Letter: ESG studies must show scientific rigour to avoid feeding biases

The new Scientific Beta study (“ESG outperformance narrative ‘is flawed’, new research shows”, ETF Hub, May 3) has caused a splash by showing that common claims of an environmental, social and governance alpha are much more fragile than portrayed.

But their findings will not be surprising to anyone who has analysed the ESG data carefully. Scientific Beta is right that claims of outperformance must control for other factors, and be based on long time horizons. Indeed, this is the role of academic peer review — no top journal should publish a study that failed to do this. In contrast, the papers that Scientific Beta correctly criticises are either unpublished, or in practitioner outlets. However, given confirmation bias — the temptation to believe even flimsy evidence if it supports your preferred viewpoint — many people latched on to these studies, despite their basic errors.

Views on ESG swing from one extreme to the other. One month, there are claims that ESG clearly improves performance; the next, all these studies are said to be flawed. The more sweeping the conclusion, the more headlines it grabs, but reality is somewhere in the middle. Certain ESG factors do pay off, over long time horizons and even after controlling for other variables. But others — particularly those that are not material to a company’s business model — do not. To avoid such black-and-white thinking, we should pay attention to studies based on their scientific rigour, rather than whether they support our personal beliefs on ESG.

Professor Alex Edmans
London W2, UK