The Federal Trade Commission has released a report examining the benefits, potential risks, and legality of the use of big data in business.

Big Data: A Tool for Inclusion or Exclusion? Understanding the Issues focuses on how big data is used after it is collected and how that information could result in discrimination against consumers.

The primary goal of the report is to provide businesses with important information on the relevant laws to big data analytics, as well as guidelines on how to use big data effectively while remaining compliant and non-discriminatory, according to the FTC.

“Big data’s role is growing in nearly every area of business … The potential benefits to consumers are significant, but businesses must ensure that their big data use does not lead to harmful exclusion or discrimination,” said FTC Chairwoman Edith Ramirez.

Some of the potential beneficial uses include providing access to credit through non-traditional methods, better access to employment to underserved populations, increased educational attainment and specialized healthcare.

However, the report also emphasizes how big data use could yield biases or inaccuracies about certain groups. Discrimination from data usage could result in assisting in the targeting of vulnerable consumers for fraud, creating higher prices for goods and services in lower-income communities and weakening the effectiveness of consumer choice.

Laws pertaining to the use of big data include the Fair Credit Reporting Act, the Equal Credit Opportunity Act and the Federal Trade Commission Act. The report provides guidance for each.

Fair Credit Reporting Act

The Fair Credit Reporting Act applies to consumer reporting agencies that compile and sell consumer reports containing consumers’ information intended to be used to determine consumers’ eligibility for credit, employment, insurance and housing, among others.

As of late, there has been a trend toward using predictive analysis to determine eligibility, rather than the traditional credit scoring method. The traditional method compares known credit characteristics of a consumer to historical data that shows how people with the same characteristics met their credit responsibilities.

However, while predictive analysis includes measures of credit characteristics, non-traditional characteristics such as zip code, shopping history and social media use may also be used to make decisions about a consumer’s eligibility. As of the report, the FTC has established that “only a fact specific analysis” will be able to determine if a practice is subject to or violates the FCRA.

Thus, companies should be aware of the FCRA when using predictive analysis or other methods that use big data, especially non-traditional characteristics, when making eligibility determinations that might be subject to the FCRA.

Equal Opportunity Laws

While there are many equal opportunity laws that prohibit discrimination based on protected characteristics such as race, color, sex or gender, religion and more, the law most relevant to this report is the Equal Credit Opportunity Act, which the FTC enforces. ECOA prohibits credit discrimination against protected characteristics, and a violation can arise when a plaintiff can show that there was disparate treatment or impact; either the lender treated an applicant differently based on a protected characteristic or the company employs “factually neutral” characteristics that have a disproportionate or adverse impact on a protected class.

The report expresses concerns that advertising and marketing practices could produce problematic lending patterns that could be subject to action. More specifically, the report points to Regulation B, the implementing regulation for ECOA that is in part concerned with what criteria creditors use to select potential recipients for prescreened solicitations.

With regard to big data, if it is being used to determine advertising or marketing, or other practices that could result in disparate impact, lenders using big-data, non-traditional characteristics should be mindful of their relationship to equal opportunity laws. While determination of the lawfulness of a practice is currently case specific, according to the FTC, in some cases the Department of Justice has pointed to a creditor’s advertising choices as evidence of discrimination.

The Federal Trade Commission Act

The report is also concerned with the use of big data analytics in relation to Section 5 of the Federal Trade Commission Act, which prohibits unfair or deceptive acts or practices. The report recommends that companies should consider:

  • whether they are violating any material promises to consumers
  • whether they have failed to disclose material information to consumers

It is also recommends that companies should “take care to reasonably secure consumers’ data,” and at a minimum, companies “must not sell their big data analytics products to customers if they know or have reason to know that those customers will use the products for fraudulent or discriminatory purposes.”

In possible instances of alleged violation, cases will be fact specific, and will be concerned with whether the company engaged with big data in a deceptive or unfair way.

Four Questions Companies Should Consider

The report recommends that companies using big data consider the following four questions to help efforts to maximize the benefits and limit the harms of big data analytics:

  • How representative is your data set?
  • Does your data model account for biases?
  • How accurate are your predictions based on big data?
  • Does your reliance on big data raise ethical or fairness concerns?

In a separate statement accompanying the release of the report, Commissioner Maureen Ohlhausen’s spoke to the issues raised by the report in the context of free-market competition:

“Concerns about the effects of inaccurate data are certainly legitimate, but policymakers must evaluate such concerns in the larger context of the market and economic forces companies face … to the extent that companies today misunderstand members of low-income, disadvantaged, or vulnerable populations, big data analytics combined with a competitive market may well resolve these misunderstandings rather than perpetuate them.”

The Commission voted to issue the report 4-0.