How Digital Marketing Can Elevate Fair Lending Risk
On August 17th, HUD filed a housing discrimination complaint against Facebook. HUD claims Facebook allows housing-related advertisers to use unlawful preferences and limit housing options through “targeted advertising”. This results in Facebook enabling advertisers to control which users receive ads based on race, color, religion, gender, familial status, national origin, and/or disability — all of which are prohibited under the Fair Housing Act.
Identifying and managing fair lending risk associated with digital marketing and the increased use of artificial intelligence continues to evolve. HUD’s complaint is an indicator of the potential risks associated with the use of digital marketing. The Federal Reserve’s Consumer Compliance Outlook publication, Keeping Fintech Fair: Thinking about Fair Lending and UDAP Risks demonstrates these regulatory concerns.
How Digital Marketing Increases Fair Lending RiskDigital marketing is becoming one of the more effective and utilized forms of advertising because it allows companies to target potential customers with heretofore unseen precision. By using data and analytical tools, marketers can define who will see ads based on characteristics such as interests, gender, occupation, income, education level, hobbies, shopping habits, etc. The more targeted a marketing campaign, the better the results. In fact, "78% of consumers say personally relevant content increases their purchase intent” (Marketing Insider Group).
The following image provides a brief overview of how marketers can define who sees their ads:
Prohibited factors available for ad targeting include age and gender. Marketers often analyze customer data to determine characteristics shared by existing customers and market to a similar customer profile.
For example, lenders could market mortgages to couples, ages 30-40 (prime homebuying years), who are college graduates with incomes over $70,000, who reside in a selected geography.
Fair lending risk associated with the use of prohibited bases or implied prohibited factors is not a new concept. Selecting zip codes and/or geographies for targeted marketing increases the risk of redlining and/or disparate impact.
Both the Fair Housing Act and the Equal Credit Opportunity Act (ECOA) prohibit discrimination based on race, color, religion, national origin, gender, and familial/marital status. The ECOA additionally prohibits discrimination based on age, source of income, and the exercise of rights under the Consumer Credit Protection Act, while the FHA includes prohibitions for the handicapped/disabled.
So, it is easy to see how marketers may unintentionally violate fair lending laws with targeted marketing campaigns that rely on a consumer’s interests or behaviors on social media.
How to Manage Fair Lending Risk in MarketingDetermining the best way to assess fair lending risk associated with marketing should be the same whether traditional or digital marketing efforts are utilized. The same prohibitions apply regardless of the method of advertising. Prohibited bases cannot be used to target potential customers. Geographic targets should be reviewed to ensure no redlining risk exists in the selection.
Compliance and/or fair lending officers should review marketing initiatives to ensure fair lending risk is addressed before campaigns are finalized.
Risk management efforts should include marketing policies that outline acceptable and unacceptable criteria, fair lending training for the marketing department, and compliance approval requirements. Through careful analysis, review, and documentation, lenders can demonstrate efforts to provide equal access to credit through oversight of marketing efforts, including the use of digital marketing, to regulators
Recommended Reading: 4 Key Risks Observed by the OCC for 2019.
Institutions may utilize mapping software to identify potential gaps in marketing efforts. It is likely that regulators will provide additional guidance and/or update regulations to address risks associated with the changing technological landscape in the near future.