Is fraud contagious? career networks and fraud by financial advisors

We show that the propensity to commit fraud is transmitted through career networks.  We use a novel dataset of U.S. financial advisors, which includes individuals' employment histories and records of fraudulent behavior.  The dataset allows us to identify co-worker networks at the branch level within firms.  To identify the effect of career networks on fraud we use changes in co-workers caused by mergers of financial advisory firms; the tests include merger-firm fixed effects to exploit the variation in changes to career networks across different branches of the same firm.  The probability an advisor commits fraud increases if his new co-workers, encountered in the merger, have a history of fraud.  This result holds even controlling for the adviser’s own history of fraud, the fraudulent behavior of his pre-merger co-workers, individual characteristics such as age, experience, and assets under management, and firm-level effects.  Further, this effect is stronger between demographically similar employees.  Our tests indicate that both similarities in age and in ethnicity are important factors.  In tests of robustness, we examine alternative explanations to the career network hypothesis.  We find that within-firm, cross-branch variation in supervision and oversight and variation in branch-level policies and controls do not explain observed changes in fraudulent behavior.  We find the results are similar for both target firm and acquiring firm advisors. Additionally, we show the results are robust to including merger firm-county fixed effects, effectively reducing the comparison to different branches of the same firm within the same county. Our results provide evidence that social transmission of criminal behavior occurs even within a professional setting. 

Keywords: Financial advisors, Financial misconduct, Fraud, Social networks, Peer effects, Career networks