Selected Outside Research
Are Stablecoins Efficient for Remittances? Evidence from a Mystery Shopping Exercise
This paper examines whether stablecoins are more efficient, in terms of cost and speed, than traditional remittance channels. To do this, the authors conduct a mystery shopping exercise executing transfers of 200 USDC across ten corridors linking Italy with Argentina, Brazil, South Africa, the United Arab Emirates and Japan. Stablecoins are shown to provide no systematic cost advantage over traditional channels with total costs ranging from 0.3 percent to 9 percent of the transferred amount. Execution speed is similarly heterogeneous and largely determined by the quality of domestic payment infrastructures. On and off ramp foreign exchange frictions are found to be the main source of cost and transfer duration.
Are Stablecoins Efficient for Remittances? Evidence from a Mystery Shopping Exercise
Firm Dynamics and Private Credit
This paper links loan-level data from two major private credit databases to U.S. Census microdata, creating a longitudinal dataset of over 17,000 borrowing firms. It then applies a synthetic control approach to compare these firms to similar untreated firms drawn from the broader universe of U.S. companies. The analysis finds that private credit financing boosts short-term employment growth and increases innovation (measured by patent activity). Effects are concentrated among smaller, younger and collateral-poor firms and are notably stronger when borrowers are matched with lenders who specialize in their industry, suggesting private credit primarily reaches firms that traditional banks are unable or unwilling to finance. A companion theoretical model shows this lender segmentation—banks serving safer firms, private credit serving riskier ones—emerges naturally from differences in funding costs and capital regulation, rather than regulation alone.
Firm Dynamics and Private Credit
Buying from the Family: Private Equity-Owned Insurers and Their Affiliated Investments
It has been increasingly common for private equity firms to own insurance companies while also running private credit businesses. This research study builds a novel dataset linking PE ownership of U.S. insurers to detailed records of insurers’ investment holdings, transactions and performance, drawing on insurance-industry ownership filings, statutory transaction filings and regulator-run stress tests for structured credit securities. The study finds that PE-owned insurers invest heavily in structured securities and predominantly purchase these from affiliated credit businesses, paying noticeably higher prices for these affiliated purchases than independent insurers pay for the identical securities on the same day—amounting to an estimated $27 million per year in financing benefits for the affiliated businesses. These affiliated investments carry higher promised returns but greater downside risk, with modeling showing they could reduce insurers’ capital cushions by nearly nine percentage points in a severe economic downturn, roughly wiping out the average life insurer’s capital buffer. The authors conclude that PE-owned insurers effectively function as a captive funding source for their owners’ credit businesses, bearing outsized risk that isn’t fully captured by current capital requirements, with broader implications for financial stability as private equity’s role in credit markets continues to grow.
Buying from the Family: Private Equity-Owned Insurers and Their Affiliated Investments
Direct from Syndication: Syndication Costs, Liquidity, and Growth of Direct Lending
There has been some substitution in corporate credit markets from bank-arranged syndicated loans toward direct lending by private credit funds, a trend often attributed to bank capital requirements. This paper instead argues that a key driver is the risk and cost that banks bear when trying to place syndicated loans with investors, which have increased since bank liquidity regulations took effect. Using a dataset of over 9,000 syndicated loan deals and 8,000 direct lending deals from 2010 through 2025, the study tracks firms that switch from syndicated borrowing to direct lending and compares them with similar firms that continue to borrow in the syndicated market. It finds that borrowers whose prior syndicated loans required more last-minute adjustments or took longer to place with investors were significantly more likely to switch to direct lending later, and that this relationship—along with banks’ fees for bearing this risk—grew notably stronger after new liquidity rules took full effect in 2017. Banks more reliant on volatile, non-deposit funding (and thus more exposed to these liquidity rules) pulled back further from syndicated lending than deposit-funded banks, supporting the idea that liquidity regulation, not just capital rules, pushed lending activity outside the banking sector.
Direct from Syndication: Syndication Costs, Liquidity, and Growth of Direct Lending
Decentralized Exchanges for Stablecoins
This paper studies decentralized platforms that specialize in secondary market trading of stablecoins using automated pricing mechanisms. The paper builds a theoretical model of how liquidity and arbitrage traders allocate activity between centralized and decentralized platforms, and then tests the model’s predictions. The empirical analysis, using trade-level data from a leading decentralized platform (Curve) and a major centralized exchange (Binance) from 2020 through early 2023, confirms key implications of the model. In particular, average trade size is much larger on the decentralized exchange, while deviations of stablecoin values from their price peg are about the same on average across the two exchanges. When a stablecoin briefly lost its peg following the Silicon Valley Bank collapse in 2023, trading activity on the decentralized platform surged relative to the centralized exchange, consistent with it absorbing more of the price-correcting activity as implied by the model. The authors also find a direct trade-off in platform design: pricing mechanisms that keep trading costs low and support peg stability also expose liquidity providers to larger losses during a de-pegging event. The paper concludes that stablecoin platform design involves an inherent balancing act between price stability and liquidity provision.
