FinRegLab is extending its investigation of the adoption of artificial intelligence in financial services through a policy analysis focused on the growing use of machine learning models for underwriting credit and a January 17 webinar with senior federal financial regulators to discuss generative AI and other recent developments.
A new research paper underscores the importance of evaluating whether the most vulnerable and distressed borrowers need longer term repayment plans soon after they first enroll in natural disaster or other emergency relief programs with credit card lenders.
FinRegLab has issued two papers that examine lenders’ ability to build, understand, and manage machine learning models to ensure that they can be trusted to underwrite applications for credit by millions of consumers and small businesses.
A new study finds that more consumers obtained short-term payment relief on their credit cards during the first 18 months of the pandemic than on any other type of loan except student debt, where forbearances were mandated by federal law. The study also finds evidence that pandemic relief initiatives may have reduced damage to the credit reports of consumers who sought long-term assistance through credit counseling and debt management programs.
FinRegLab has launched a new research project using data from the National Foundation for Credit Counseling (NFCC) to evaluate ways to help consumers recover more quickly from personal and economic crises such as COVID-19. The project will analyze pilot initiatives by nonprofit counseling agencies and other data sources as a springboard for considering broader market and policy changes.
Machine learning models are being used to evaluate the creditworthiness of tens of thousands of consumers and small business owners each week in the U.S., increasing the urgency of answering key questions about their performance, governance, and regulation.
FinRegLab is working with researchers from Stanford Graduate School of Business to launch a ground-breaking evaluation of emerging market practices to improve the transparency and fairness of machine learning underwriting models in consumer credit.
Data is being used by both new entrants and traditional lenders to extend smaller loans to smaller businesses and increase credit to underserved communities.
This report analyzes mortgage default using information taken from the JPMorgan Chase Institute housing finance research to evaluate the relationship between liquidity, equity, income level, and payment burden and default. Across all four groups, the report finds that liquidity may be more predictive for determining the likelihood of mortgage default particularly among borrowers with little post-closing liquidity and little liquidity but high equity. Overall, the report determines that alternative underwriting standards incorporating a minimum amount of post-closing liquidity may be a more effective way to prevent mortgage default compared to using DTI thresholds at origination.