulate the propane market, and frequent safety violations (e.g., Cherry and Sneirson 2011). This illustrates that firms may fall short of their sustainability commitments and save money at the expense of their stakeholders, which in the extreme may lead to sizable losses and disastrous societal outcomes.
In my paper, ESG Incidents and Shareholder Value, I use a novel dataset of negative incident news to study poor ESG practices. I examine whether a firm’s incident history can predict future ESG incidents; how high incident rates affect firm value over time; whether stock markets properly price incident-based ESG information and if not, what prevents investors from doing so. The answers to these questions help us to understand why some firms have higher incident rates and are therefore important, not only for corporate managers but also for investors and policy makers that attempt to curb negative externalities on stakeholders.
From a theoretical perspective, it is unclear whether poor ESG practices increase or decrease shareholder value. On the one hand, companies may maximise shareholder value by spending less on stakeholders’ needs (e.g. employees or the community) or on protecting the environment. High ESG incident rates would then imply a strong shareholder focus and high cost-efficiency. On the other hand, poor ESG practices may be the result of managerial short-termism (Bénabou and Tirole 2010). While ignoring ESG policies and practices can grow profits over the short term, it may lower long-term value by increasing the chances of new ESG incidents, reputational damage, and less social capital and trust.
My paper measures poor ESG practices based on a firm’s ESG incident history, using data from RepRisk, which collected over 80,000 news reports about incidents involving publicly listed US firms between 2007 and 2017. I study the value implications of a firm’s past ESG incident rate, which reflects the frequency and severity of past ESG incidents (e.g., environmental pollution, poor employment conditions, or anti-competitive practices).
Incident-based ESG measures directly capture poor practices, through ESG-related business risks that materialise, and stakeholder criticism. In contrast, conventional ratings (e.g., MSCI IVA or ASSET4) aggregate hundreds of ESG criteria into one company score using very different approaches and consequently, they often disagree with each other, which raises concerns about their validity (e.g. Chatterji et al. 2016). As such, incident measures may be a better predictor of future incidents than ESG ratings.
Studying the value implications of a firm’s incident history, I find five key results:
Past ESG incident rates are associated with higher disagreement among conventional ESG ratings. This suggests that ESG rating providers disagree more on how sustainable a company is when it has had more incidents in the past. Put differently, not every rating provider gives a company with a history of incidents a lower ESG rating. This disagreement makes it more difficult for investors to identify and value companies with poor ESG practices.
Past ESG incident rates predict more future incidents and lower operating profits. This suggests that poor ESG practices negatively affect firm performance through a higher probability of new incidents, and consistent with this, stock markets react negatively to news of a company’s ESG incident.
Past ESG incident rates predict negative risk-adjusted stock returns. A value-weighted US portfolio with high ESG incident rates underperforms the stock markets by about 3.5% per year, while a European portfolio underperforms by 2.5% per year. These results account for differences in risks and are robust to industry controls, outliers, and alternative model specifications.
Past ESG incident rates predict negative analyst forecast errors and lower stock returns when companies announce their earnings or have new ESG incidents. This suggests that market participants overestimate the earnings of firms that have high ESG incident rates and underestimate the probability that they will have new incidents. Taken together, this highlights that stock markets are surprised by the negative value implications of poor ESG practices, which leads to predictable negative stock returns at firms with high ESG incident rates.
Past ESG incident rates predict more pronounced negative stock returns in firms that have more short-term oriented investors, are more difficult to value, or receive less investor attention. Investors – especially those with short horizons – do not pay enough attention to a firm’s ESG-incident history. Additional tests show that short-term investors cannot predict when a company will have more incidents. These findings are consistent with behavioural finance theory arguing that investors absorb salient, easily processable information better than non-salient information. Investors may find it difficult to value poor ESG practices due to rating disagreements, the need to separate material from immaterial ESG information, the long-term nature of ESG issues, a lack of standardised ESG reporting, and potential greenwashing activities by corporations.
Corporate and investor implications
Overall, my findings suggest that poor ESG practices negatively affect long-term value, which is not fully reflected in stock prices. Short-term stock market participants do not pay enough attention to a firm’s ESG practices, which results in overvalued stocks and overly optimistic earnings forecasts for firms with high ESG incident rates. This mispricing eventually leads to predictable negative stock returns once these companies present lower earnings and more ESG incidents than anticipated.
For investors, historic incident-based ESG information is clearly material and warrants increased attention in investment decisions. Furthermore, excluding firms with high ESG incident rates from a portfolio may result in better investment performance if a significant number of investors continue to neglect incident-based ESG information.
This blog is written by academic guest contributors. Our goal is to contribute to the broader debate around topical issues and to help showcase research in support of our signatories and the wider community.
Please note that although you can expect to find some posts here that broadly accord with the PRI’s official views, the blog authors write in their individual capacity and there is no “house view”. Nor do the views and opinions expressed on this blog constitute financial or other professional advice.
If you have any questions, please contact us at [email protected]
 One caveat remains – there are also non-causal explanations for my findings. The correlation between poor ESG practices and future stock performance could be driven by an unobserved third variable. Alternatively, companies that predict lower future stock performance may reduce spending on ESG today. Although I conduct several empirical tests to make my results as robust as possible against these non-causal explanations, I cannot rule them out completely.