ICYMI | Using ESG Ratings to Build a Sustainability Investing Strategy

Background

Over the past few years, there has been a significant increase in the number of investors motivated to incorporate sustainable considerations into their decision-making process. As investors seek to diversify their portfolios, the global demand for information related to the roles played by ESG factors in corporations has increased enormously in prominence and consideration. What began as a small segment of investors disassociating their investments from contentious businesses has now spread to conventional investing. Today, global investors include companies’ sustainable strategies in their investment decisions (Gregory Unruh et al., “Investing for a Sustainable Future,” MIT Sloan Review, May 11, 2016, http://bit.ly/2sofq6W). Furthermore, executives from BNY Mellon Corporate Trust have acknowledged an increased investor demand, given that institutional investors are looking to maintain a percentage of socially responsible companies in their portfolios (James Maitland et al., “Contributing Solutions to Global ESG Challenges,” June 2014, https://bny.mn/2tA1gPZ).

It is evident that sustainable investing has evolved from a limited range of investments focused on screening undesirable exposures to a multitude of companies developing means to achieve sustainable outcomes. Consequently, investors are demanding a variety of sustainable products, and investment professionals are increasingly observing the links between sustainable initiatives and a company’s operational strength, efficiency, and management of long-term financial risks.

Despite this increasing demand, there is still much ambiguity as to what exactly is meant by sustainability in corporations and how investors can gather relevant sustainability and related ESG data. The current authors have previously written about this investor demand (Silvia Romero et al., “An Overview of Sustainability Reporting Practices,” CPA Journal, March 2014, http://bit.ly/2Nicm4r), the oversight of sustainability reporting (Beixin Lin et al., “Sustainability Reporting Standards: Who Should Be in Charge?” Today’s CPA, September 2015), and practices in the United States and Europe (Silvia Romero et al., “A Comparison of Sustainability in European and U.S. Corporations,” International Journal of Data Analysis and Information Systems, January 2014, http://bit.ly/2KG74xV).

To date, there is still no consistent framework for sustainability reporting by corporations. It is also well known that the ranking of companies by rating agencies are utilized by money managers as part of their investment strategies. The authors’ previous study noted that a significant lack of consistency and transparency has been found among the various ratings agencies, thus impeding greater efficiency in the capital markets. That study found little consistency across the rankings published by Newsweek, Forbes, and CSR Magazine Global, and little transparency in the underlying methodologies used in compiling the rankings. Information is the lifeblood of capital markets; if the information is short-term and thin, then these same characteristics will define the market. Unfortunately, the current reporting model, framed by IFRS, national standards, and stock exchange rules, does not provide the necessary framework to enable nonfinancial factors to be systematically taken into account in decision making. To this end, there is increased demand for more information regarding corporate social responsibility (CSR) as well as ESG rankings; this study attempts to bridge this information gap.

Overview of Sustainability Metrics Used

The source data was retrieved from Bloomberg Professional Services (a.k.a. the Bloomberg Terminal), generally perceived as the professional investor’s tool of choice. Company selection was driven by impact; with this as the guide, market capitalization was used as the primary selection criterion based upon the belief that size and impact are related. Undoubtedly, there are other metrics that could be used, such as enterprise value, brand value, or geopolitical reach; however, market cap brings several benefits to the analysis. The market cap number can be traced directly to the company, is a straightforward calculation devoid of estimates and assumptions, and is generally perceived as the value of the company.

Companies were selected on April 24, 2018, during active trading hours. The Bloomberg Terminal was used to search all the major worldwide exchanges for companies with a market capitalization greater than $15 billion based on actively traded common stock. There were 968 companies, with a total market cap of $48.5 trillion; all major sectors and industries were represented. Since not all 2017 data is available, data from 2016 was utilized. Furthermore, not all companies are scored by all of the three rating agencies used in the study; of the selected companies, 925 had Bloomberg ESG disclosure scores, 894 had RobecoSAM ESG rating/ranking scores, and 691 firms had Sustainalytics rating/ranking scores.

