Economic theory predicts that, in the absence of mispricing, investing in socially responsible businesses should have lower expected returns in equilibrium. In contrast, in an influential paper, Edmans (2011) shows that a portfolio of the “100 Best Companies to Work For in America” (BCs) earns a positive and significant Carhart four-factor alpha. More than one decade later, we ask whether this result (a) holds out-of-sample, (b) is driven by exposure to newly discovered factors and characteristics, and (c) depends on the state of the economy. We find that up to 22% of the documented BC portfolio’s Carhart alpha could be attributed to exposures to more recently discovered factors such as investment, profitability, and quality. Nevertheless, using the state-of-art factor models and a sample from the period 1984 to 2020, an equal-weighted BC portfolio earns an abnormal return of 2% to 2.7% per year. The abnormal returns are not driven by firm characteristics, industry composition, or micro-cap stocks. The estimated alphas are positive in almost all periods within our sample (with no upward or downward trend) and are particularly large in “bad” times such as in the crisis periods of 2000-2002 and 2008-2009. Overall, our results suggest that the stock market still undervalues employee satisfaction, which seems to have the greatest value in “bad” times. We conclude with proposing potential reasons behind the (surprising) persistent outperformance of BCs.