Research Exchange: February 2024

Selected Outside Research

CECL Implementation and Model Risk in Uncertain Times: An Application to Consumer Finance

The CECL loss reserving framework requires projection of lifetime loan losses based in part on economic forecasts, increasing the sensitivity of the allowance to economic forecasting and model error relative to the incurred loss framework it has replaced. This paper investigates the increased sensitivity to model and forecasting error of the CECL framework. In addition, the paper develops a modeling framework that relies on machine learning strategies and statistical principles that that can be applied to mitigate the effects of model uncertainty on CECL projections. The approach allows for quick and efficient deployment of an array of models and emphasizes “resiliency and adaptability of models and model infrastructure to novel shocks and uncertain economic conditions.” The approach is illustrated through application to a portfolio of auto loans using 20 years of portfolio performance data. CECL Implementation and Model Risk in Uncertain Times: An Application to Consumer Finance (

Cities Disrupted: The Diverse Impact of the PPP and Other Pandemic Support Programs

This paper analyzes the role of three key public support programs—the Paycheck Protection Program (PPP), expanded unemployment benefits and tax rebates—on the economic recovery from the COVID-19 downturn. In particular, the analysis assesses the effects of a critical, 10-day interruption in PPP funding. The analysis finds that the delay had “pronounced and enduring adverse effects concentrated in a few most populous urban areas, likely due to their structural vulnerability to pandemic-induced disruptions.” The analysis finds that the PPP program successfully supported business recovery despite this delay and without relation to the timing of PPP fund disbursement across localities. In addition, the study finds that compared to the PPP, the other two support programs “were quantitatively more important for the overall path of the recovery” compared to the PPP.

Size-Based Regulation and the Reallocation of Bank Fragility: Evidence from the Wells Fargo Asset Cap

Since 2018, U.S. regulators have imposed a punitive, $1.95 trillion asset size cap on Wells Fargo and Company, which has constrained its deposit growth and caused it to contract geographically. This study demonstrates that the cap also led to a reallocation of uninsured deposits to smaller banks and “thus contributed to the rise in banking fragility exposed by the 2023 regional bank crisis”. In particular, the analysis indicates that banks geographically closer to Wells Fargo experienced a stronger influx of flighty uninsured deposits after imposition of the cap, and particularly during the COVID-19 period. Moreover, their deposit growth was highest at branches located in closest proximity to Wells Fargo branches within the same county. The analysis also finds that once the Fed began tightening monetary policy in 2022, these same banks experienced larger deposit outflows. Following the failure of Silicon Valley Bank during the first quarter of 2023, these deposit outflows accelerated, and the banks also experienced lower equity returns.

Private Credit: Characteristics and Risks

Private credit refers to debt-like, non-publicly traded instruments provided by nonbank entities, such as private credit funds or business development companies (BDCs), to fund private companies. Private credit typically has been a source of funds for middle-market firms with annual revenues between $10 million and $1 billion, but in recent years it has been increasingly used by larger companies that were traditionally funded by leveraged loans. This article provides an expansive overview of the private credit market. Topics discussed include market size and growth, key characteristics of loans extended as private credit, loan pricing and rollover risk, default and loss rates and financial stability implications, including interconnections with banks. One key takeaway is that “banks are progressively selling complex debt instruments to private fund managers in so-called ‘synthetic risk transfers’ in order to reduce regulatory capital charges on the loans they make. Such instruments have limited transparency and pose hidden risks to the financial system, especially as the industry has yet to endure a prolonged recession”. The Fed – Private Credit: Characteristics and Risks (

Short Selling and Bank Deposit Flows

Although some observers have argued that the short selling of bank stock contributes to bank runs and bank failures, there is a scarcity of empirical analysis addressing this issue. This paper uses confidential Federal Reserve data on deposit flows and publicly available data on short interest and short-sale volumes to investigate the empirical relationship between short selling and deposit flows. The analysis focuses on 2021Q1-2023Q3, a period associated with three major run-driven bank failures and heightened concern about bank solvency and liquidity. The analysis finds no evidence of a material increase in short selling preceding the three bank failures. Although the authors does not claim to have ruled out such a relationship, they argue that “the burden of proof that short selling contributes to bank distress on proponents of restrictions on short selling”. Short Selling and Bank Deposit Flows (

Bank Deposit Rates Haven’t Kept Pace with Yields on Other Investments, but Depositors Are Staying Anyway

This article reviews the recent dynamics in bank deposit flows and deposit interest rates and the outlook for the near term. The dynamics during the recent monetary tightening cycle are shown to have mirrored those of past cycles—whereas money market fund yields quickly accelerated, bank deposit rates have lagged, resulting in a current spread of more than 2 percentage points. Because of the lagging deposit rates and also because of an outflow of uninsured deposits following the high-profile bank failures in March 2023, money market fund assets have grown, while deposits have declined since the start of the current cycle. Deposit levels during the second half of 2023 stabilized with continuing increases in deposit rates. They are expected to remain stable since “deposits remaining at banks are more likely to be held by depositors who value the convenience and safety of bank deposits over higher yields offered by alternative investments”. Bank Deposit Rates Haven’t Kept Pace with Yields on Other Investments, but Depositors Are Staying Anyway – Federal Reserve Bank of Kansas City (