The Bloomberg ESG Disclosure Score ranges from 0.1 for companies that disclose a minimum amount of ESG data to 100 for those that disclose every data point collected by Bloomberg. Each data point is weighted in terms of importance, with data such as greenhouse gas emissions carrying greater weight than other disclosures. The score is also tailored to different industry sectors. The RobecoSAM Total Sustainability Rank, ranging from 0 to 100, is converted from total sustainability score, based on the RobecoSAM Corporate Sustainability Assessment. The Sustainalytics rank also ranges from 0 to 100 and indicates the overall percentile rank assigned to the company based on its ESG total score relative to its industry peers; hereinafter, these rankings are simply referred to as “scores.”

Exhibit 1 illustrates the scope of coverage of each rating agency. For comparison, 96% of sample companies are covered by Bloomberg, 92% by RobecoSAM, and 71% by Sustainalytics.

Data Analysis

Exhibit 2 depicts the descriptive statistics of the three sustainability metrics. Bloomberg scores range from 3 to 74 [average 38, standard deviation (SD) 15], and RobecoSAM scores range from 3 to 100 (average 52, SD 30). In comparison, Sustainalytics scores have a similar range, but the average is higher at 62, with a smaller SD of 27.

Correlation Analysis

Correlation analyses were undertaken between the three ESG metrics. The results of the analysis indicate strong positive correlations exist between the Bloomberg scores and the other two rankings. Furthermore, the two sustainability rankings are also highly correlated with each other. This finding presents a sharp contrast to the previous study, where there were large variances between ranking agencies.

A series of scatter diagrams provides a visualization of the correlations among the three metrics. Exhibit 3 compares each company’s Sustainalytics score with its RobecoSAM score. Exhibit 4 compares each company’s Bloomberg score with its RobecoSAM score, and Exhibit 5 compares each company’s Bloomberg score with its Sustainalytics score. The linear shape of the scattered dots depicted in these three diagrams shows the positive correlation between the two metrics compared in each of the three scenarios and suggests that the scores and ratings from these agencies are consistent. Given the independence of these three rating agencies, these results give a level of comfort as to the rating agencies’ indication of a company’s ESG profile.

Is There a Relationship between Rankings and ESG Disclosure Scores?

Next, the authors undertook a regression analysis. Because a company’s ESG initiatives and reporting can be affected by other factors, regression is necessary to control for size and industry, which have been found in previous research to affect ESG disclosure and reporting quality (David Campbell, “Intra- and Intersectoral Effects in Environmental Disclosures: Evidence for Legitimacy Theory?” Business Strategy and the Environment, 2003, http://bit.ly/2MtdL76; Craig Deegan and Ben Gordon, “A Study of the Environmental Disclosure Practices of Australian Corporations,” Accounting and Business Research, vol. 26, no. 3, 1996, http://bit.ly/2lBb4Vq). The size variable is calculated as a logarithm of total assets. The industry variable, ESSI (ethics, sustainability, and social impact), adopts a value of 1 if the company belongs to an industry identified as an environmentally sensitive industry, and 0 if not. These industries are agriculture, automotive, aviation, chemical, construction, construction materials, energy, energy utilities, forest and paper products, logistics, metal products, mining, railroad, waste management and water utilities. Companies belonging to those industries provide better sustainability disclosures, one of the suggested reasons being the need to legitimize their activities given that the public considers them potentially harmful (Torbjörn Tagesson et al., “What Explains the Extent and Content of Social and Environmental Disclosures in Corporate Websites: A Study of Social and Environmental Reporting in Swedish Listed Corporations,” Corporate Social Responsibility and Environmental Management, November/December 2009, http://bit.ly/2tHYeID; Ramin Gamerschlag et al., “Determinants of Voluntary CSR Disclosure: Empirical Evidence from Germany,” Review of Managerial Science, July 2011, http://bit.ly/2Kgj6BO; Manuel Castelo Branco and Lúcia Lima Rodrigues, “Factors Influencing Social Responsibility Disclosure by Portuguese Companies,” Journal of Business Ethics, December 2008, http://bit.ly/2Kn3icF).

The first two regressions examine the relationship between Bloomberg scores and the other two scores. These regression results (presented in this online-only supplement to the article) show that, consistent with the scattergrams, both the RobecoSAM and Sustainalytics scores are strongly associated with the Bloomberg scores, even after controlling for factors that are found to affect ESG disclosures. This positive correlation is an important finding, because the Bloomberg score evaluates the level of ESG information disclosed by a company, but not the quality of ESG practices, whereas RobecoSAM and Sustainalytics present ESG scores based upon a company’s sustainability practice. Thus, the analysis confirms that good ESG disclosures are positively correlated with higher rankings. This result affirmatively answers the first question: “Is there a relationship between sustainability rankings and ESG disclosure scores?”

