Search Results
Discussion Paper
Quantifying Cyber Risk in the Financial Services Industry
The Consumer Finance Institute hosted a workshop in February 2017 featuring James Fox, partner and principal at PricewaterhouseCoopers (PwC) and a leading authority on cybersecurity in the financial services industry. He discussed the importance of measuring cyber risk, highlighted some challenges that financial institutions face in measuring cyber risk, and assessed several leading cyber-risk management methodologies. Fox also provided some recommendations for bank exams and insights into how federal agencies might begin to quantify systemic cyber risk. This paper summarizes Fox?s ...
Working Paper
One Threshold Doesn’t Fit All: Tailoring Machine Learning Predictions of Consumer Default for Lower-Income Areas
Modeling advances create credit scores that predict default better overall, but raise concerns about their effect on protected groups. Focusing on low- and moderate-income (LMI) areas, we use an approach from the Fairness in Machine Learning literature — fairness constraints via group-specific prediction thresholds — and show that gaps in true positive rates (% of non-defaulters identified by the model as such) can be significantly reduced if separate thresholds can be chosen for non-LMI and LMI tracts. However, the reduction isn’t free as more defaulters are classified as good risks, ...
Discussion Paper
Can Data Sharing Help Financial Institutions Improve the Financial Health of Older Americans?
This paper explores how increased data sharing among financial institutions could improve the financial outcomes of older adults suffering from cognitive impairment. Among the first signs of cognitive impairment in older adults is a decline in financial capacity, which is also a risk factor for abuse or exploitation. Banks and other financial institutions are at the front lines to monitor and detect changes in financial capacity and susceptibility to fraud and abuse. However, industry experts have found that, in many cases, no mechanism exists for financial service providers to communicate ...
Report
How Has the COVID-19 Pandemic Affected the Supply of Consumer Credit?
For credit card companies and other lenders, the pandemic has created a variety of risks and challenges. This report examines how credit card lenders have responded to the COVID-19 pandemic.
Report
Secured Card Market Update
For more than 40 years, secured credit cards have enabled borrowers with limited or damaged credit histories to obtain credit, often with many of the benefits of traditional, unsecured credit cards. Despite the recent development of alternative credit-building products, the secured card has not been left behind and, in fact, continues to be a focal point of fintech innovation and new product offerings. In this special report, we update previous research by the Consumer Finance Institute on secured cards using data from the country’s largest financial institutions. As of September 2023, ...
Discussion Paper
Combining AI and Established Methods for Historical Document Analysis
This paper examines methodological approaches for extracting structured data from large-scale historical document archives, comparing “hyperspecialized” versus “adaptive modular” strategies. Using 56 years of Philadelphia property deeds as a case study, we show the benefits of the adaptive modular approach leveraging optical character recognition (OCR), full-text search, and frontier large language models (LLMs) to identify deeds containing specific restrictive use language— achieving 98% precision and 90–98% recall. Our adaptive modular methodology enables analysis of ...
Report
How Has the COVID-19 Pandemic Affected the Supply of Consumer Credit?
For credit card companies and other lenders, the pandemic has created a variety of risks and challenges. This report examines how credit card lenders have responded to the COVID-19 pandemic.
Discussion Paper
How Prevalent Were Racially Restrictive Covenants in 20th Century Philadelphia? A New Spatial Data Set Provides Answers
One of the tools used by early 20th century developers, builders, and white homeowners to prevent African Americans from accessing parts of the residential real estate market was the racially restrictive covenant. In this paper, we present a newly constructed spatial data set of properties in the city of Philadelphia with deeds that contained a racially restrictive covenant at any time from 1920 to 1932. To date, we have reviewed hundreds of thousands of property deeds and identified nearly 4,000 instances in which a racial covenant had been included in the deed. The covenanted properties ...
Report
Overdraft Use During the Pandemic: Insights from the CFI COVID-19 Survey of Consumers
This research brief examines consumers’ use of checking account overdrafts since the beginning of the COVID-19 crisis, using responses gathered from a special module in the Federal Reserve Bank of Philadelphia’s Consumer Finance Institute COVID-19 Survey of Consumers conducted between July 5 and 16, 2021. The overdraft module includes responses from a national sample of 3,615 consumers, 1,054 of whom reported experiencing at least one overdraft charge during the crisis. Respondents provided their demographic and employment characteristics as part of the broader survey. Survey participants ...
Working Paper
Foreclosure Kids: Examining the Early Adult Credit Usage of Adolescents Affected by Foreclosure
We investigate the long-term effects of foreclosure-induced relocations on adolescents and their subsequent use of credit. We ask whether individuals who experience a foreclosure-induced move between the ages of 10 and 17 are more likely to exhibit signs of credit scarring later in life. To establish a set of counterfactual outcomes, we implement propensity score matching with exact matching on certain characteristics and regression adjustment of the remaining covariate imbalances. We then compare the credit behavior of individuals who experienced a foreclosure-induced move in adolescence to ...