Research Exchange: June 2024

Selected Outside Research

Of Data, Limits and Models: A Note on “The Limits of Model-Based Regulation”

The Basel II IRB (Internal Ratings Based) capital framework, which determines capital requirements based on banks internal ratings or risk parameter quantifications, has been criticized based on the view that the approach is overly reliant on banks’ providing accurate risk assessments. One major study offering evidence in support of this view found that the introduction of the IRB framework in Germany in 2007 led affected banks to systematically underreport risk in their IRB portfolios and thus economize on regulatory capital. This paper critiques that particular study and finds the evidence presented in it be lacking, in that its analysis contains two serious flaws: the sample for the analysis is defined incorrectly and excludes relevant observations, and the empirical model used to test for downward bias in the risk parameter calibrations is mis-specified. Thus, this paper finds that the notion of underreporting credit risk via the IRB approach is unsubstantiated. Rather, it suggests that that the rigorous supervisory standards and validation processes inherent in the internal ratings based approach ensure its integrity and validity in evaluating credit risk specifically.

Of Data, Limits and Models: A Note on ‘The Limits of Model-Based Regulation’ by Stefan Blochwitz, Christine Fremdt, Jürgen Prahl, Marcus Pramor :: SSRN

Government Litigation Risk and the Decline in Low-Income Mortgage Lending

The US Department of Justice’s suit against prominent lenders associated with the Federal Housing Administration mortgage insurance program alleging fraudulent activities during the mortgage crisis period resulted in settlements amounting to approximately $5 billion. The targeted banks and many of their peers subsequently withdrew from the FHA loan market. This paper investigates the effect of their exit on credit supply in geographic areas where they previously had a significant presence, employing  a “difference-in-difference” empirical approach. The analysis indicates that their exit led to a 20% reduction in FHA lending in these areas, despite expansion by nonbanks in place of the exiting banks. Moreover, the contraction in FHA credit supply was “not associated with improved underwriting standards or lower default rates” and  was a salient factor in the share of home purchase mortgages for low-income borrowers dropping by nearly 50%, decreasing from 11% in 2009 to 6% in 2017, The study concludes that “large banks’ FHA exit has significantly reduced low-income households’ overall access to mortgage credit”.

Government Litigation Risk and the Decline in Low-Income Mortgage Lending by W. Scott Frame, Kristopher Gerardi, Erik J. Mayer, Billy Xu, Lawrence Chengzhi Zhao :: SSRN

De-Risking Banks Through Synthetic Securitization and Credit-Linked Note Issuance

U.S. banks are increasingly using synthetic securitization in the form of credit-linked notes to transfer credit risk to outside investors. This post summarizes the structure of these transactions, discusses the underlying economic motivations, and compares them to similar types of credit risk sharing activities undertaken in the secondary mortgage market by Fannie Mae and Freddie Mac. The two main vehicles for credit linked note transactions are direct issuance of the notes by the banks to the investors and issuance through a special purpose entity, where the bank purchases loss protection from the SPE, while the SPE itself issues the credit linked notes. In the case of direct issuance, the issuing bank is exposed to counterparty risk; whereas in the case of an SPE, counterparty risk is substantively addressed and mitigated. The discussion highlights how the transfer of credit risk to external investors through banks issuing credit linked notes reduces banks’ regulatory capital requirements, bringing closer alignment between their regulatory capital requirements and internal capital allocations.

De-Risking Banks through Synthetic Securitization and Credit-Linked Note Issuance by W. Scott Frame :: SSRN

Sending Out an SMS: Automatic Enrollment Experiments for Overdraft Alerts

This paper describes a field experiments conducted at major UK banks studying the effects of automatic enrollment into “just-in-time” text message alerts and other types of “early warning” alerts designed to prevent the occurrence of overdrafts. The just-in-time messages were found to reduce the incidence of overdraft charges by 4-8% on accounts with pre-arranged overdraft protection (a line of credit with a borrowing limit) and by 17-19% on accounts without pre-arranged protection, implying a substantial annual market-wide savings of £170–240 million. Incremental benefits from additional “early warning” alerts triggered by low account balances were not statistically significant, although not ruled out. These findings are consistent with findings from previous research that overdrafts could often be avoided “by using lower-cost liquidity available in savings and credit card accounts, and they suggest that “default enrollment” in such an alerts program would help consumers achieve some of these potential savings.

Sending Out an Sms: Automatic Enrollment Experiments for Overdraft Alerts by Michael D. Grubb, Darragh Kelly, Jeroen Nieboer, Matthew Osborne, Jonathan Shaw :: SSRN

Can Discount Window Stigma Be Cured?