Is There Commonality between Rating Agencies?

To address the second question, a third regression is run to investigate the relationship between the RobecoSAM and Sustainalytics scores. These regression results (presented in the online-only supplement) also confirm the positive association between these two sustainability rankings. This analysis evaluates the quality of ESG/sustainability initiatives, not the level of ESG disclosure. ESSI appears to be nonsignificant, highlighting the fact that the industry effect is present in disclosure and not in the sustainability rankings.

Overall, the regression results lend further support to the above correlation analysis, but are also in contrast with the results of the previous study. Again, the possible explanation for the alternative results is that the previous study used publicly available information that was designed for the general public, while this study uses proprietary information constructed by professionals who are focused on providing reliable information for the investment community.

Does Good Corporate Citizenship Support Value Creation and Better Performance?

Previous studies indicate that socially responsible companies perform better; a major assumption with this statement, however, is that the available information is accurate, relevant, complete, timely, and independently constructed (i.e., all of the typical attributes reviewed when evaluating the quality of audit evidence). The absence of mandatory guidelines and the lack of required assurance make it somewhat difficult to consistently measure social responsibility within and across companies.

Notwithstanding this level of assurance, the means of the return on equity (ROE) and the enterprise value of companies at the top and bottom of the three ESG metrics were compared to validate the relationship between the Bloomberg score and the RobecoSAM and Sustainalytics scores with performance/value. The results show a significant relationship between the enterprise value, the Bloomberg score, and the RobecoSAM score. Companies with top ESG rankings have significantly higher value than companies at the bottom. There is, however, no correlation to ROE. On the other hand, Sustainalytics scores show positive correlation to ROE, but no correlation to enterprise value.

Despite these mixed results, and the fact that there are many other factors that affect ESG disclosure and sustainability implementation, the results suggest that companies with higher sustainability scores/rankings tend to have higher ROE and enterprise value. (The supplement  presents the mean values of the variables used to divide the sample into two groups and then presents the differences in means for each of three sustainability metrics.) These results provide some evidence that good ESG practice and reporting have positive impact on value creation and performance in companies. Furthermore, this relationship may also indicate that companies with higher enterprise value have the means and resources to better implement CSR practices and disclosures than companies with lower enterprise value.

Limitations

The results of this study suggest that the ESG rankings data provided by the examined ranking agencies is not updated on a timely basis and standardization is missing; thus, although the data was sourced on April 24, 2018, few companies had data for 2017. Given the rapid dissemination of information worldwide, the value of these metrics is lessened due to the lack of timeliness. The lack of standardization also could have affected the results of the study. This validates the findings of the previous study, underscoring the value that a standardized CSR/ESG framework, reporting standards, and governance would have if it were in place and followed by corporations, ratings agencies, and other interested parties.

It is also clear that robust, efficient markets require the necessary pillars of reporting. These pillars, as discussed in the previous study, are as follows:

• A widely accepted conceptually sound framework
• Standardized reporting and disclosure
• Management assurance
• Independent assurance
• Regulatory oversight with enforcement powers.

As noted in the previous study, SASB has issued voluntary reporting standards in various sectors. It is apparent, however, that equilibrium in information can only be achieved if a set of consistent standards with strong independent oversight is implemented.

Finding a Reliable Compass

This article sought to answer three questions. Regarding the first, the authors found that significant relationships indeed exist between companies’ sustainability ratings and their ESG disclosure scores. Given the independence of these three rating agencies, the positive correlation results give a level of comfort. For the second, the authors found commonality between three well-known providers in the global investment community: Bloomberg, RobecoSAM and Sustainalytics. This should give an additional level of confidence to investors regarding the consistency of ESG ratings. Finally, although the results on the third question were mixed, the authors found some evidence that good corporate citizenship indeed led to increased enterprise value and ROE.

The findings of this study can be useful to investors and help them navigate the difficult territory of sustainability and impact investing. Despite the fact that ESG metrics are an integral part of the formula used in sustainability investment strategies, there are other factors to consider. It is thus necessary for investors to tread carefully, do their necessary homework, and confer with their investment advisors prior to embarking on ESG/sustainable investment strategies.