Abstract

Using a sample of US stocks over the period 1991–2019, we test whether stocks with high exposure to a social index exhibit high returns. Using a univariate analysis, our in‐sample results show that stocks with high sensitivities to the MSCI KLD 400 Social Index underperform stocks with low sensitivities by an annual risk‐adjusted performance of 7.02%. The negative premium is also larger in the post‐crisis period of 2007–2019 and is equal to 10.25%. The out‐of‐sample results offer, however, only weak evidence of such a finding, with a risk‐adjusted performance difference of merely −0.84% over the full sample period and no significant differences between the pre‐crisis and post‐crisis periods. In the multivariate regression, we find evidence of a negative relationship between exposure to the social index and stock performance. Moreover, we find that stocks with high exposure to the social index display a low corporate social responsibility score, a high Tobin’s Q, high long‐term debt, a large size, high total risk, a high market beta, a high SMB coefficient, a low HML coefficient, and a small MOM coefficient.