Author Details

Guido Giese

Guido Giese
Executive Director, MSCI Research

Zoltán Nagy

Zoltán Nagy
Executive Director, MSCI Research

Linda-Eling Lee

Linda-Eling Lee
Managing Director, MSCI Research

Social Sharing

Extended Viewer

Which ESG Issues Mattered Most? Defining Event and Erosion Risks

  • Very different ESG issues can be material for different industries. Our research suggests that risks can be divided into two main types: “event” risks and “erosion” risks to companies’ long-term competitiveness.
  • Environmental Key Issues were purely erosion-driven (they unfolded over time). Social Key Issues showed a mix of event-driven and erosion-driven characteristics, while Governance issues had the highest share of event risks.
  • Highly active portfolio managers may want to focus on mitigating short-term event risks. In comparison, portfolio managers building diversified portfolios with long investment horizons may be more focused on long-term erosion risks.

Which environmental, social and governance (ESG) issues have mattered most for companies’ financial performance?

Very different ESG issues can be material for different industries, a view adopted by organizations such as the Sustainability Accounting Standards Board,1 which have promoted industry-specific indicators for corporate disclosure. These issues range from pollution (environmental) and labor rights (social) to board diversity (governance). Our research suggests that, rather than categorizing issues by E, S or G, ESG risks might be better divided into two main types: “event” risks and “erosion” risks to companies’ long-term competitiveness.

 

Industry Differences

Our research used scores from the MSCI ESG Ratings of the MSCI ACWI Investable Market Index’s constituents from 2012 to 2019. In the MSCI ESG Ratings methodology, Key Issues are selected and weighted for each GICS®2 sub-industry based on potential exposure to the respective issue. Some GICS sub-industries are exposed to as few as four underlying Key Issues, while other industries are exposed to as many as eight underlying Key Issues (see examples in the exhibit below).

 

ESG Key Issues per GICS Sub-industry for Selected GICS Sectors

 

GICS Sector GICS Sub-Industry Code GICS Sub-Industry Name Environmental Social Governance
Carbon Emissions Product Carbon Footprint Climate Change Vulnerability Water Stress Biodiversity Land Use Raw Material Sourcing Financing Environmental Impact Toxic Emissions & Waste Packaging Material Waste Electronic Waste Opportunities in Clean Tech Opportunities in Green Building Opportunities in Renewable Energy Labor Management Health & Safety Human Capital Development Supply Chain Labor Standards Product Safety & Quality Chemical Safety Financial Product Safety Privacy & Data Security Insuring Health & Demographic Risk Responsible Investment Controversial Sourcing Access to Communications Access to Finance Access to Health Care Opportunities in Nutrition & Health Corporate Governance Corruption & Instability Financial System Instability
Energy 10101010 Oil & Gas Drilling                                                              
10101020 Oil & Gas Equipment & Services                                                              
10102010 Integrated Oil & Gas                                                              
10102020 Oil & Gas Exploration & Production                                                              
10102030 Oil & Gas Refining & Marketing                                                              
10102040 Oil & Gas Storage & Transportation                                                              
10102050 Coal & Consumable Fuels                                                              
Materials 15101010 Commodity Chemicals                                                              
15101020 Diversified Chemicals                                                              
15101030 Fertilizers & Agricultural Chemicals                                                              
15101040 Industrial Gases                                                              
15101050 Specialty Chemicals                                                              
15102010 Construction Materials                                                              
15103010 Metal & Glass Containers                                                              
15103020 Paper Packaging                                                              
15104010 Aluminum                                                              
15104020 Diversified Metals & Mining                                                              
15104025 Copper                                                              
15104030 Gold                                                              
15104040 Precious Metals & Minerals                                                              
15104045 Silver                                                              
15104050 Steel                                                              
15105010 Forest Products                                                              
15105020 Paper Products                                                              
Health Care 35101020 Health Care Supplies                                                              
35102010 Health Care Distributors                                                              
35102015 Health Care Services                                                              
35102020 Health Care Facilities                                                              
35102030 Managed Health Care                                                              
35103010 Health Care Technology                                                              
35201010 Biotechnology                                                              
35202010 Pharmaceuticals                                                              
35203010 Life Sciences Tools & Services                                                              
Utilities 55101010 Electric Utilities                                                              
55102010 Gas Utilities                                                              
55103010 Multi-Utilities                                                              
55104010 Water Utilities                                                              
55105010 Independent Power Producers & Energy Traders                                                              
55105020 Renewable Electricity                                                              

 

Data as of December 2019. Source: MSCI ESG Research LLC

Our analysis focused on the 11 Key Issues that are most commonly used in calculating a company’s ESG Rating and thus show the broadest historical coverage of companies within MSCI’s dataset:

  • Environmental (“E pillar”): Carbon emissions, water stress, toxic emissions and waste
  • Social (“S pillar”): Labor management, health and safety, human-capital management and privacy and data security
  • Governance (“G pillar”): Corporate governance, business ethics, corruption and instability and anticompetitive practices

