A new paper from Colleen Honigsberg and Matthew Jacob sheds light on how to think about FINRA's controversial expungement process.  As I explained in a post a couple years back, FINRA's expungement process leaves its own paper trail behind, making it possible for firms and more sophisticated individuals to see whether a broker has scrubbed complaints from their record:

You should also know that these studies work off compromised databases.  BrokerCheck only shows a partial picture because many financial advisers have managed to have complaints expunged from their records.  In instances where an investor settles an arbitration claim against a financial adviser, FINRA arbitrators routinely agree to expunge the existence of the complaint from the public record.  One study found that FINRA arbitrators granted 90% of these requests for expungement.  In some instances, state regulators have even struggled to block the expungement of complaints from the public record.

Like Harry Potter's Mad-Eye Moody, you too can see the invisible.  You can uncover whether (and possibly how many times) a financial adviser has used this process to scrub records by checking a different database.  FINRA makes its arbitration awards available.  While complaints may not show up on BrokerCheck, you may find whether a financial adviser has had complaints expunged by running their name or registration number through this awards database.  The arbitration award granting expungement still shows up in that database.  Wizardry.  

 Of course, some financial advisers may have used the expungement process to remove truly frivolous complaints.  Still, you probably want to know if a particular financial adviser regularly attracts complaints from disgruntled customers. 

Honigsberg and Jacob examined the expungement process and looked to see what having received an expungment meant about a broker.  In theory, expungment is an extraordinary remedy to remove baseless complaints from a broker's record.  If the process works correctly, a broker that receives an expungement would probably be about as likely as a broker without any complaints to have a future complaint.  (Unless, of course, there is some other reason why a subset of brokers keep attracting baseless complaints.) On the other hand, if the process is scrubbing complaints truly generated by broker behavior–that is to say complaints that would warn other investors that they might have a problem with a broker in the future–brokers that have their slates wiped clean will be more likely to attract a new complaint than the average broker with a clean record.

The paper seems to indicate that the process now suppresses the signal sent by these complaints.  Removing them makes it harder for customers to use the BrokerCheck database to protect themselves.  The paper has some interesting findings–including "that removing this public information increases recidivism."  Fascinatingly, brokers that successfully expunge complaints from their record may be "more likely to reoffend than a broker denied expungement."  Honigsberg and Jacob point out that this might be explained by the behavioral economics literature indicating that success can breed overconfidence and excessive risk taking.  If a broker has been able to suppress a complaint in  the past, she might be a bit more aggressive in the future because she might believe she could suppress a future complaint as well.

Congratulations to Honigsberg and Jacob for this contribution.  The Wall Street Journal picked up on it already.  Hopefully it'll also end up on the desks of regulators at the SEC and FINRA thinking about how to reform the expungement process.