Session 10: Financial Regulation

We show that widely used macroprudential regulations that rely on historical cost accounting (HCA) — to insulate banks' balance sheets from financial market volatility — significantly affect the transmission of monetary policy. Using detailed supervisory data from Italian banks, we find that HCA mutes the transmission of quantitative easing on bank lending supply, weakening the effectiveness of monetary policy in reducing firm credit constraints. We also show that a drop in the market price of HCA-valued securities is equivalent to a reduction in capital requirements, which is large and nearly identical in Italy and the US.

We develop a two-layer asset demand framework to analyze fragility in the corporate bond market. Households allocate wealth to institutions, and institutions then allocate funds to specific assets. The framework generates tractable joint dynamics of flows and asset values, featuring amplification and contagion. The framework can be estimated using micro-data on bond prices, investor holdings, and fund flows, allowing for rich parameter heterogeneity across assets and institutions. We match the model to the March 2020 turmoil and quantify the equilibrium effects of unconventional monetary and liquidity policies on asset prices and institutions.

Over the past two decades, the mortgage industry has been transformed from the traditional bank-centered deposit taking, lending, and servicing model to a fragmented market with high non-bank participation. We document a novel mechanism for this unbundling – mortgage servicing transfers – and study the role of bank regulation in transforming servicing. Using a near universe of consumer credit records, we show that banks increase transfers of mortgage servicing rights (MSRs) to non-banks following the announcement of Basel III’s higher regulatory costs of holding MSR assets for banks. Based on predictions of a simple model of servicing transfers, we demonstrate which types of banks and loans experience the highest transfer rates. We find that banks selectively transferred below-median income, subprime, and 60+ day delinquent MSRs to non-banks. Loans subject to transfer due to regulatory pressure experienced more foreclosures and personal bankruptcies. Our results suggest that growth in the unbundling of mortgage servicing increased existing disparities in financial risks across households.

Bank branch density, defined as the number of bank branches to total deposits, has significantly declined over the past decade, fueled by a confluence of branch closings and the almost doubling of deposits between 2016 and 2022. During this period, banks with low branch density benefited from large deposits inflows, leading to even lower density. But the virtuous cycle of deposit growth in these banks stopped spinning when investors became wary about their financial health. Stock prices of banks with low branch density plummeted during the 2023 Banking Crisis as these banks experienced larger outflows of uninsured deposits. Our results suggest that digital banking enabled banks to grow faster and attract uninsured deposits, but those large deposit inflows took the form of “hot money” that changed its course when economic conditions worsened.

When race is not directly observed, regulators and analysts commonly predict it using algorithms based on last name and address. In small business lending — where regulators assess compliance with fair lending laws using the Bayesian Improved Surname Geocoding (BISG) algorithm — we document large prediction errors among Black Americans. The errors bias measured racial disparities in loan approval rates downward by 40%, with greater bias for traditional vs. fintech lenders. Since errors correlate with socioeconomic characteristics, basing regulation on self-identified race would increase lending to Black borrowers, but also shift lending toward affluent areas. Our results highlight systematic problems with policies based on race proxies.

Over 30 million US adults do not use formal consumer credit. How many of these are inefficiently excluded because they lack a credit history or have a poor credit score? We develop a framework to characterize the efficiency-maximizing system of credit histories and credit scoring, subject to the constraints imposed by the severity of adverse selection and by the ability of credit histories to predict future risk. We find US consumer credit features a moderate amount of adverse selection and persistent consumer types. This adverse selection generates substantial welfare loss: a majority of today's non-borrowers would be first-best efficient to lend to. Credit reporting helps alleviate the costs of adverse selection, with the current US system recovering roughly two-thirds of the welfare that would be lost in a no-credit-reporting counterfactual, relative to a full-information first-best. We find that requiring histories to be shorter – or to forget past defaults sooner – would induce some market unraveling but also would help non-borrowing consumers escape the "no history trap."

