Abstract:
Lins, Servaes and Tamayo (2017) (LST) use the 2008-2009 global financial crisis (GFC) as an exogenous shock where public trust in corporations, capital markets, and institutions unexpectedly declined. Their main hypothesis is that if a firm’s social capital helps build stakeholder trust and cooperation it should pay off when being trustworthy is more valued, such as in an unexpectedly low-trust period. Consistent with this hypothesis, LST find that during the GFC, stock prices of US companies with high social capital decreased by less than the stock prices of peers with low social capital. We replicate the results in LST, improve on their research design and examine external validity by testing the hypothesis for a sample of Japanese stocks. Their main results are based on panel regressions, which is surprising given that the GFC is a major event that affected all economic entities simultaneously. Since the GFC is a clustered event, abnormal returns to US stocks during the GFC are cross-correlated. Because panel regressions give biased results in this setting, the event study literature advocates the use of calendar-time portfolio regressions. LST also use this approach but only show the results of equally weighted portfolios, without controlling for industry. We propose to replicate this result. We also propose to present the results for value-weighted portfolios, with and without controlling for industry. Our initial results suggest that the results in LST are not robust: the impact of CSR during the financial crisis is concentrated in small firms and in certain industries. Next, we will replicate the tests in Japan which is of interest because Japan i) is an early promoter of CSR disclosure, ii) was severely affected during the 2008 crisis, and iii), according to some research, has a lower level of social trust than the US – making CSR as substitute for social trust more important.
