Research Exchange: February 2026

Selected Outside Research

Strong Wills, Strong Outcomes? Bank Resolvability and Credit Supply

Regulatory "living will" requirements mandate that large, systemically important banks prepare detailed emergency resolution plans. This study uses an unexpected policy change in May 2018, when banks below a $250 billion asset threshold were suddenly exempted from living will requirements, as the basis for an analysis of how living will requirements affect the behavior of U.S. banks. More specifically, the study conducts a difference-in-differences type regression analysis comparing newly exempt banks to those that remained subject to the rules. It finds that the profitability and loan volume of these banks increased significantly, and they provided loans at more favorable terms. In addition, the exempt banks greatly increased their off-balance-sheet activities and expanded their supply of credit in the syndicated loan market, particularly by providing riskier and less resolvable loans. The authors conclude that living-will requirements suppress credit supply and bank profitability by disincentivizing more complex lending relationships.

Strong Wills, Strong Outcomes? Bank Resolvability and Credit Supply

Model Risk under CECL: A Consumer Finance Perspective

The Current Expected Credit Loss (CECL) framework for bank loan loss provisioning requires banks to estimate lifetime expected losses using forward-looking macroeconomic scenarios. Large U.S. banks adopted it starting in 2020, replacing the older, "incurred loss" framework. The paper describes how CECL's heavy reliance on macroeconomic forecasts makes it highly vulnerable to two sources of model error during periods of crises: (1) forecast errors from genuinely uncertain economic conditions, and (2) model misspecification caused by government interventions such as stimulus payments. Using auto loan and credit card data spanning more than 20 years, the paper demonstrates that models tend to underperform in novel economic environments, underscoring the need for adaptability. Notably, models without macroeconomic drivers sometimes outperformed more complex ones during COVID-19, precisely because government intervention distorted those macro signals. The paper also develops an illustrative CECL modeling framework using simple machine learning techniques (decision tree classifiers), designed to be re-estimated quickly as new data emerges.

Model Risk Under CECL: A Consumer Finance Perspective

Who's the Boss? The Role of Dividend Clienteles in Banks' Lending Decisions

This paper examines how a regulatory ban on bank dividend payments during COVID-19 affected the funding costs and lending behavior of banks in the UK. The analysis finds that restricted banks with significant exposure to income-oriented mutual funds - a proxy for "dividend clienteles" - saw their share valuations fall by 6-10 percentage points, and their cost of equity rise by up to 1.7 percentage points. The increased cost-of-equity was offset by a decline in debtholder required returns due to restricted banks becoming better capitalized, but the offset was partial and overall funding costs rose for the most exposed banks. In addition, the authors find that less exposed banks increased SME lending by 40 percent relative to unaffected peers, as the funding cost pressure fed into lending decisions. The analysis further suggests that these lending differences had real effects: firms borrowing from less exposed banks increased their capital expenditure more than firms borrowing from heavily exposed banks.

Who's the Boss? The Role of Dividend Clienteles in Banks' Lending Decisions

Beyond Hot Money: Brokered Deposits and Bank Funding Stability

The conventional view is that brokered deposits (BDs) are unstable and run-prone. This paper examines the use of brokered deposits by U.S. banks during 2007 to 2023, challenging this view. The analysis demonstrates that aggregate BD balances rose during both the Global Financial Crisis and the 2023 banking turmoil, moving in the opposite direction to uninsured deposits during these stress episodes. Moreover, access to BDs did not freeze up for vulnerable banks - in fact, banks losing uninsured deposits were able to turn to BDs as a substitute funding source. Interest rates on brokered CDs also showed no signs of stress-driven spikes for BD-reliant banks, again in contrast to uninsured deposits. Although high-BD banks did experience larger uninsured deposit outflows and worse stock returns during the 2023 turmoil, these negative outcomes were driven almost entirely by banks that had recently ramped up their BD usage, not those with long-term reliance on them.

