Selected Outside Research
Current Expected Credit Losses and Consumer Loans
Using data from a large U.S. consumer credit bureau, this study explores whether the implementation of the Current Expected Credit Loss accounting standard for loss reserving has affected the pricing of auto and personal unsecured loans. The analysis first applies a difference-in-differences approach that compares the interest rates set by CECL-adopting and non-adopting banks before and after the policy change and finds no evidence that CECL affected loan pricing. The analysis then tests for differential price response for different loan maturities, exploiting the fact that CECL has a larger effect on loss reserves for longer-maturity loans, and again finds no statistically or economically significant effect. The study notes that “a possible explanation for the small, estimated impact of the adoption of CECL on consumer interest rates is that most banks were well-capitalized and therefore faced a low shadow cost of regulatory capital during the transition to the new standard.” Read More ⇨
Fintech and Gender Discrimination
This paper provides a case study of how lending decisions based on a machine learning system may result in gender discrimination. The peer-to-peer consumer lending platform that is the subject of this study had switched from a human-based to a machine learning-based system for assigning a credit rating and interest rate to each loan application. After the switch, the platform assigned higher interest rates and better credit ratings to female borrowers. This study attributes this result to “the platform’s attempt to maximize its revenue,” whereby the machine learning algorithm assessed not only consumers’ risk of default but also their response to offered loan terms led to price discrimination. Thus, higher interest rates and better credit ratings were assigned to less price-sensitive borrowers, who were disproportionately female. Read More ⇨
Out of Sight, Out of Mind? Banks’ Private Information, Distance and Relationship Length
Two empirical measures of banks’ private information about borrowers for C&I loans are developed in this paper using Federal Reserve supervisory, loan-level data, and assessed in relation to the business’s physical distance from the bank and the length of the borrowing relationship. One measure, labeled “incongruity,” is the variability of internal ratings relative to expected ratings based on public information. The second is “unfavorability,” the extent to which borrowers are assigned a lower internal rating relative to the expected ratings based on public information. The analysis indicates that incongruity increases in physical distance and length of the bank-firm relationship while unfavorability increases in distance and decreases in relationship length. In addition, the analysis demonstrates that incongruity and unfavorability also significantly affect maturity and loan amounts. Read More ⇨
Limitations of Implementing an Expected Credit Loss Model
This paper uses supervisory, loan-level data from Germany to investigate how adoption of the expected credit loss accounting standard for loss reserving (IFRS 9) affected banks’ internal ratings assignments for corporate exposures. The analysis exploits a cutoff for the level of provisions at the investment-grade threshold based on banks’ internal rating of a borrower, comparing the rating assignments of adopting versus non-adopting banks in a neighborhood of this threshold. Results indicate that the banks required to adopt the new rules assign better internal ratings to the same borrowers compared to non-adopters, consistent with a strategic use of ratings discretion. These banks also reduce their lending exposure to borrowers at the highest risk of being downgraded to below the cutoff, such that these loans would be associated with additional provisions in future periods. Read More ⇨
The Information Value of Past Losses in Operational Risk
The predictive relationship between past and future operational losses of U.S. banks is explored in this study. The analysis finds that “inclusion of past operational losses improves the performance of operational risk models, even when accounting for a wide range of quantifiable controls.” In addition, the analysis finds that past losses are predictive of future losses across a range of event types, and that they are informative up to three years prior. The study also shows that past losses suffered by peers are informative, and that historical loss frequency is generally more predictive of future exposure than historical loss severity, “likely because loss frequency is a more stable metric of exposure than loss severity.” Read More ⇨
The Ring-Fencing Bonus
In the United Kingdom, large banks are required to ring-fence their retail banking from their investments and international banking operations. This paper examines the effect of ring-fencing on bank risk as reflected in short-term repo rates, using confidential data on nearly all sterling-denominated repo transactions in the U.K. The analysis finds “compelling evidence that banking groups subject to ring-fencing are perceived to be safer; repo investors lend to ring-fenced groups at lower rates, controlling for bank characteristics and collateral risk.” In addition, the study finds that ring-fenced groups charge more to supply liquidity, and that the banking groups reduce their risk-taking after the imposition of the fence. Overall, the results suggest that ring-fencing reforms have had “a significant beneficial impact on risk in the banking system.” Read More ⇨
The Future of Money and Its Implications for Society, Central Banks, and the International Monetary System
This article discusses the broad implications for society, banking and central banking of recent innovations in payments technology. The article highlights how digital platforms can “ease entry for financial services providers, increase transactional efficiency, and widen access to and participation in the financial system.” The broad theme is that the payments sector is entering an era of increased competition in money, with both domestic and international dimensions. Domestically, “digital payment systems, perhaps stablecoins, perhaps even a CBDC” may emerge as important, alternative means of payment, but the fundamental store of value will remain the fiat currency. In the international monetary system, certain currencies, including emerging market currencies, will become more important in terms of international payments, but not as stores of value or reserve currencies. Read More ⇨
Useful, Usable and Used? Buffer Usability during the COVID-19 Crisis
The effectiveness of regulatory capital buffers in helping to support lending provision by U.K. banks through COVID‑19 is examined in this paper using data on U.K. mortgages. The analysis applies a difference‑in‑differences empirical strategy, comparing the effects across banks from the regulatory cut to the countercyclical capital buffer implemented during the pandemic. The results indicate that banks receiving greater capital relief from the cut to the countercyclical buffer, due to being less capital-constrained at the outset, maintained more stable capital ratios, loss provisions and risk‑taking capacity. In contrast, “more constrained banks defended their capital surpluses to a greater extent during the pandemic, and did so by maintaining higher loan rates, lower loan values and tighter terms on riskier lending.” The findings suggest that “regulatory buffers may be less usable than intended, but buffer releases can dampen these unintended consequences.” Read More ⇨
Chart of the Month
The common equity tier 1 capital ratio is the ratio of common stock (the most loss-absorbing type of capital) to risk-weighted assets. This chart shows the evolution of common equity tier 1 capital ratio over the last two decades. Since the GFC, the six largest banks in the U.S. have almost doubled their CET1 capital ratios. In addition, banks increased their capital ratios in 2022 in response to increases in capital requirements (higher stress capital buffers and capital surcharge for global systemically important banks).
