As the Financial Industry Regulatory Authority pushes its BrokerCheck database as a resource for investors to check out a broker’s disciplinary history before investing, a big question arises: Can all that data even predict whether a broker will do investors wrong?

To a “significant” degree, it can, two FINRA economists found in a study.

Hammad Qureshi and Jonathan Sokobin argue in their working paper, “Do Investors Have Valuable Information About Brokers?,” that the information available to investors through the online tool BrokerCheck — which includes disciplinary records, financial disclosures and employment history — has “significant power” to separate brokers associated with “investor harm events” from other brokers.

In their working paper, the two FINRA economists sought to answer two questions: Do investors have access to valuable information about brokers through BrokerCheck today? Would adding other nonpublic information filed in the Central Registration Depository, such as test scores, enhance the value of BrokerCheck to investors?

The questions are crucial ones as FINRA pushes to expand investor use of BrokerCheck. In October, the SEC approved a rule to require broker-dealers to link to the service on their websites.

Critics have argued that investors don’t understand the context of the information presented in BrokerCheck. For example, a broker may be the target of a complaint if the market tanks and a client loses money. The FINRA economists controlled for market direction in the study.

More than 29 million broker searches were conducted on BrokerCheck in 2014, with approximately 18.9 million summary records viewed and approximately 7 million downloads of detailed reports on brokers, the FINRA economists state.

The economists surveyed annually the activities and disciplinary histories of 181,133 brokers registered with FINRA from 2000 and 2013. The review tracks brokers who registered with FINRA in 2000 or later and looks at their information since their first registration, including data publicly released on BrokerCheck as well as other nonpublic CRD data.

The economists note that they gauge “investor harm” via complaints filed by customers against their brokers and their subsequent outcomes. “Since some customer complaints may lack merit or suitable evidence of investor harm, we only count complaints that led to awards against brokers or settled above a de minimis threshold,” the two write. “This allows us to focus our analysis on outcomes that are likely associated with material investor harm.”

Less than 1.5% of brokers in the sample met that definition of being associated with investor harm, the paper said.

The 20% of brokers with the highest probability of investor harm were associated with more than 55% of investor harm events as well as total dollar harm, the two write.

Brokers who settled an investor complaint had a greater propensity to harm investors in the following year, the economists found. Other traits associated with investor harm included associations with expelled firms, being dually registered or having a large number of previous employers. Male brokers had a greater propensity to harm investors than female brokers. The economists stress that the predictive value of these traits doesn’t mean the traits cause investor harm.

BrokerCheck does not include certain CRD information about brokers, such as some financial events and performance on qualification examinations such as the FINRA Series 7, the two note.

Investor advocates have called on FINRA to add information about brokers’ exam scores and the number of times they take the exams to BrokerCheck. FINRA has said it would not do so.

Brokers’ average exam scores are negatively associated with investor harm, “but there is no statistically significant association between the number of times a broker failed the exams and investor harm,” the economists write.

More importantly, they argue, “including exam performance generally leads to a reduction in predictive power.”

— Check out Stock Jocks Get Their Knocks at BDs on ThinkAdvisor.