 

Transmission Channels

We extended previous research on how the overall MSCI ESG Rating has reflected financial risk and returns through three economic-transmission channels:

  1. The cash-flow channel, whereby companies better at managing intangible capital (such as employees) may have been more competitive and hence more profitable over time
  2. Idiosyncratic risk, whereby companies with stronger risk management practices may have experienced fewer incidents that triggered unanticipated costs, such as accidents
  3. Systematic risk, whereby companies that used resources more efficiently may have been less susceptible to market shocks such as fluctuations in energy price

We examined how each of the three economic-transmission channels performed across these 11 Key Issues by comparing the top- and bottom-scoring equal-weighted quintiles against the financial variables associated with each channel (see the exhibit below).3 We employed profitability, residual volatility and systematic volatility as target variables to test the financial significance of each of the three transmission channels.

The exhibit below shows the corresponding t-statistics, where the colors indicate the direction of alignment: Blue bars indicate results where ESG characteristics were aligned with the expected financial results, and red bars indicate a misalignment.4

 

Most Common ESG Key Issues and Significance of Their Exposure to Target Financial Variables

Data from 2012 to 2019 on all constituents of the MSCI ACWI IMI. Source: MSCI ESG Research LLC

We found that, of the three transmission channels, the idiosyncratic-risk channel showed the most significant results across the 11 Key Issues tested. Furthermore, the Key Issues categorized under the Governance pillar showed, on average, the most significant results of all three channels. Companies with strong corporate governance had significantly better profitability, lower stock-specific risk and lower systemic risk than low-scoring companies across the Governance Key Issues, during the seven-year study period between December 2012 and December 2019. In addition, within the set of Social Key Issues, health and safety showed significant empirical results.

These results indicate that, during the study period, financial markets were largely focused on events that could immediately affect company valuations. What about Key Issues that capture intangible ESG characteristics over longer periods?

 

Event vs. Erosion Risks

Some Key Issues aim to capture risks related to events — such as fraud and oil spills — that can affect companies’ stock price over short periods. Other Key Issues, such as resource efficiency, relate more to long-term risks that can erode a company’s stock price over long periods. Some Key Issues may exhibit characteristics of both event and erosion risk.

Are we able to verify these two types of ESG risks? To do so, we need to quantify event and erosion risks using appropriate risk and performance measures. Below, we used equal-weighted quintile portfolios for each Key Issue and compare the Q5 (best ESG characteristics) to the Q1 quintile (worst) using the following measures:

Event risk: For each Key Issue, we created equal-weighted quintile portfolios, which we rebalanced on a monthly basis. We compared the frequency with which the top-scoring (Q5) quintile and the bottom-scoring (Q1) companies experienced a 90% or greater stock-price drawdown in the 36 months after the publication of a company’s rating. We used the Q1/Q5 frequency ratio as a measure for the effectiveness for identifying event risk — the higher the ratio, the more effective.

Erosion risk: We compared the annualized cumulative financial performance of these top-scoring (Q5) versus bottom-scoring (Q1) companies for each Key Issue as a measure for erosion risk.

The following exhibit summarizes both risk measures for the 11 Key Issues under consideration, with the log of the drawdown-frequency ratio indicating event risk and the performance difference indicating erosion risk. We found that nine out of 11 issues showed a positive Q5-Q1 performance contribution cumulatively over the study period, while seven out of the 11 issues showed a positive contribution in terms of identifying differences in event-driven (i.e., stock-specific) risks.

 

Event Risk Versus Erosion Risk of 11 MSCI ESG Key Issues

Data from 2012 to 2019 on constituents of the MSCI ACWI IMI. Source: MSCI ESG Research LLC

We saw distinct differences across the issues categorized under the E, S and G pillars:

  • Environmental Key Issues:
    • The three environmental Key Issues — carbon emissions, water stress and toxic emissions — were driven by erosion risk. They showed positive long-term differences between the top- and bottom-scoring companies. However, both top- and bottom-scoring companies showed negligible differences in propensity to event risks.
  • Social Key Issues:
    • The Key Issues under the Social pillar showed more balanced results, with differences in labor management (e.g., mitigating labor conflicts) showing strong event- and erosion-risk characteristics. In fact, top-scoring companies on labor management not only outperformed bottom-scoring companies by an average of 3% per year, but also showed significant reductions in event risks — i.e., the top-scoring companies experienced severe stock-price losses only one-fifth as frequently as the low scorers.
    • Health and safety showed equally strong positive long-term performance differences; however, it showed minimal differentiation on event risk between high- and low-scoring companies.
    • Interestingly, though high-profile cases such as Equifax and Facebook might have suggested otherwise, differences in companies’ management of privacy and data security have not historically contributed to positive performance during this period, nor have the top-scoring companies avoided more negative events than low-scoring companies.
  • Governance Key Issues:
    • Governance-related Key Issues as a whole showed the strongest results along both risk dimensions. However, business ethics and anticompetitive practices showed much stronger event-risk characteristics with minimal erosion risk. The bottom-scoring companies on business ethics were approximately four times more likely than top-scoring companies to experience a severe stock-price loss during this period, while the bottom-scoring companies on corruption were only about 1 1/2 times more likely than the top-scoring companies to do so.
    • In contrast, corporate governance — and especially the corruption Key Issue — showed positive long-term differences, with stronger erosion-risk characteristics and less event-driven risk differentiation.