Beyond Hot Money: Brokered Deposits and Bank Funding Stability

Bank Failures: The Roles of Solvency and Liquidity

This paper provides a comprehensive literature review, supplemented by presentation of historical data, addressing the question of whether banks fail primarily because of illiquidity (runs by depositors on otherwise solvent banks) or insolvency (fundamental weaknesses impeding profitability). The discussion is organized around a simple theory which predicts that runs are more likely in banks with weak fundamentals, and which suggests that insolvency is distinguished by low recovery rates on bank assets. The main finding is that weak fundamentals, not illiquidity, are almost always the root cause of bank failures. Historical recoveries averaged only 75 cents on the dollar, even for banks that experienced runs before failing, consistent with the view that most failed banks were already insolvent. Moreover, across the entire historical record, banks that failed consistently showed weak capitalization, poor profitability and reliance on expensive funding - regardless of whether a run occurred. Post-mortem assessments by bank examiners of the time further reinforce this view. Conversely, runs rarely brought about the demise of inherently healthy banks. The authors conclude by discussing the policy implications of these findings and offering directions for future research.

Bank Failures: The Roles of Solvency and Liquidity

Factors Affecting Regional Bank Health and Supervisory Rating: An Exploration

This paper examines determinants of the health of regional banks (those with $10-100 billion in assets) during 2008 through 2023 using a bank's supervisory composite CAMELS rating as the measure of bank health. It investigates which regional economic conditions predict overall bank health, as well as which components of the CAMELS rating matter most for the composite score. A key finding is that the health of local commercial real estate (CRE) markets - particularly the industrial property sector - has a strong positive impact on regional bank health. Among the individual CAMELS components, the management rating stands out as particularly important for the overall composite score.

Factors Affecting Regional Bank Health and Supervisory Rating: An Exploration

Short-Circuiting Short-Term Funding

Stability of the U.S. tri-party repo market, which facilitates over $10 trillion in daily transactions, depends on time-sensitive daily settlements between a relatively small and concentrated set of counterparties. This paper examines the vulnerability of the market to cyber-induced operational disruptions that take a systemically important institution offline. The analysis combines actual counterparty-level transaction data with firm-level cybersecurity ratings to identify which institutions are most vulnerable to cyberattacks. It then simulates the market-wide impact of various outage scenarios. One key takeaway is that, whereas cybersecurity ratings flag bank-dealers and bank borrowers as having the highest inherent cyber risk, asset managers - primarily money market funds - exhibit the greatest systemic importance in this context. The simulations also show that some outages can affect more than twenty borrowers and over $100 billion in funding, and that a large lender going offline can push aggregate repo rates up by more than 50 basis points.

Short-Circuiting Short-Term Funding

Bank Runs and Interest Rates: A Revolving Lines Perspective

Corporate revolving credit lines represent about 20 percent of bank liabilities and, like uninsured deposits, are demandable claims that can be drawn down suddenly - as happened during both the 2008 financial crisis and the COVID pandemic in early 2020, creating significant liquidity stress for banks. This paper examines how interest rates affect the likelihood of "runs" on revolving credit lines, using detailed Federal Reserve supervisory loan data (FR Y-14Q) from 2015 to 2023. By exploiting interest rate floor provisions that prevent credit line contract rates from falling below a set threshold, the analysis identifies the causal effect of interest rates on drawdown behavior. The key finding is that credit line drawdowns are highly responsive to changes in interest rates. Additionally, the analysis finds that have more rate-sensitive deposits tend to face less risk-sensitivity on their revolving lines. The latter finding suggests that for these banks, reduced drawdowns on credit lines provides some offset to the deposit outflows they tend to experience when market interest rates are rising.

Bank Runs and Interest Rates: A Revolving Lines Perspective

Featured BPI Research

Does Mandating Interest Payments on Mortgage Escrow Accounts Benefit Consumers?

This working paper examines whether laws mandating interest payments on mortgage escrow funds benefit consumers, focusing on the effects of Iowa's 2022 repeal of its interest-on-escrow requirement. Estimates obtained using 2018-2024 Home Mortgage Disclosure Act (HMDA) data in a difference-in-differences framework suggest that lenders largely offset the associated revenue loss by increasing up-front origination fees. Low-income borrowers are the most affected, indicating a potentially regressive cross-subsidy associated with the regulation. The rule is also found to lower the likelihood that mortgage applications are originated as loans, again with the greatest impact on low-income applicants. Overall, these effects point to potential reductions in net social benefit stemming from the policy.

Does Mandating Interest Payments on Mortgage Escrow Accounts Benefit Consumers?

