How to examine external validity within an experiment

A fundamental concern for researchers who analyze and design experiments is that the estimate obtained from the experiment might not be externally valid for other policies of interest. Researchers often attempt to assess external validity by comparing data from an experiment to external data. In this paper, I discuss approaches from the treatment effects literature that researchers can use to begin the examination of external validity internally, within the data from a single experiment. I focus on presenting the approaches simply using stylized examples.     

Fraud Firms' Non-Implicated CFOs: An Investigation of Reputational Contagion and Subsequent Employment Outcomes*

We investigate labor market consequences for CFOs employed by fraud firms, focusing on reputational contagion for those who are not implicated. These individuals provide an opportunity to understand reputational contagion and the nuanced meaning of “guilt” because the labor market may suspect complicity or infer negligence regardless of whether that is truly the case. We compare these CFOs to a matched sample of non-fraud CFOs and track both turnover and subsequent employment positions. Non-implicated CFOs are more likely to experience turnover compared to non-fraud CFOs, driven in particular by the public revelation of fraud to the labor market. We further find that non-implicated CFOs are more likely to obtain comparable subsequent employment than non-fraud CFOs before the fraud is publicly revealed, but not after. In supplementary analyses, we find that turnover rates are highest for non-implicated CFOs who started their employment with the firm before the fraud began as compared to non-implicated CFOs who started their employment after the fraud began. These results highlight the labor market significance of the public revelation of fraud and imply that the labor market does not fully distinguish between fraud firm association and general firm performance when making executive hiring decisions.