The performance impact of traditional social screens appears negligible

June 13, 2012 | 2 min read

The authors find that unexplained returns are not statistically different from zero in either the full time period or in either of the two subperiods (January 1992–November 1999 and December 1999–June 2010) which coincide with periods of nominal outperformance and underperformance by the index. This is beneficial for investors motivated by social values because it suggests that the risk exposures created by social screens can be managed through careful portfolio construction. It is less encouraging for investors seeking a performance advantage through the use of social or environmental factors — the analysis suggests that, for this universe at least, market valuations already correctly incorporate this information.

In Kurtz’s and diBartolomeo’s earlier studies of this index, in 1996¹ and 1999², they found that a conventional multifactor model explained virtually all of the variation in its performance relative to the intentionally similar S&P 500 Index. They concluded that corporate social responsibility had been a free good and that social investors could therefore achieve competitive returns by using modern portfolio construction techniques (i.e., optimization) to reduce tracking error to acceptable levels. They returned to the topic because several recent studies, notably Edmans and Guenster et al, offered strong evidence that some social factors have been beneficial for portfolio performance.

To ensure their earlier results were not an artifact of the market-capitalization weighting of both the KLD400 and S&P 500, Kurtz and diBartolomeo repeated the entire attribution study and equally weighted the stocks within both indices. The results were essentially identical with the KLD400 having a return advantage of 8 bps per month on average, but this difference was not statistically significant. The stock-specific portion of return that might be attributed to the existence of a social factor was only 2 bps per month, and again not statistically significant. On an equal-weighted basis, the return benefit of tobacco stocks was negligible.

Kurtz and diBartolomeo contend that “an alpha-seeking social investor must be disappointed that a purpose-built social index has no observable alpha over a long time period.” They suggest that alpha-seeking investors must also face the fact that the field is getting crowded. They believe this is what the future will look like for those seeking ESG alpha: effects will be identified in relatively narrow issue areas, away from widely studied datasets, and may or may not coincide with the areas that are of concern to values-driven investors. According to them, this suggests that divergences in practice between investors driven by social values and alpha-seeking social investors will increase over time.

References

¹ 1996 : Lloyd Kurtz and Dan diBartolomeo: “Socially Screened Portfolios: An Attribution Analysis of Relative Performance”.

² 1999 : Dan diBartolomeo and Lloyd Kurtz “Managing Risk Exposures of Socially Screened Accounts”.

This article may be reproduced according to our terms of use with attribution (and link, if online) to www.tias.edu. To be cited as: “The performance impact of traditional social screens appears negligible”, Lloyd Kurtz, Dan DiBartolomeo, www.tias.edu, June 13, 2012.

Research Paper