The Upside-Down World of the Liquidity Coverage Ratio: Some Unpleasant Arithmetic

This note evaluates the liquidity coverage ratio (LCR) requirement through a simple two-period framework and a set of numerical examples, arguing that the rule's implementation makes the liquidity buffer effectively unusable. Because banks are expected to hold enough high-quality liquid assets (HQLA) to maintain an LCR above 100 percent even during stress, they must hold substantial excess HQLA in normal times - an implied LCR tax that can reduce long-run credit provision. The analysis highlights several related distortions: the implied tax is larger for banks funded with more stable liabilities, and meeting the requirement in stress can induce fire sales of illiquid assets, potentially increasing fragility and the likelihood of central bank intervention - the opposite of the rule's stated objective. The note suggests policy adjustments aimed at restoring buffer usability, including to allow the minimum LCR requirement to adjust downward in stress environments and ex-ante recognition in the LCR of proven discount window capacity.

The Upside-Down World of the Liquidity Coverage Ratio: Some Unpleasant Arithmetic

How 2025 Bank Regulatory Reform Has Enhanced Economic Growth

This blog post argues that the pickup in bank lending during 2025 coincided with a meaningful change in the regulatory and supervisory backdrop that banks had increasingly viewed as constraining credit supply in 2022-2024. Using data on loan growth from the Federal Reserve's H.8 report and information from the Senior Loan Officer Opinion Survey, the post documents that banks tightened standards significantly over the latter time frame even as the economy continued to expand. Moreover, this tightening was tied to the regulatory environment - which encompassed Basel III reforms that would increase required capital, along with heightened supervisory pressures - rather than purely macro fundamentals. In contrast, the agencies' 2025 actions that emphasized transparency and reduced regulatory uncertainty, alongside reported easing in lending standards beginning in early 2025, were accompanied by a notable improvement in loan growth. To translate these shifts into macroeconomic terms, the post updates the Bassett et al. (2012) VAR framework and constructs counterfactual GDP paths, estimating that the 2022-2024 tightening lowered the level of real GDP by about 1.8 percentage points by late 2024, while the easing in 2025 is projected to lift GDP by roughly 0.6 percentage points by the end of 2026.

How 2025 Bank Regulatory Reform Has Enhanced Economic Growth

Outside Blog Posts and Research Notes of Interest

Nonbank Mortgage Companies: Ginnie Mae and FHFA Could Enhance Financial Monitoring

Financial Stability Considerations on Bail-In

Stablecoins and the Exorbitant Privilege of Money Creation

Buy Now, Pay Later: Recent Developments and Implications

Assessing Bank Resilience to a Funding Shock

Conferences & Symposiums

3/25/2026 - 3/26/2026
2026 Federal Reserve National Community Investment Conference
Phoenix, AZ
More Information

3/26/2026 - 3/27/2026
2026 Georgia Tech-Atlanta Fed Household Finance Conference: Announcement and Call for Papers
Atlanta, GA
More Information

4/3/2026
Fintech and Financial Institutions Research Conference: Announcement and Call for Papers
Federal Reserve Bank of Philadelphia and University of Delaware
More Information

5/7/2026 - 5/8/2026
13th Annual Conference on Financial Market Regulation: Announcement and Call for Papers
The Securities and Exchange Division of Economic and Risk Analysis
Washington, D.C.
More Information

5/14/2026 - 5/15/2026
Mortgage Market Research Conference
Federal Reserve Bank of Philadelphia
More Information

5/18/2026
2026 Research Conference on Banking Regulation
European University Institute and Bank Policy Institute
Florence, Italy
More Information

5/27/2026 - 5/29/2026
Boulder Summer Conference on Consumer Financial Decision Making
University of Colorado, Boulder
More Information

6/22/2026 - 6/23/2026
Fifth Conference on the International Roles of the U.S. Dollar: Announcement and Call for Papers
Federal Reserve Board, Washington, D.C.
More Information

6/25/2026 - 6/26/2026
2026 New York Fed Innovation Conference
Federal Reserve Bank of New York
More Information

9/2/2026 - 9/3/2026
Edinburgh Financial Technology Conference: Announcement and Call for Papers
University of Edinburgh Business School, Edinburgh, Scotland
More Information