Navigating Asia’s ESG data puzzle

It is no secret that the future growth of ESG investing is linked to data. In a 2019 survey¹, respondents listed data as one of the key obstacles to greater ESG integration. See Fig. 1.

Fig. 1. What are the obstacles to developing ESG integration?

For investors and fund managers, ESG ratings provide valuable inputs into the investment framework and influences portfolio outcomes. For companies, ESG ratings help them understand whether their efforts are achieving the desired ESG impact and outcomes. It is no surprise therefore that the number of third party ESG rating providers has increased significantly in recent years.

This in turn creates other issues which are important for investors to understand.

Unlike financial ratios, which are largely standardised and where there is clarity over what each ratio measures, there is yet to be a global taxonomy for ESG data (although the EU published a Sustainable Finance Taxonomy Regulation in June 2020). As such, there are many ways to report the same issue which ESG ratings providers need to interpret and compare. Just in the area of diversity for example, meaningful metrics extend beyond gender and can extend to include race, nationality, educational attainment, age, disability, sexual orientation and employment status. At the same time, the unique historical, political, and cultural environments of different countries and regions determine which diversity issues are more relevant. For example, while race may be a predominant diversity issue in the United States, multiculturalism is more important in Western European countries with a large proportion of migrants. In China however, the urban versus rural divide may be more relevant while religion and caste could be more meaningful diversity metrics to measure in India².

ESG ratings are also impacted by how ESG rating providers define peer groups. Most ESG rating providers do not provide transparency over how they define peer groups, even though the peer groups have a significant impact on companies’ ESG ratings. In Asia, there is an added complexity for conglomerates with diversified operations and businesses across multiple countries.

Meanwhile, ESG rating providers have different approaches to fill data gaps. One provider may assume that missing incidents data for example, implies a clean safety record. Another may accord a higher score to a company operating in a more regulated country, even if data was missing. For example, in the event of missing data, a large bank in China may be naturally given a lower governance score versus a bank in Singapore. Yet another approach provides an estimate of the missing data by regressing other observable data and predictive factors³.

ESG ratings may also be biased against the smaller capitalised companies which may not have the resources to provide comprehensive reports on ESG issues. At the same time, unlike financial data, the unaudited nature of ESG data also raises concerns over the reliability and usability of the reported data.

This explains why there is significantly greater variance in ESG ratings (due to the reasons listed above) compared to credit ratings (which are based on standardised financial ratios). It is estimated that there is a positive 0.9 correlation between the credit ratings of Moody’s and Standard & Poor’s (S&P)⁴. The same cannot be said for ESG ratings. For example, Sustainalytics rates an Indian company, Adani Ports and Special Economic Zone, as having low ESG risk while MSCI assigned the worst ESG rating of CCC to the company. On the other hand, the company has similar credit ratings of BBB- by Moody’s and Fitch and BB+ by S&P.

All these factors, plus the fact that ESG rating providers currently only have about a 60 to 70% coverage of the investment universe in Asia, highlight Asia’s ESG data challenges.

Fig. 2. Factors that can affect an ESG rating

ESG data standards will continue to be refined and harmonised. Over in Asia, the Asia Securities Industry and Financial Markets Association (ASIFMA) is leading efforts to address ESG data deficiencies and have made the following recommendations⁵:

• Greater convergence towards a principles-based global (or at least regional) taxonomy
• Higher, more consistent corporate disclosure standards between jurisdictions and sectors
• Higher standards of analysis, with incentives for more holistic and robust approaches to ESG measurement and analysis
• Higher standards and accountability for ESG rating providers, potentially including regulation, as well as clear and harmonised requirements for product disclosure
• Harmonisation between ESG standards and frameworks such as UN SDGs, and policy on climate change and bank supervision at the systemic level, including climate risk
• Ongoing partnership and dialogue between the public and private sectors, as well as between stakeholders such as companies and investors on disclosure and reporting standards

Perhaps the solution is not necessarily more data, but agreeing on a set of standardised data. Interestingly, research shows that companies that provide more ESG disclosures tend to have greater variation in their ESG ratings among the different providers⁶.

Against this backdrop, it is key that active managers build their own ESG evaluation framework. Mercer, a company that provides fund ratings and helps asset owners in manager selection, encourages fund managers to have their own formulas to determine ESG weightings and scoring. In addition, Mercer advocates having ESG as a value-added component of the investment decision-making process instead of having a dedicated ESG team⁷. This is very much in line with our team’s approach. The fixed income team based in Singapore has a proprietary ESG evaluation framework which is a natural extension of our credit work and leverages on our credit research team’s deep understanding of issuers in our investment universe.

The proliferation of non-standardised ESG data has partly resulted in incidences of “greenwashing”, a corporate practice of making sustainability claims to cover a questionable environmental record. Despite having a robust Green, Social and Sustainability (GSS) framework and principles, even GSS labelled bonds are not insulated from the risk of greenwashing. This is why our team has not prioritised the inclusion of GSS labelled bonds in our ESG-focused bond strategy, preferring instead to apply our proprietary ESG evaluation framework. We hope to mitigate the risks of greenwashing by performing our own analysis of ESG risks which can impact the long-term viability of a business and its ability to pay debt.

While third party ESG rating providers offer key inputs into the ESG investing process, investors need to understand the factors that affect the ratings.

At the same time, there is no “one size fits all” in ESG data. The different strategies that managers use to integrate ESG in their investment processes (e.g. exclusion, norms-based screening, active ownership and engagement etc), also mean that they will require different types of ESG data⁸.

While Asia navigates data inefficiencies and continues to improve data standards, there is no substitute for old fashioned credit work. After all, active managers have relied on their interpretation of financial data for decades in order to find unique investment opportunities. As such, Asia’s ESG data challenges today potentially provide room for active bond managers to deliver alpha.