ESG scores, scandal probability, and event returns

Wenya Sun, Yichen Luo, Siu Ming Yiu, Luping Yu, Wenzhi DingFinancial 2024 · DOI

TL;DR

High ESG scores act as a protective shield by reducing the likelihood of corporate scandals, but they become a double-edged sword if a scandal actually occurs. This paper demonstrates that "highly rated" firms suffer a significantly harsher "fall from grace" in the stock market compared to their low-rated peers, creating a complex optimization problem for firm managers.

What Problem Does This Paper Address?

The financial world is currently obsessed with Environmental, Social, and Governance (ESG) metrics, yet empirical evidence on their actual impact remains frustratingly mixed. Some studies suggest ESG leads to outperformance, while others find it has a muted or even negative effect on returns.

The authors tackle a specific gap: the informativeness of these scores. If ESG ratings are truly informative, they should predict future risks—specifically, the probability of a firm being embroiled in a scandal. Furthermore, the paper explores the "reputation trap": does having a high ESG score make the market more or less forgiving when things go wrong? By bridging theoretical modeling with large-scale empirical data, the researchers aim to explain why different firms make such heterogeneous decisions regarding their ESG investments.

Methods

The researchers employed a two-pronged approach:

  1. Theoretical Modeling: They developed a three-period model where investors trade based on noisy ESG signals. The model predicts that while high ESG scores lower the probability of a scandal, they also increase the "valuation shock" if a scandal is revealed, as the market must aggressively correct its over-optimistic expectations.
  2. Empirical Analysis:
    • Scandal Probability: Used a panel logistic Poisson regression with firm and year fixed effects to see if ESG scores could predict future scandals.
  3. Event Study: Conducted an Ordinary Least Squares (OLS) regression to analyze Cumulative Abnormal Returns (CAR) around the dates of ESG scandals.
  4. Aggregation Techniques: Because ESG raters often disagree, the authors used Principal Component Analysis (PCA) to extract a "consensus" signal from four major rating agencies, alongside simple averaging.

Datasets

The study utilized a robust combination of financial and ESG-specific datasets:

  • ESG Ratings: Data from four major providers: KLD (MSCI), ASSET4 (Refinitiv), Sustainalytics (Morningstar), and S&P Global.
  • Scandal Data: RepRisk, which monitors over 100,000 public sources to identify ESG-related incidents (2007–2020).
  • Financial Data: CRSP for stock returns and Compustat for firm fundamentals (size, leverage, profitability, etc.).
  • Sample Size: The probability analysis included 127,478 firm-month observations, while the event return analysis covered 40,045 scandal events.

Key Insights and Ideas

  • The "Fall from Grace" Effect: The most striking finding is that the market penalizes "good" firms more than "bad" ones. When a high-ESG firm hits a scandal, the stock price drop is more severe because the scandal provides a stronger signal that the firm's reputation was overestimated.
  • ESG as a Risk Predictor: Despite the "noise" and disagreement between rating agencies, ESG scores are genuinely informative. They successfully predict a lower probability of future negative news.
  • The Cost of "Being Good": The authors model ESG as "reputation capital." Improving this capital isn't free—it involves adjustment costs.
  • Two Equilibria: The paper explains why some firms stay "brown" (low ESG). For these firms, the cost of improving their ESG score outweighs the potential savings from reduced scandal risk. Conversely, "green" firms find it optimal to maintain high standards because their "fall from grace" would be too costly to bear.
  • The Disagreement Red Flag: When rating agencies disagree significantly on a firm's ESG performance, the market reaction to a scandal is even more negative. Disagreement acts as a proxy for uncertainty.

Key Results

  • Scandal Reduction: A one-standard-deviation increase in ESG scores corresponds to a 0.1% to 1.1% decrease in the probability of a scandal.
  • Market Penalty: On average, an ESG scandal leads to a cumulative abnormal return of -0.29%.
  • The High-ESG Penalty: A one-standard-deviation increase in a firm's ESG score is associated with an additional 7 to 20 basis point decrease in stock returns when a scandal occurs.
  • Disagreement Impact: A one-standard-deviation increase in ESG rater disagreement was associated with a further 22–26 basis point drop in returns during a scandal.

Limitations and Future Work

The authors acknowledge that their findings are primarily descriptive rather than strictly causal. While they use fixed effects to control for many variables, the inherent endogeneity of ESG decisions remains a challenge.

Future Directions: * Causal Exploration: Using more rigorous identification strategies to pin down the exact causal mechanisms. * Capital Structure: Incorporating how a firm's debt-to-equity ratio influences its ESG investment decisions. * Global Preferences: Using datasets like the World Values Survey to see how different national "ESG preferences" change the optimal investment level for firms in different countries. * Dynamic Modeling: Transitioning the model from a static three-period snapshot to a dynamic model that accounts for changing market sentiments over time.

Summary updated

Abstract

Abstract The informativeness of environmental, social, and governance (ESG) scores and their actual impact on firms remains understudied. To address this gap in the literature, we make theoretical predictions and conduct empirical research revealing that a high ESG score is associated with a lower probability of ESG scandals and lower stock returns during a scandal event. Our results suggest that ESG scores are heterogeneous but informative, and that a strong ESG reputation may have both positive and negative consequences for firms. Drawing on our findings, we develop a model and showcase that firms face an optimization problem when determining optimal ESG investment levels. Two equilibria may exist based on the trade-off between ESG scandal losses and ESG adjustment costs. Our model explains why certain firms make heterogeneous ESG decisions
BibTeX
@inproceedings{sun2024esg,
  author = {Wenya Sun and Yichen Luo and Siu Ming Yiu and Luping Yu and Wenzhi Ding},
  title = {{ESG scores, scandal probability, and event returns}},
  booktitle = {Financial},
  year = {2024},
  doi = {10.1186/s40854-024-00635-1},
}
Figure