Decentralized Exchanges for Stablecoins
Repo Rate Dynamics: The Role of Dealers, Hedge Funds, and Issuance
The U.S. repo market has undergone major structural shifts in recent years, including rapid growth in Treasury debt issuance and a rise in hedge funds using leveraged trading strategies that rely heavily on repo borrowing, straining dealer banks’ capacity to finance repo activity. This paper jointly analyzes reserves, dealer balance-sheet capacity, hedge fund activity and Treasury issuance together, using a statistical technique that assesses how these forces affect funding-market spreads differently depending on whether conditions are calm or stressed. The analysis finds that higher reserve levels consistently ease funding pressures and become especially important when conditions tighten, confirming reserves as the primary stabilizing force. It also finds that in calm periods, growing hedge fund borrowing tends to strain dealer balance sheets and push funding costs up, but as conditions tighten, hedge funds pull back and Treasury issuance becomes the more important source of strain, as dealers must absorb a growing supply of securities. The paper concludes that “repo market outcomes reflect the interaction between liquidity supply, leveraged demand, and intermediation capacity.”
Repo Rate Dynamics: The Role of Dealers, Hedge Funds, and Issuance
The Declining Role of Deposits in Credit Creation
This paper examines the feedback loop through which bank deposit growth affects credit creation, which in turn leads to additional deposits, and how that has evolved as financial activity has moved outside the traditional banking system. The analysis uses aggregated weekly U.S. banking data from 2005 to 2019, estimating a model of Treasury cash flows (movement of federal government funds in and out of private bank deposits) to construct a proxy for exogenous deposit shocks. This is combined with other banking time-series data to parameterize a structural economic model in which households and firms choose between banks and nonbank alternatives for saving and borrowing. The analysis finds that while a dollar of new deposits generates over two dollars in cumulative deposits through repeated rounds of lending, this amplification effect weakened substantially over the sample period, and the overall amount of bank credit generated by each deposit dollar fell by more than half. The decline reflects two forces: banks now lend out less of each new deposit dollar, likely due to higher costs of maintaining capital and liquidity buffers, and less of the money from new loans returns to banks as stable deposits, as savers increasingly park funds in alternatives like money market funds.
The Declining Role of Deposits in Credit Creation
Stablecoins under Stress in a National Economy: Transaction-Level Evidence from Austrian Crypto-Asset Service Providers
How demand for crypto assets responds to financial shocks has been hard to study directly, because blockchain transactions are anonymous. This paper exploits an Austrian regulation requiring licensed crypto companies to report all blockchain addresses they control, allowing the authors to directly reconstruct nearly 12 million transactions worth over $50 billion, involving all 12 licensed crypto firms. The analysis finds that this activity is overwhelmingly connected to the global crypto market rather than to other domestic firms; is dominated in dollar terms by a tiny number of large, frequent counterparties even though the vast majority of participants trade rarely and in small amounts; and that ordinary users and large institutional players reacted quite differently to crises. For instance, retail-like users tended to pull funds out of custodial accounts after a major exchange collapsed, while institutional flows looked more like routine repositioning. Notably, the study finds little support for the idea that stablecoins reliably serve as a “safe haven” during turmoil: during the Silicon Valley Bank crisis, for example, retail users added funds to one major stablecoin while institutions withdrew from it, a split tied to that stablecoin’s redemption rules favoring institutional accounts.
Bank Specialization and SMEs’ Employment Stability
This paper contributes to a growing body of research which suggests that banks concentrating their lending in a particular region or industry develop specialized knowledge that helps them better serve harder-to-assess borrowers. Specifically, the paper asks whether that specialized knowledge also helps small and medium-sized businesses protect jobs when they’re hit by a sudden drop in sales. Drawing on detailed, proprietary French banking and credit-registry data that tracks each bank’s lending by region and industry, the analysis finds that firms borrowing from more geographically or industry-specialized banks are better able to maintain their workforce when experiencing a sharp, firm-specific sales decline. These banks continue extending credit that helps firms cover short-term cash needs rather than cutting off support under such circumstances, but this protective effect fades when a sales decline turns out to be long-lasting. The analysis also shows that this employment-stabilizing effect is separate from the general benefits of having a long-standing relationship with a lender, and it holds up across firms of different sizes, ages and borrowing structures, though it is somewhat weaker for the smallest firms.
Bank Specialization and SMEs’ Employment Stability
Macroeconomic Parameter Instability in Auto Loan Loss Models
Using loan-level data from consumer credit reports spanning 2000 through 2025, this paper develops a statistical model that tracks the repayment performance of auto loans (transitions from current to delinquent, defaulted or paid off early), re-estimating the model repeatedly using rolling windows of data to assess model stability. The analysis finds that the relationship between unemployment and loan default shifted during both the Great Recession and the Covid pandemic, causing a prior modeled relationship to overstate realized defaults. These shifts persisted—they do not appear to be explained solely by expansion of loan forbearance during the downturns. The authors opine that these findings have implications for “stress testing models and loss forecasting practices that rely on stable unemployment-default relationships.”