Featured BPI Research
Basel Finalization: The History and Implications for Capital Regulation – Part I
In his inaugural speech as the new Federal Reserve Vice Chair for Supervision, Michael Barr pointed to releasing a proposal to implement the 2017 reforms to the Basel III capital framework as one of his key priorities. The latter set of reforms is commonly referred to as the “Basel Finalization” package. Barr also announced that the Federal Reserve would undertake a “holistic review” of the U.S. capital framework that presumably would inform the Basel Finalization package. Both this holistic review and implementation of the Basel Finalization package will likely breathe new life into long-debated questions about the appropriate scope, structure and content of the U.S. capital requirements. This is the first in a series of posts to help readers assess the forthcoming proposal implementing the Basel Finalization package, which is expected in the first or second quarter of 2023. This first post reviews the purpose of capital requirements and the principles underlying the Basel framework and its evolution over the last several decades, culminating in the 2017 Basel Finalization revisions issued by the Basel Committee on Banking Supervision. Read More ⇨
Basel Finalization: The History and Implications for Capital Regulation – Part II
Aside from the resulting level of overall capital requirements, the most fundamental questions regarding U.S. implementation of the Basel Finalization revisions are how they would affect the overall structure of the U.S. capital framework and which banks would be subject to the changes. As detailed in the Part I post, the largest banks must comply with two “stacks” of risk-based capital requirements. These combinations of capital minimum and buffer ratios relative to specific RWA calculations, taken together, make up the total ratio of capital to risk-weighted assets a bank must comply with to avoid negative consequences, such as limitations on capital distributions. Whether the U.S. regulators retain the dual RWA approach, and its resulting two-stack set of capital requirements—and, if so, which stack (or stacks) it will assess using the Basel Finalization revisions to calculate RWAs—will significantly influence firms’ capital requirements. Read More ⇨
Basel Finalization: The History and Implications for Capital Regulation – Part III
In this third installment of our series on the Basel Finalization agreement, we explore potential modifications to enhance the sensitivity to risk and reduce the procyclical impact of capital requirements, so as to prevent worsening of economic downturns. These modifications also aim to avoid raising borrowing costs for businesses and households, maintain market liquidity and promote economic growth. As discussed in the note, banking regulators have two broad options for mitigating capital increases while maintaining the risk sensitivity of the U.S. framework: (1) adjustments to risk-weighted assets and (2) adjustments to capital buffers. Read More ⇨
The Role of Credit Card Late Fees in Encouraging Timely Repayment Is Essential to Efficient Functioning of the Market
The current safe harbor policy governing credit card late payment fees was implemented by the Federal Reserve Board, pursuant to the requirements of the 2009 Card Act, based on thoughtful considerations of cost versus benefit. This research note explains how late payment fees within the current safe harbor limits satisfy CARD Act requirements that they be reasonable and proportional. In addition to covering costs associated with accounts that are not paid on time, they have important incentive effects that allow the market to function more safely: (1) Late fees incentivize consumers to overcome or moderate behavioral biases that are potentially harmful to their financial health. (2) Late fees also may incentivize consumers to limit the number of open and active credit card accounts they hold, and a credit card market populated by consumers who are thus more disciplined will allow issuers to compete more vigorously. (3) Late fees encourage customers to only miss unavoidable payments and to work with their bank to identify solutions such as renegotiation opportunities. Compelling banks to reduce their late fees could weaken these incentive effects. Read More ⇨