Federal Reserve Losses and Monetary Policy

As described in this paper, past monetary policy decisions have resulted in the Federal Reserve suffering more than $140 billion in accumulated cash losses, in addition to $1 trillion in unrealized losses on its securities portfolio, leaving the Fed system and the majority of Reserve Banks technically insolvent on a GAAP basis. The paper examines what this means for the Fed’s monetary policy and other functions. It challenges the view of Fed officials that Fed’s negative GAAP capital does not compromise its ability to conduct monetary policy because the Fed can create money to cover its losses. After calculating the GAAP capital of each Reserve Bank and the system and estimating depositors’ loss exposures under current law, the paper reviews the current legal framework in place for addressing insolvent Reserve Banks. The authors argue that “the Fed’s narrative leaves out important details including that the Fed’s ability to print paper currency is limited by law and deposits held at insolvent Reserve Banks are unsecured liabilities that are legally at risk because they lack a federal government guarantee.”
Federal Reserve Losses and Monetary Policy by Paul Kupiec, Alex Pollock :: SSRN

Primary and Secondary Markets for Stablecoins

Stablecoins are designed to perform a mechanically complex function – to remain pegged to the dollar, even during periods of market volatility – but during recent periods of intense stress, several stablecoins have de-pegged on secondary markets. This paper focuses on the events of March 2023 when Circle’s U.S. Dollar Coin de-pegged significantly off the U.S. dollar, and markets for other stablecoins fluctuated considerably, in response to news that Circle was unable to wire out a portion of USDC reserves held at the troubled Silicon Valley Bank. The paper analyzes factors that contributed to the instability of USDC, “with an emphasis on distinguishing between primary and secondary market dynamics during a stablecoin crisis”. In particular, the paper examines four stablecoins with different technical designs, providing a detailed account of their dynamics on primary points of issuance and secondary markets. This analysis yields “several general insights about the nature of stablecoin markets during periods of stress”.
The Fed – Primary and Secondary Markets for Stablecoins (

Precautionary Debt Capacity

This study investigates the propensity of small and medium enterprises with bank credit lines to retain unused line amounts. The analysis is based on a field experiment conducted in collaboration with a bank, in which credit limits were randomly expanded on SME credit lines. Treated SMEs on average drew on 35 percent of the expanded debt capacity in the short run and 55 percent in the long run, despite having balances well far below their limits prior to the expansion. In addition, the study applies a cross-sectional analysis based on risk of hitting a credit limit, which finds further support for the notion that SMEs seek to preserve financial flexibility. Precautionary Debt Capacity :: SSRN

Chart of the Month

Projections of Peak-to-Trough CET1 Ratios Over the 9 Quarters

Estimates of Maximum Declines in Capital Ratios in the 2024 Stress Tests

The chart illustrates the maximum declines in the common equity tier 1 capital ratios for banks across Categories I through IV in the 2024 stress tests, as projected by BPI. Anticipated to face larger capital depletions in this year’s stress tests are some of the largest banks, due to the greater severity of market shocks under the severely adverse scenario. Also, this year’s stress test includes four exploratory scenarios: two macroeconomic and two market shocks. Our projections indicate that the reductions in the common equity tier 1 capital ratio, especially for Category I banks under the second exploratory macroeconomic scenario, could match or surpass those anticipated under the severely adverse scenario.

Featured BPI Research

Deep Dive: DFAST 2024 Stress Test Scenarios

The Federal Reserve recently disclosed the severely adverse scenario and Global Market Shock (GMS) for the 2024 stress tests on covered banks. Four exploratory scenarios were introduced to assess bank resilience to funding stress, in addition to rising interest rates and severe global recession scenarios. Though not directly affecting stress capital charges, the exploratory scenarios could influence bank performance. The 2024 stress test appears more severe than the previous year, especially with regards to equity prices and corporate bond spreads. Challenges include higher noninterest expenses, potential offset to revenue gains and a unique FDIC special assessment. Projections indicate a slight increase in capital ratio declines compared to the prior year, potentially influenced by the FDIC charge and exploratory scenarios. However, despite heightened market shocks, increased loan loss projections and provisions, higher pre-provision net revenue might offset capital ratio declines. Exclusion of the FDIC charge from noninterest expenses is recommended to maintain accurate projections. In conclusion, this year’s stress test might elevate capital requirements for large U.S. banks, influenced by market shocks, provisions and revenue projections, but excluding the FDIC charge from noninterest expenses could mitigate this impact. Uncertainty surrounds the exploratory scenarios, particularly the second macroeconomic scenario and the default of five largest hedge fund exposures, suggesting potential challenges for banks in maintaining capital ratios. Deep Dive: DFAST 2024 Stress Test Scenarios – Bank Policy Institute (

