ESG PERFORMANCE AND TAIL RISK MITIGATION: EMPIRICAL EVIDENCE FROM THE S&P 500

ABSTRACT

The article examines modern approaches to analyzing the relationship between Environmental, Social, and Governance (ESG) performance and corporate resilience to extreme market shocks. Based on data from 496 constituents of the S&P 500 index, a comprehensive analysis of "tail risks" is conducted using a quantile regression framework. Special attention is given to the methodology of intra-sector normalization (Z-score), which allows for the isolation of the ESG effect regardless of sectoral affiliation. It is empirically substantiated that high ESG ratings serve as a tool for reducing the Expected Shortfall (ES), providing an "insurance effect" during periods of high volatility. The research findings confirm the necessity of integrating ESG metrics into the prudential risk management systems of institutional investors in Kazakhstan and international markets.

АННОТАЦИЯ

В статье рассматриваются современные подходы к анализу взаимосвязи между показателями экологической, социальной и управленческой ответственности (ESG) и устойчивостью компаний к экстремальным рыночным шокам. На основе данных 496 компаний индекса S&P 500 проведен комплексный анализ «хвостовых рисков» (tail risks) с использованием модели квантильной регрессии. Особое внимание уделено методологии внутрисекторальной нормализации (Z-score), позволяющей выявить чистый эффект ESG-факторов вне зависимости от отраслевой принадлежности. Эмпирически обосновано, что высокие ESG-рейтинги выступают инструментом снижения показателя Expected Shortfall (ES), обеспечивая «эффект страхования» в периоды высокой волатильности. Результаты исследования подтверждают необходимость интеграции ESG-метрик в системы пруденциального риск-менеджмента институциональных инвесторов в Казахстане и на международных рынках.

Keywords:  ESG performance, tail risk, Expected Shortfall, quantile regression, S&P 500, risk management, sustainable development.

Ключевые слова: ESG-рейтинг, хвостовой риск, Expected Shortfall, квантильная регрессия, S&P 500, риск-менеджмент, устойчивое развитие.

1. INTRODUCTION

In the modern financial landscape, the evaluation of corporate performance has transcended beyond traditional balance sheet metrics. As global markets face increasing volatility and "black swan" events, institutional investors are shifting their focus from simple alpha generation to robust risk mitigation. Within this context, Environmental, Social, and Governance (ESG) criteria have emerged not merely as ethical benchmarks but as critical indicators of corporate resilience.

The core of the current financial debate lies in the relationship between a firm’s non-financial disclosures and its market stability. Traditional risk models, such as Value-at-Risk (VaR), have historically been criticized for their inability to accurately capture the severity of extreme losses during market crashes-a phenomenon known as "tail risk." To address this limitation, modern risk management increasingly relies on Expected Shortfall (ES), which provides a more comprehensive measure of the average loss in the worst-case scenarios of the return distribution.

This research aims to investigate whether superior ESG performance serves as a "protective shield" for investors in the S&P 500 index. While numerous studies have explored the impact of ESG on average stock returns, there is a significant gap in literature regarding its effectiveness in thinning the "left tail" of the distribution. We hypothesize that companies with high ESG scores possess greater social capital and stronger governance structures, which act as a form of "intangible insurance" during periods of systemic stress.

A unique contribution of this study is the application of intra-sector normalization (Z-scores). By comparing firms only against their direct industry peers, we isolate the idiosyncratic "ESG effect" from broader sectoral trends. Furthermore, by utilizing Quantile Regression, this study provides a more nuanced analysis of how ESG influences the 5th percentile of returns, rather than just the mean.

The findings of this paper are particularly relevant for institutional fund managers and policymakers in emerging markets, including the Republic of Kazakhstan, who are currently integrating ESG frameworks into their prudential oversight systems. By providing empirical evidence from the S&P 500, this study demonstrates that ESG integration is a mathematically sound strategy for managing extreme market shocks and ensuring long-term portfolio stability.

2. LITERATURE REVIEW

The academic discourse regarding Environmental, Social, and Governance (ESG) factors has evolved from viewing them as purely ethical considerations to recognizing them as fundamental determinants of financial risk. The theoretical foundation of this study rests on two primary pillars: the Social Capital Theory and the Risk Mitigation Hypothesis.

2.1. Social Capital and Corporate Resilience. A seminal contribution to this field was made by Lins, Servaes, and Tamayo (2017), who analyzed the performance of firms during the 2008-2009 financial crisis. Their research demonstrated that high-social-capital firms-measured by high ESG scores-experienced stock returns that were four to seven percentage points higher than low-social-capital firms. The authors argued that ESG investments build a "reservoir of trust" with stakeholders, which acts as an informal insurance policy during periods of systemic stress. When market trust collapses, investors are less likely to liquidate positions in companies that have consistently demonstrated high governance standards and social responsibility.

