the divergence of esg ratings

In our working paper “Aggregate Confusion: The Divergence of ESG Ratings” we disentangle the disagreement between ESG raters into three different sources: scope, weights, and measurement. Scope relates to what you measure (do you include the issue Bribery?), weights to how important an issue is (do you give more weight to CO2e Emissions than Bribery?), and measurement to how you measure an issue. We find that scope and measurement divergence are the main drivers, while weights divergence is less important. Scope and weight divergence is nothing but your personal preferences. Everyone is entitled to their preferences. A quick non-scientific survey at the table I am currently sitting at indicates more different preference sets than people. Moreover, when the conversation started, people had different opinions than at the end of the conversation, i.e., preferences also change over time. The more severe problem is measurement divergence. For example, if we want to measure how women are treated in a company, rating agencies typically resort to proxies such as the gender-pay-gap as one of the data points. However, a gap of zero does not guarantee equal treatment. Measurement divergence indicates different attempts to measure the same issue. When it comes to ESG issues, we often cannot observe the distribution ex-post. For example, six months after measuring the gender-pay-gap, we are not smarter about the treatment of women in the workforce. In this case, there is no way we can improve ESG ratings through backtests.

I invite you to read the paper if you have more questions than answers now!