Macroeconomic Parameter Instability in Auto Loan Loss Models
Featured BPI Research
A Post-Pandemic Consumer-Centered View of Frauds and Scams: Incidence, Exposure and Reporting
This blog post analyzes new survey data from the Consumer Financial Protection Bureau on the frequency and nature of fraud and scams targeting consumers in the years since the pandemic, including the kinds of fraud and scams consumers encounter, how much money is at stake and whether they report these incidents. The research found that roughly 78 million adults—nearly one in three—experienced some form of fraud or scam in 2024. Incidents involving online shopping, phishing, social media and impostor scams were common types. Total financial exposure reached about $67.1 billion, with a minority of incidents accounting for most of the financial losses. Fraud and scams involving bank transactions, such as wire transfers and cashier’s checks, tended to be more financially devastating than card-related fraud. Moreover, most victims report the incident to their bank or financial institution rather than to law enforcement or federal agencies. They are also more likely to feel supported by financial institutions. This pattern makes it harder for regulators to get a full, accurate picture of fraud trends and respond effectively on their own.
A Post-Pandemic Consumer-Centered View of Frauds and Scams
The 2026 Federal Reserve Stress Test Results: A Framework in Transition
The Federal Reserve’s 2026 supervisory stress test, which covered 32 large banks, indicated that all of them would remain well capitalized under a severely adverse macroeconomic scenario. Notably, this year’s stress tests will not affect banks’ capital requirements as the Federal Reserve finalizes revisions to its stress testing framework. Overall capital losses in the 2026 stress test were slightly smaller than projected in the 2025 test, with this improvement driven mainly by a rise in projected net interest income thanks to a steeper gap between short- and long-term rates, which outweighed somewhat higher projected loan losses and smaller valuation gains on securities holdings. Results varied across banks, with the very largest institutions seeing a small uptick in projected losses while others improved.
The 2026 Federal Reserve Stress Test Results: A Framework in Transition
Outside Blog Posts and Research Notes of Interest
Findings from the 2026 TIAA Institute-GFLEC Personal Finance Index
The Rise of Tokenization
Artificial Intelligence and Cybersecurity in the Financial Sector
Are U.S. Consumers Ready to Use Pay-by-Bank at the Point of Sale?
How Basel III Changes Where Capital Sits: Nonbank Subsidiaries as Equity Reservoirs
On the Comparison of Capital Requirements for Global Systemically Important Banks
Conferences & Symposiums
9/2/2026 – 9/3/2026
Edinburgh Financial Technology Conference: Announcement and Call for Papers
University of Edinburgh Business School, Edinburgh, Scotland
More Information
9/10/2026 – 9/11/2026
Higher Education Finance Research Conference: Announcement and Call for Papers
Federal Reserve Bank of Philadelphia
More Information
9/13/2026
Workshop on Innovations in Credit Scoring
Federal Reserve Bank of Philadelphia (Virtual)
More Information
9/18/2026 – 9/19/2026
2026 Wharton-Chicago-Harvard Insolvency and Restructuring Conference: Announcement and Call for Papers
University of Pennsylvania, Philadelphia, PA
More Information
9/24/2026 – 9/25/2026
Inflation Drivers and Dynamics Conference 2026: Announcement and Call for Papers
Federal Reserve Bank of Cleveland
More Information
9/24/2026 – 9/25/2026
25th Annual Bank Research Conference: Announcement and Call for Papers
FDIC Center for Financial Research, Arlington, VA
More Information
9/24/2026 – 9/25/2026
10th Annual Fintech Conference
Federal Reserve Bank of Philadelphia
More Information
10/6/2026 – 10/7/2026
2026 Annual Community Banking Research Conference
Federal Reserve Bank of St. Louis
More Information
10/9/2026 – 10/10/2026
2026 Wharton Conference on Liquidity and Financial Fragility
University of Pennsylvania, Philadelphia, PA
More Information
10/29/2026 – 10/30/2026
MIT GCFP 13th Annual Conference | “Financial Regulation in an Era of Innovation and Disruption”: Announcement and Call for Papers
Cambridge, MA
More Information
10/29/2026 – 10/30/2026
Sixth Biennial Conference on Auto Lending: Announcement and Call for Papers
Federal Reserve Bank of Philadelphia
More Information
11/5/2026 – 11/6/2026
2026 Federal Reserve Stress Testing Research Conference
Federal Reserve Bank of Boston
More Information
11/13/2026
NY Fed-ECB Conference on Nonbank Financial Institutions: Announcement and Call for Papers
Federal Reserve Bank of New York
More Information
11/18/2026 – 11/19/2026
15th Annual Research Workshop – Efficient and proportionate regulation for a competitive financial sector
European Banking Authority (Paris, France)
More Information
11/19/2026 – 11/20/2026
International Conference on Payments and Securities Settlement: Announcement and Call for Papers
Deutsche Bundesbank (Conference Center in Eltville, Germany)
More Information
11/19/2026 – 11/20/2026
2026 Financial Stability Conference
Federal Reserve Bank of Cleveland
More Information
11/20/2026
Pacific Basin Research Conference: Announcement and Call for Papers
Federal Reserve Bank of San Francisco
More Information