2.2. The Risk Mitigation Hypothesis. Building on this, Giese et al. (2019) from MSCI provided empirical evidence that ESG information influences the valuation and risk profile of companies through three main channels: the cash-flow channel, the idiosyncratic risk channel, and the valuation channel. Most relevant to this study is the idiosyncratic risk channel, which suggests that high-ESG firms are better at managing operational risks, leading to a lower frequency of "tail events" such as regulatory fines, environmental disasters, or governance scandals. Consequently, these firms exhibit lower systematic risk and a more stable return distribution. Moreover, the risk mitigation hypothesis implies that superior ESG performance reduces idiosyncratic risk by enhancing internal control systems. While systematic risk affects all market participants, firms with robust governance (G) and environmental (E) risk management protocols are less susceptible to firm-specific shocks. By normalizing ESG scores within sectors, this study isolates this idiosyncratic resilience, proving that "best-in-class" companies are inherently more stable than their industry peers during periods of heightened market volatility.

2.3. Beyond Traditional Metrics: Expected Shortfall and Quantile Regression. While much of the early literature focused on the relationship between ESG and average returns or standard volatility, recent studies have begun to shift toward "tail risk" metrics. However, traditional linear models often fail to capture the behavior of returns during extreme market downturns. As noted by Koenker and Bassett (1978) in their development of Quantile Regression, financial variables often exhibit non-linear relationships at the edges of the distribution. By focusing on the 5th percentile of returns (Expected Shortfall), this study addresses a critical gap in the existing literature, providing a more precise assessment of how ESG performance thins the "left tail" of market risk.

Furthermore, this research addresses the "sector bias" often found in ESG studies. By utilizing intra-sector Z-scores, we build upon the methodology suggested by recent institutional frameworks, ensuring that our findings reflect corporate efficiency rather than a mere preference for "low-impact" industries like technology or finance over "high-impact" sectors like energy or materials.

3. DATA AND METHODOLOGY

3.1. Sample Selection and Data Sources. The initial universe for this study consists of the companies comprising the S&P 500 index. After filtering for data availability of both daily stock prices and ESG ratings, the final sample includes 496 firms. The time horizon for stock price analysis spans from early 2025 to March 2026, capturing a full year of market dynamics. The ESG scores and financial data for the S&P 500 constituents were extracted from the Bloomberg Terminal, ensuring high data integrity and consistency for the period of 2025-2026.

The data collection process involved a multi-stage filtering approach. Initially, companies with more than 10% of missing daily price observations or inconsistent ESG disclosure scores were excluded to maintain the statistical power of the model. The use of the Bloomberg Terminal allowed for the retrieval of 'hard' ESG disclosure data, which is less prone to subjective reporting biases compared to survey-based ratings. All financial time series were cross-referenced for corporate actions, such as stock splits and dividends, to ensure that the log-returns accurately reflect the economic reality of the 2025-2026 market period.

3.2. Quantitative Measure of Tail Risk: Expected Shortfall (ES). To capture the extreme downside risk, this study utilizes Expected Shortfall (ES) at the 5% confidence level (q=0.05). Unlike Value-at-Risk (VaR), which only identifies the threshold of loss, ES calculates the average loss in the worst 5% of cases, providing a more coherent measure of tail risk. The daily log-returns () were calculated as:

The Expected Shortfall is then defined as the conditional expectation of loss exceeding the VaR:

3.3. Sector-Neutral ESG Normalization (Z-Score). To eliminate industry-specific bias (where some sectors naturally have higher ESG scores than others), we applied intra-sector normalization. We calculated a Z-score for each firm relative to its GICS sector peers:

Where ​ is the mean ESG score of sector s, and is the standard deviation. This approach ensures that a "leader" is defined by its performance relative to its direct competitors. This intra-sector normalization is critical because ESG ratings are highly heterogeneous across industries; for instance, a 'Materials' company might have a lower absolute score than a 'Technology' firm due to inherent environmental footprints. By applying the Z-score transformation, we effectively reposition each firm within its own 'playing field,' identifying leaders who outperform their immediate competitors in sustainable management regardless of their industry's baseline impact.

3.4. Quantile Regression Framework. The primary hypothesis is tested using Quantile Regression, which allows for the estimation of the relationship between variables at specific points of the distribution. We focus on the τ=0.05 quantile to analyze the impact of ESG on the "left tail." The model is specified as follows:

By including sector dummy variables (​), we control for any remaining unobserved industry fixed effects, ensuring that the coefficient reflects the true impact of ESG performance on risk mitigation.

Unlike Ordinary Least Squares (OLS) regression, which estimates the conditional mean of the dependent variable, Quantile Regression provides a comprehensive view of the entire distribution. In financial risk management, the 'average' relationship is often misleading, as the factors that stabilize a stock during normal times may differ from those that protect it during a crash. Our focus on the τ=0.05 quantile specifically targets the regime of financial distress, where the risk-mitigating properties of ESG are expected to be most potent.

4. RESULTS AND DISCUSSION

4.1. Descriptive Statistics and Visual Analysis

Before conducting the formal econometric analysis, we visualized the return distributions of two distinct groups: ESG Leaders (top 20% by sector-neutral ESG score) and ESG Laggards (bottom 20%).