Erosion-driven ESG issues understandably grab fewer headlines than the more abrupt and sometimes dramatic event-driven issues. How have these Key Issues performed over the entire seven-year study period? The exhibit below shows the performance of the equal-weighted Q5-Q1 portfolios over time.

Carbon emissions (E pillar) showed the most significant gross outperformance of all Key Issues, with health and safety and labor management (both S pillar) and corruption (G pillar) in second, third and fourth place, respectively. Between 2012 and 2019, for these four top-performing Key Issues, the outperformance of the top-scoring companies over the bottom companies ranged from 26% (carbon emissions) to 21% (labor management) cumulatively, or 3.8% and 3.0% on an average annualized basis. These top-performing Key Issues indicate that erosion risks were more evenly distributed across the three pillars than event risks, which were more concentrated in the G pillar.

With the exception of labor management, which also exhibited strong event-risk characteristics, the performance effects in the next exhibit were not driven by large losses suffered from distinct negative events during the study period. As the exhibit above indicated, these Key Issues reflected minimal event-risk differences; instead, the gains in returns accumulated gradually throughout the study period. This confirms the view that these Key Issues can present erosion risks that can more subtly but cumulatively influence stock prices over longer periods of time.

 

Cumulative Q5-Q1 Performance for Erosion-Driven Key Issues

Data from 2012 to 2019 on all constituents of the MSCI ACWI IMI. This exhibit shows how the top-scoring quintile (Q5) minus the bottom-scoring quintile (Q1) performed for erosion-driven Key Issues. Source: MSCI ESG Research LLC

In addition to considering absolute-performance differentials, we also employed MSCI’s Barra Global Equity Model for Long-Term Investors (GEMLT) to parse the extent to which these performance differentials could be attributed to common factor exposures or were related to stock-specific performance. After controlling for other factors, we found that 10 out of 11 Key Issues showed a positive stock-specific performance contribution, with only privacy and data security showing a negative result. In fact, of the 11 Key Issues, five showed a stock-specific performance contribution of at least 1% per year over the study period.

 

Implications for Portfolio Managers

Empirical evidence suggests that ESG risks can come in different flavors: event-driven, erosion-driven or a mixture of both. While Environmental Key Issues such as carbon emissions were purely erosion-driven (i.e., they unfolded continuously over time), Key Issues in the Social pillar were more balanced; some, such as labor management, showed both strong event-driven and erosion-driven performance characteristics. Governance-related Key Issues ran the gamut: All four Key Issues categorized under Governance showed event- and erosion-driven performance characteristics, with the highest share of event risks among the three pillars. The Governance Key Issues’ event-risk characteristics help explain why Governance as a whole has consistently shown the strongest significance for stock-price risks over shorter periods of time.

Our finding has important implications for the relevant time horizon for ESG integration in portfolio construction. Highly active portfolio managers building concentrated stock portfolios with relatively high turnover may want to focus on identifying and mitigating short-term event risks. This group may find erosion risks to be less relevant if these risks unfold over longer time horizons.

In comparison, portfolio managers building broad diversified portfolios with long investment horizons (e.g., indexed or buy-and-hold investors) may be more focused on long-term erosion risks in their choice of ESG criteria and ESG integration and may aim to mitigate event risks through diversification. They may face headline risk from unexpected events, but may feel it’s a reasonable tradeoff against potential erosion risks.

 

 

1For example, SASB states that “SASB standards differ by industry, enabling investors and companies to compare performance from company to company within an industry.”

2The Global Industry Classification Standard (GICS®) was jointly developed by MSCI and Standard & Poor’s.

3The financial variables are transformed into z-score format, or standardized, by subtracting the mean and dividing by the standard deviation. Thus financial variables are brought to a common scale and are directly comparable with each other.

4T-statistics indicate the likelihood that the difference between Q5 and Q1 is not just accidental. A t-statistic over 2 is viewed as significant.

 

 

Further Reading

Deconstructing ESG Ratings Performance

Is ESG All About the ‘G’? That Depends on Your Time Horizon.

Foundations of ESG Investing: How ESG Affects Equity Valuation, Risk and Performance

Has ESG affected stock performance?

ESG and the cost of capital

Building Better ESG Indexes: 30 Years On

MSCI ESG Indexes during the coronavirus crisis

Understanding MSCI ESG Indexes

Regulation