Historically, the Federal Reserve’s role as “lender of last resort” has been impeded by banks’ reluctance to borrow from the Federal Reserve’s discount window, even for benign reasons, out of concerns that it could be interpreted as a sign of financial weakness, a phenomenon known as “discount window stigma.” This post addresses the effectiveness of alternative strategies to counter stigma and conclude that pre-existing stigma is difficult to break.  Specifically, the post describes the results of an experimental game in which participants played the role of investors making decisions whether to invest in a bank in part based on the bank’s history of discount window borrowing. The results indicate that, while requiring banks to borrow from the discount window at random increments may prevent stigma formation, neither this strategy nor reducing the cost of borrowing to zero are adequately effective at curing pre-existing stigma. Finally, the post suggests that it may be preferable for the Federal Reserve to focus on effective alternatives to discount window borrowing. These could include the introduction of temporary emergency facilities when liquidity markets become severely strained (similar to the Term Auction Facility introduced at the onset of the Global Financial Crisis in December 2007, or the Bank Term Funding Program in March 2023) or new, permanent backstop facilities designed to be “stigma-proof”. An example of the latter would be an overnight facility providing liquidity while also having high quality collateral.

Can Discount Window Stigma Be Cured?  – Liberty Street Economics (

The Growing Risk of Spillovers and Spillbacks in the Bank-NBFI Nexus

This post discusses the increasing interconnectedness via portfolios, funding sources, and liabilities between bank and non-bank financial entities and its systemic implications. During times of stability, bank entities are able to offload risk exposures to non-bank financial entities. However, during times of stress, spiking demand for liquidity from non-bank financial entities will cascade towards the banking sector so that, in effect, “bank-NBFI dependencies turn into vectors of shock transmission and amplification”. The post provides evidence of increasing interdependence between banks and NBFI’s with systemic risk implications, supporting the view that “effective financial regulation and systemic risk surveillance requires a holistic approach that recognizes banks and NBFIs as highly interdependent sectors”.

The Growing Risk of Spillovers and Spillbacks in the Bank-NBFI Nexus – Liberty Street Economics (

Negative Equity Findings From the Auto Finance Data Pilot

In February 2023, the Consumer Financial Protection Bureau launched a data gathering initiative, called the “auto finance data pilot”, seeking to “better understand loan attributes that may result in increased consumer distress” in the auto finance market. For the pilot study, the CFPB issued nine market monitoring orders to three banks, three finance companies, and three captive lenders to provide information about their auto lending portfolios. This report, which is the first of a planned series using data from the pilot collection, provides a detailed analysis of lending characterized by negative equity, where the trade-in value offered for a consumer’s vehicle is less than the outstanding loan balance and the unpaid balance is rolled into the new loan. This form of financing “can place consumers further underwater on their next loan which, for example, may increase the risk of a deficiency balance if the consumer cannot repay the loan”. Among multiple findings of interest are that more than 10% of borrowers financed negative equity from a prior vehicle loan into a new loan; and that these borrowers financed larger loans with higher loan-to-value and payment-to-income ratios. They also exhibited a significantly higher likelihood of having their account assigned to repossession within two years. Overall, the data suggest that “negative equity may lead to worse consumer outcomes”.

Negative Equity in Auto Lending (

Paying Too Much? Borrower Sophistication and Overpayment in the Mortgage Market

This paper compares mortgage rates that borrowers obtain to rates that lenders could offer for the same loan, using  data sourced from a major industry platform that connects mortgage lenders with whole loan investors and provides detailed information on mortgage pricing. On the originator side, this platform is used primarily by nonbank lenders and secondarily by community banks and credit unions. The analysis finds that many homeowners significantly overpay for their mortgage from these originators, with overpayment varying across borrower types and with market interest rates. Overpayment is most pronounced among FHA borrowers, for whom the study finds a 72 basis point difference between the 90th and 10th percentile interest rate that observably identical borrowers lock in for the same loan in the same market on the same day. In addition, the paper presents evidence, based in part on analysis of data from the National Survey of Mortgage Originations (a program of the Federal Housing Finance Agency), which ties overpayment to inadequate shopping behavior or negotiating effort on the part of borrowers.

Paying Too Much? Borrower Sophistication and Overpayment in the U.S. Mortgage Market (

Chart of the Month

The chart depicts year-over-year changes in bank-specific capital requirements, as implied by comparing stress test outcomes between last year and the current year. Banks are ordered from left to right in the chart based on the change in their stress capital buffer, with the leftmost bank showing the largest reduction in the stress capital buffer compared with the previous test and the rightmost the largest increase. The results show significant dispersion across banks—over half of the banks face an increase in their SCB requirement and about one third of the banks face an increase of 70 basis points or more. This variability underscores the uncertainties banks face due to lack of transparency around the factors driving the changes in their capital requirements.

Featured BPI Research

BPI’s Stress Testing Testimony and 2024 Stress Test Results

On June 26, Francisco Covas from BPI testified at a House Financial Services Committee subcommittee hearing to address concerns about the lack of transparency in the Federal Reserve’s stress tests and its impact on bank credit availability and cost. Later that day, the Federal Reserve released the 2024 stress test results, revealing higher capital depletions under stress scenarios, which will lead to increased capital requirements for many banks effective Oct. 1, 2024. This article provides an overview of the Fed’s 2024 Stress Test Results and demonstrates how these results exemplify the key points presented in BPI’s expert testimony pertaining to volatility of the stress test results and lack of transparency around the stress tests.  During the hearing, BPI presented several examples highlighting the volatility inherent in the supervisory stress-testing models and the need for increased transparency around the models. BPI emphasized that one of the main drivers of excess volatility in capital requirements is the lack of granularity in the Fed’s revenue projection models, and heavy reliance of these models on bank performance in the preceding year. This “momentum” effect was particularly evident in the Fed’s projections of noninterest income and noninterest expense in the 2024 stress test results. Another noteworthy result from this year’s stress tests is the substantial increase in capital requirements for several intermediate holding companies (IHCs) of foreign banks over the past few years. In 2023, the Fed implemented significant changes to the PPNR modeling for these firms. However, the lack of transparency in the Fed’s methodology makes it impossible to understand the factors driving the changes in these banks’ capital requirements.

