Materiality and Corporate Sustainability : A Model Uncertainty Analysis

For more than thirty years, scholars have investigated the connection between corporate sustainability and financial performance. In 2016, Khan, Serafeim, and Yoon published what appeared to be a major breakthrough in this quest. They reported that guidance from the Sustainability Accounting Standards Board (SASB) enabled the formation of scales of sustainability measures that robustly predicted stock returns. Today, their publication is widely interpreted as evidencing a real connection between corporate sustainability and financial performance, as well as demonstrating the practicality of determining, in advance, which elements of sustainability are material for investors. Yet, for all of its influence, the study remains, in its authors’ own words, just “first evidence”, and thus provides an uncertain basis for inference. In this article, we further explore the relationship between material-sustainability and stock return by performing a “model uncertainty analysis”. We show that we can replicate the published 2016 estimate, but also reveal that it is unrepresentative of results from other reasonable models, and there is strong evidence to support the inference that it is a statistical artifact. We then use machine learning to test whether historical associations between sustainability measures and stock return are predictive of returns in a holdout sample, and we find that they are not. In total, our analysis suggests that for one popular source of data on corporate sustainability, accurate guidance on materiality may be difficult to achieve