Log in     Support     Status

How to Improve Racial Bias in Mortgage Lending

by | Jul 22, 2021

Loan discrimination or discrimination in lending is nothing new. The Fair Housing Act of 1968 outlawed discrimination concerning the sale, rental, and financing of housing based on race, religion, national origin, or sex. However, despite this civil rights victory, Black and Latino homebuyers have continued to struggle to access fair lending practices. Now, with the U.S. housing boom and the uneven impact of the pandemic on minorities, the disparity in lending opportunities has become even more prevalent. Minority applicants face higher interest rates and fees and are more likely to have their loan denied.

One way that banks and other mortgage lenders can reduce bias is to use AI-powered mortgage support automation platforms such as Capacity.

How can AI tools help to reduce discrimination?

Laws such as the Fair Housing Act and the Equal Credit Opportunity Act have attempted to end discrimination in lending. However, humans have conscious and unconscious racial biases that are impossible to eradicate with the rule of law.

Even when a lender implements diversity training, a mortgage application from a member of a minority group is less likely to succeed.

AI tools, on the other hand, can be trained to evaluate a loan application based on purely objective criteria like the credit history or income of borrowers. This approach helps banks and other lenders eliminate racial biases or other personal discrimination in the process.

The Financial Services Innovation Coalition (FSIC) and Creative Investment Research recently released a report that covered the impact of artificial intelligence (AI) and algorithmic lending on reducing racial disparities in mortgage lending. The report found that AI-based systems reduce bias on the basis of race by 40 percent.

AI lending platforms evaluate loan applicants as a whole and not on an individual basis. This makes it easier for mortgage lenders to comply with the Equal Credit Opportunity Act.

Mortgage companies can also use AI to identify and correct patterns of historic discrimination against mortgage applicants. That means training and testing AI systems not merely on an applicant’s credit report or credit score but other factors like background or minority status. AI can combine these factors to create a system that analyzes disparities in lending on a deeper level than conventional systems. Then, lenders can use AI systems to adjust their mortgage application processes.

Can AI loan tools be biased?

AI tools can reflect the biases of programmers and users as well as historical biases against particular groups of borrowers. After all, any tool can only be as unbiased as the data used to train and test it. That’s why at Capacity, we are committed to the responsible use of AI. By focusing on diverse data sets and trainers with different backgrounds, you can reduce biases.

To learn more, read our  7 guiding principles around how we put AI into practice. > 

The benefits for borrowers.  

AI tools can help lenders transparently evaluate applications and offer better loans. When bias and other personal prejudices are eliminated, minority applicants will have a real chance to take out mortgages that they otherwise could not afford. This will help minority applicants to feel more confident when buying a new home.

The benefits to the economy.  

Minorities tend to be more concentrated in areas of distressed economic conditions. The disparity of unfair lending practices can contribute to unstable communities and overall economic hardship. Integrating AI into lending procedures helps to boost economic growth in poor neighborhoods and also helps the U.S. economy as a whole.

Other benefits of AI tools in mortgage lending.

In addition to racial bias reduction, using AI tools in mortgage lending can help lenders with other issues.

Due to the complexity of the process and the number of people involved, manual underwriting is a time-consuming process that delays lending. An automated support system allows for simultaneous approval by multiple parties and an objective review of the application. This approach results in a more streamlined version of the loan application process. 

This helps both minority and non-minority borrowers get loans approved faster. It also decreases costs for lenders.

AI tools like Capacity also improve the borrower experience by providing 24/7 support via a chatbot that links to key systems and learns on the job.

What this means for mortgage companies.

Mortgage institutions must work diligently to ensure fair treatment so all customers get an equal opportunity at creating wealth.

Artificially intelligent systems can make great strides toward reducing discrimination in lending without the need for compliance litigation. By empowering loan officers to run automated and algorithmically based mortgage applications, lenders can find increasingly successful revenue streams while reducing bias.

How can mortgage companies implement AI?

Capacity helps financial institutions refinance and create loans faster and at a lower cost. We’ve built a complete support automation platform to automate the underwriting process and free up loan officers to do higher-value work such as creating relationships with borrowers.