³ÉÈË´óƬ

We examine the economic consequences of a controversial process, known as expungement, that allows brokers to remove evidence of financial misconduct from public records. From 2007 through 2016, we identify 6,660 expungement attempts, suggesting that brokers attempt to expunge 12% of the allegations of misconduct reported by customers and firms. Of these attempts, 70% were successful. We show that successful and, to a greater extent, unsuccessful expungement attempts, are a significant predictor of future misconduct. Further, using an instrumental variable based on the random assignment of arbitrators, we show that a broker who receives expungement is more likely to reoffend than a broker denied expungement. This is consistent with the expungement process harming the ability for regulators and consumers to monitor brokers. By contrast, there is only limited evidence that successful expungements improve career prospects. This is consistent with anecdotal evidence that firms ask about expunged infractions during the hiring process, and suggests that expunging a misconduct does not entirely remove the reputational consequences of that misconduct.