Figure 1. Comparison of Return Distributions: ESG Leaders (Top 20%) vs. ESG Laggards (Bottom 20%)

As illustrated in Figure 1, there is a visible disparity in the "thickness" of the left tails. The distribution for ESG laggards (red area) exhibits a more pronounced extension into the extreme negative zone, suggesting a higher frequency and magnitude of tail events. In contrast, ESG leaders (green area) show a more compressed distribution in the loss zone, providing preliminary evidence for the risk-mitigation hypothesis.

4.2. Quantile Regression Results

The core hypothesis was tested using a quantile regression at the 5th percentile (τ=0.05). The results of the model are summarized in Table 1.

Table 1.

 Quantile Regression Results (Quantile = 0.05)

The empirical results reveal a statistically significant positive relationship between the intra-sector ESG Z-score and Expected Shortfall (β=0.0006, p=0.038). In the context of financial risk, where ES is a negative value representing potential losses, a positive coefficient signifies a reduction in risk. This indicates that as a firm's ESG performance improves relative to its sector peers, its exposure to extreme downside shocks decreases.

4.3. Sector-Specific Robustness

To ensure that these findings are not driven by a single industry, we analyzed the average Expected Shortfall across all GICS sectors.

Figure 2. Average Expected Shortfall (ES 5%) by GICS Sector

Figure 2 confirms that the "ESG protection effect" is consistent across diverse industries. In sectors traditionally considered high-risk, such as Energy and Materials, the gap between ESG leaders and laggards remains significant. This underscores the importance of our Z-score methodology, proving that ESG is a relevant risk differentiator even within carbon-intensive or volatile industries.

 The consistency of results across GICS sectors is particularly noteworthy. For instance, in the Energy and Materials sectors-often criticized for high environmental footprints-the 'Leaders' still exhibit superior tail-risk resilience. This implies that even in traditionally 'brown' industries, superior governance and operational transparency (the G and S components of ESG) differentiate firms that can manage crises from those that cannot. This proves that ESG is a relative performance indicator rather than an absolute industry filter.

4.4. Discussion: The Mechanism of Resilience

The statistical significance of our findings (p<0.05) supports the "Insurance Effect" theory. High ESG scores act as a proxy for management quality and operational transparency. During market downturns, these firms benefit from lower idiosyncratic risk and higher investor trust, which prevents aggressive capital outflow. For institutional investors, particularly in developing markets like Kazakhstan, these results suggest that ESG metrics should be integrated into prudential risk frameworks to enhance portfolio stability against systemic shocks.

5. Conclusion

This study provided an empirical analysis of the relationship between ESG performance and tail risk mitigation within the S&P 500 index. By utilizing Expected Shortfall (ES) and a Quantile Regression framework, we demonstrated that companies with superior ESG ratings relative to their sector peers exhibit significantly lower exposure to extreme market shocks. The statistical significance of our findings (p=0.038) reinforces the "insurance effect" hypothesis, suggesting that corporate sustainability practices build institutional resilience and stakeholder trust.

Specifically, the positive and statistically significant coefficient of the ESG Z-score (β=0.0006, p=0.038) confirms that for every standard deviation improvement in a firm’s relative ESG performance, there is a measurable reduction in the average magnitude of tail losses. This empirical evidence supports the transition from qualitative ESG assessments to quantitative risk integration.

For financial managers and institutional investors, the implications are clear: ESG integration is a vital component of modern risk management. It allows for the identification of firms that are better prepared for "black swan" events. As global markets, including those in Central Asia and Kazakhstan, increasingly adopt sustainability reporting, the findings of this research provide a mathematical justification for the shift toward ESG-centric investment strategies.

References:

1. Bloomberg Terminal. Environmental, Social, and Governance (ESG) Data and Market Prices for S&P 500 Constituents / Bloomberg Professional Service. – 2026. – [Electronic resource]. – Mode of access: Bloomberg Terminal (date of access: 15.03.2026).
2. Giese, G. Foundations of ESG Investing: How ESG Affects Equity Valuation, Risk, and Performance / G. Giese, L. Hancock, S. Kahn, C. Meyer // Journal of Portfolio Management. – 2019. – Vol. 45, No. 5. – P. 69–83. – DOI: 10.3905/jpm.2019.45.5.069.
3. Koenker, R. Regression Quantiles / R. Koenker, G. Bassett Jr. // Econometrica. – 1978. – Vol. 46, No. 1. – P. 33–50. – DOI: 10.2307/1913643.
4. Lins, K. V. Social Capital, Trust, and Firm Performance: The Value of Corporate Social Responsibility during the Financial Crisis / K. V. Lins, H. Servaes, A. Tamayo // Journal of Finance. – 2017. – Vol. 72, No. 4. – P. 1785–1824. – DOI: 10.1111/jofi.12505.
5. MSCI ESG Research. ESG and the Cost of Capital: A Global Perspective / MSCI Technical Report. – 2024. – Mode of access: URL: https://www.msci.com/documents/1296102/293813/ESG_Cost_of_Capital.pdf (date of access: 18.03.2026).
6. Rockafellar, R. T. Optimization of Conditional Value-at-Risk / R. T. Rockafellar, S. Uryasev // Journal of Risk. – 2000. – Vol. 2. – P. 21–41. – DOI: 10.21314/JOR.2000.038.