Renneboog, Luc, Jenke ter Horst, and Chendi Zhang. “The Price of Ethics: Evidence from Socially Responsible Mutual Funds.” ECGI Working Paper No. 168/2007. European Corporate Governance Institute. May 2007.

This strong and thorough study reviews historical returns of socially responsible mutual funds around the world for the 1991-2003 time period. It is noteworthy not only for its headline findings, some of which are new and important, but also because it examines a number of issues that have not been studied in the past, and present some surprising answers.

The foundation of the study is a returns analysis of 463 live and dead equity mutual funds in “23 countries and offshore jurisdictions around the world.” Although the time period is reasonably long, it should be noted that, due to the recent growth in the industry, the age of the average fund is quite low seven years for U.S. and U.K. funds, and just three years for the “Rest of the World” (hereafter “ROW”). The SRI funds were, on average, smaller than their unscreened counterparts, and the U.S. and U.K. funds were significantly larger that the ROW funds.

The returns analysis is equal-weighted, uses monthly returns, and uses CRSP, Worldscope, and Datastream data. Key findings:

1) Risk adjusted returns (CAPM excess return) for U.S. and U.K. funds are not statistically different from those of their unscreened counterparts. For the ROW SRI funds, however, there is statistically significant underperformance.

2) Alphas derived from a Carhart four-factor model show the U.S. and U.K. firms underperforming their unscreened counterparts (-2.2% and -3.4% vs. -1.1% and -2.5% for the average conventional fund), although, again, this was not statistically significant. The ROW performance disadvantage is again observed, but is less pronounced than in the CAPM analysis.

3) Country-level data (limited to those countries with five or more years of performance) shows “strongly negative” four factor alphas the ROW group. Several countries show alpha of -5% or lower (Belgium, France, Ireland, Norway, Singapore, and Sweden). The authors note that negative alpha of this magnitude is too large to be explained solely by management fees.

Perhaps the most counterintuitive finding is that SRI funds with more restrictive social screens tend to outperform other SRI funds. This was particularly true when the fund benefited from in-house social research capability.

On balance this is a sophisticated and worthwhile study, and one that anyone interested in the performance of SRI mutual funds should be aware of.