10 Pitfalls to Avoid When Designing Any Additional Liquidity Requirements

This note discusses the significance of the discount window in managing liquidity risk while outlining 10 pitfalls for banking agencies to avoid when revising liquidity regulations. Key observations include the importance of clear objectives; the need to learn from historical mistakes; caution with using ratios; proper treatment of repos; reducing the stigma associated with using the discount window and avoiding the use of liquidity requirements to address unrelated issues. Other important recommendations are to consider institutional differences, make conforming changes to the rest of the liquidity assessment framework and eventually revisit international standards like the liquidity coverage ratio if banks are required to meet short-term cash outflows using reserve balances and discount window capacity. In summary, the author welcomes the renewed focus on discount window preparedness but calls for careful and thoughtful development of new liquidity requirements to enhance the resilience of the banking system. 10 Pitfalls to Avoid When Designing Any Additional Liquidity Requirements – Bank Policy Institute (

Overdependence on Short-term Wholesale Funding: A Historical Perspective  

In this blog post, the historical context of banking practices during the 1920s and the 2007–2008 Global Financial Crisis is discussed, with a view toward shedding light on recent concerns about overreliance on short-term wholesale funding. The net stable funding ratio (NSFR) was introduced after the 2008 crisis to encourage banks to diversify their funding sources and maintain a stable long-term funding structure. Drawing parallels, the blog post examines Virginia state banks in the 1920s, particularly those labeled as “habitual borrowers,” to assess the impact of overdependence on wholesale funding on the banking system’s financial stability. In 1922, Virginia state bank regulators implemented a “reluctance to borrow” approach, identifying certain banks as “habitual borrowers” and raising concerns about an overreliance on loans from other banks. The analysis compares habitual borrowers with non-habitual borrowers in terms of balance sheet ratios, profitability and deposit compositions. The findings suggest that habitual borrowers were less liquid, more reliant on borrowed money and less profitable than non-habitual borrowers. Additionally, the study indicates that habitual borrowers offered higher rates on longer-term deposit accounts. The blog post underscores the historical importance of maintaining a stable funding structure in the banking sector and the challenges associated with overdependence on short-term wholesale funding.  Overdependence on Short-term Wholesale Funding: A Historical Perspective – Bank Policy Institute (

How the Federal Reserve Got So Huge, and Why and How It Can Shrink

This working paper examines the Federal Reserve’s change in monetary policy after the 2008 Global Financial Crisis, in particular criticizing the shift to an excessive-reserves approach. It argues that the decision in 2019 to adopt this framework was a mistake, as the approach failed to simplify policy conduct and led to the evaporation of the federal funds market. The text advocates for a return to a necessary-reserves system with a corridor framework, emphasizing that the prior approach allowed for better control of overnight interest rates with a smaller balance sheet. In addition, concerns are raised about the growing size of the Federal Reserve’s balance sheet, with a ratchet effect making it challenging to normalize. The note proposes managing the balance sheet to reduce reserves, implementing a voluntary-reserve targeting system and addressing the drawbacks of the excessive-reserves system, including its impact on interbank markets. How the Federal Reserve Got So Huge, and Why and How It Can Shrink – Bank Policy Institute (

Conferences & Symposiums

Consumer Research Symposium: Announcement and Call for Papers 
Federal Deposit Insurance Corporation, Arlington, VA
More Information

Financial Stability Implications of Digital Assets Products and Activities: Stablecoins and Tokenization Federal Reserve Banks of Boston and New York (virtual)
More Information

4/18/2024 – 4/19/2024
Inaugural Fintech and Financial Institutions Research Conference
University of Delaware and Federal Reserve Bank of Philadelphia: Announcement and Call for Papers
More Information

Rethinking Optimal Deposit Insurance: Announcement and Call for Papers
Yale University
More Information

5/2/2024 – 5/3/2024
7th Annual CFPB Research Conference: Announcement and Call for Papers
Consumer Financial Protection Bureau, Washington, D.C.
More Information

5/9/2024 – 5/10/2024
Conference on Fixed Income Markets and Inflation: Announcement and Call for Papers
Federal Reserve Bank of San Francisco
More Information

5/12/2024 – 5/14/2024
Boulder Summer Conference on Consumer Financial
Decisionmaking: Announcement and Call for Papers
Leeds School of Business, University of Colorado
More Information

5/15/2024 – 5/16/2024
Mortgage Market Research Conference: Announcement and Call for Papers
Federal Reserve Bank of Philadelphia
More Information

The 17th New York Fed/NYU Stern Conference on Financial Intermediation: Announcement and Call for Papers
New York City
More Information

6/5/2024 – 6/7/2024
OCC Bank Research Symposium on Depositor Behavior, Bank Liquidity, and Run Risk: Announcement and Call for Papers
Office of the Comptroller of the Currency, Washington, D.C.
More Information

6/19/2024 – 6/21/2024
Economics of Financial Technology Conference: Announcement and Call for Papers
University of Edinburgh Business School
More Information

Next Post: Research Exchange: March 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.