BPI’s Stress Testing Testimony and 2024 Stress Test Results – Bank Policy Institute

The CFPB’s Misrepresentation about the Evidentiary Support for its Policies: Latest Examples

This post documents a pattern of misrepresenting, exaggerating or skewing the empirical support for favored policy positions on the part of the Consumer Financial Protection Bureau, which is undermines the agency’s credibility. One of the more prominent examples of this behavior has been the agency’s cherry-picked and error-ridden analysis in support of regulatory actions against credit card late fees, which it argues are excessive and necessitate a reduction. Past BPI blog posts have demonstrated that the CFPB’s analyses on late fees are flawed, ignoring or downplaying relevant evidence inconsistent with the agency’s narrative, and failing to recognize legitimate purposes of late fees. Another, more recent example is the CFPB’s claim of non-competitive interest rates by large credit card issuers, superficial and misleading for a number of reasons, including improper comparisons of large bank card program to credit union and specialty card programs. Additionally, the CFPB has exaggerated findings on the effects of price complexity on competition in consumer credit markets and has mischaracterized the prevalence and impact of consumer complaints about credit card rewards programs..

The CFPB’s Misrepresentation about the Evidentiary Support for its Policies: Latest Examples – Bank Policy Institute (

Potential Adverse Effects of the Basel III Capital Proposal on Consumer Credit Card Lines

In July 2023, U.S. banking agencies proposed a new 10-percent Credit Conversion Factor (CCF) on unused consumer credit card lines as part of the Basel III Endgame package. This charge is seen as overstating risk and could lead banks to reduce credit limits or close accounts. Consumers who rely on these unused lines for emergency liquidity, especially many low- and moderate-income households, might be particularly harmed. This analysis shows that Transactor accounts, which pay balances in full monthly and tend to have lower risk of default, have significantly lower utilization rates compared to Revolvers. The analysis also finds that these accounts are distributed broadly across all income groups, including among low- and moderate-income individuals. Thus, the benefit of reconsidering the proposed charge by, at a minimum, reducing the CCF for Transactor accounts, is clear. Reducing the CCF for these accounts seems justified by risk considerations and would prevent adverse effects on consumer access to credit. Further research is needed to fully assess the impact of the proposed CCF on different consumer groups.

Potential Adverse Effects of the Basel III Capital Proposal on Consumer Credit Card Lines – Bank Policy Institute (

Conferences & Symposiums

9th Annual Cambridge Conference on Alternative Finance
University of Cambridge, England
More Information

7/18/2024 – 7/19/2024
Exploring Conventional Bank Funding Regimes in an Unconventional World
Federal Reserve Bank of Dallas
More Information

Yale Program on Financial Stability Annual Research Conference: Announcement and Call for Papers
Yale University
More Information

9/19/2024 – 9/20/2024
23d Annual Bank Research Conference: Announcement and Call for Papers
FDIC Center for Financial Research
More Information

10/1/2024 – 10/2/2024
Conference on Technology-Enabled Disruption 
Federal Reserve Bank of Atlanta
More Information

10/11/2024 – 10/12/2024
Federal Reserve Stress Testing Research Conference: Announcement and Call for Papers
Federal Reserve Bank of Boston
More Information

10/18/2024 – 10/19/2024
The 10th Anniversary Wharton Conference on Liquidity and Financial Fragility: Announcement and Call for Papers
More Information

10/24/2024 – 10/25/2024
Workshop on Changing Demographics and Housing Demand
Federal Reserve Bank of Philadelphia
More Information

10/24/2024 – 10/25/2024
Inflation: Drivers and Dynamics
Federal Reserve Bank of Cleveland
More Information

11/21/2024 – 11/22/2024
Conference on Financial Stability: Announcement and Call for Papers
Federal Reserve Bank of Cleveland and Office of Financial Research
More Information

Pacific Basin Research Conference
Federal Reserve Bank of San Francisco 
More Information

11/25/2024 – 11/26/2024
III CEMLA-Dallas Fed-IBEFA Workshop on Financial Stability: Announcement and Call for Papers
Mexico City
More Information

10th Annual U.S, Treasury Market Conference
Federal Reserve Bank of New York
More Information

12/5/2024 – 12/6/2024
The Evolving Structure of the Financial Services Industry: Announcement and Call for Papers
Federal Reserve Bank of Atlanta and Georgia State University
More Information

Next Post: Research Exchange: May 2024 View Next Post


The views expressed do not necessarily reflect those of the Bank Policy Institute’s member banks, and are not intended to be, and should not be construed as, legal advice of any kind.