The welfare impact of increasing retirement contributions depends on how individuals adjust their spending, borrowing, and non-retirement savings in response. Using newly merged deposit-, credit-, and pension-account data from a large UK financial institution, we estimate the relevant elasticities by leveraging a policy that incrementally increased minimum retirement contributions in the U.K. from 2% to 8% of salary between March 2018 and April 2019. For every £1 reduction in monthly take-home pay due to higher employee pension contributions, consumers cut their spending by £0.34 (especially in the restaurant and leisure categories) and finance the remainder with lower deposit account balances and higher credit card debt levels. Those with lower initial deposit balances cut their spending the most, while those with significant liquid savings draw down their deposits. Interpreted via a theoretical model, these results suggest that a social planner concerned about undersaving should target retirement saving interventions at low-liquidity individuals whose spending is more elastic to increased retirement saving. In contrast, interventions that increase the retirement contributions of high-liquidity individuals are both less efficient (due to large crowd-out) and often regressive.

After the introduction of a nationwide Unified Payment Interface (UPI) in 2016, India has become one of the world’s leading economies for cashless transactions. We exploit the heterogeneity in the intensity of the adoption of digital payments across districts to show that economic outcomes, as measured by household income and small business activities, increased significantly in districts with a higher intensity of cashless transactions. These effects are stronger in financially less developed regions of the country. We achieve identification using two complementary empirical strategies. We first exploit the differences in the timing of participation on the UPI platform by different banks to obtain quasi-random variation in the level of digital payments across districts. Second, we exploit the within-district-year-quarter variation in the effect of cashless payments on economic outcomes across households who are differentially impacted by the adoption of digital payment. Specifically, we show that the impact of digital payments is stronger for self-employed households, such as hawkers and traders, compared to others. Relaxation of borrowing constraints and reduction in the transaction cost of payments are two principal mechanisms behind our findings.

We study the impact of rising mortgage rates on mobility and labor reallocation. Using individual-level credit record data and variation in the timing of mortgage origination, we show that a 1 p.p. decline in mortgage rate deltas (∆r), measured as the difference between the mortgage rate locked in at purchase and the current market rate, reduces moving rates by 0.68 p.p, or 9%. We find that this relationship is non-linear: once ∆r is high enough, households’ alternative of refinancing without moving becomes attractive enough that moving probabilities no longer depend on ∆r. Lastly, we find that mortgage lock-in attenuates household responsiveness to shocks to nearby employment opportunities that require moving, measured as wage growth in counties within a 50 to 150-mile ring and instrumented with a shift-share instrument. The responsiveness of moving rates to wage growth is nearly three times as large for households who are less locked in (above-median ∆r) than for those who are more locked in. We provide causal estimates of mortgage lock-in effects, highlighting unintended consequences of monetary tightening with long-term fixed-rate mortgages on mobility and labor markets.

Climate models predict that many natural hazards will become increasingly damaging and costly to insure as the effects of climate change manifest. We study how the cost of hedging disaster risk changes home prices by using a 2012 law that mandated flood insurance premium increases for properties discontinuously around flood zone boundaries and based on the timing of construction. With a triple-difference design, we find that homes that experience the largest increase in premiums experience the largest decline in home values. The effect is five times larger for homes that are exposed to sea level rise than those not exposed, suggesting that insurance pricing can accelerate the incorporation of climate risk in asset markets.

We evaluate the FDIC’s costs for resolving at-risk banks, as defined in Jiang et al., 2023 using the model developed in Allen et al., 2023 to determine bidder valuations of at-risk banks and simulate bidding. We show that resolution would cost the FDIC over $200 billion, if it maintained its bidder size and health restrictions, well in excess of the amount in its fund. Costs could be lowered if the FDIC relaxed these criteria and/or peeled off parts of the troubled banks. Our results illustrate that, during times of crisis, resolution costs can spiral because the set of unconstrained healthy bidders dries up.

Credit card lenders individualize contracts primarily through risk-based credit limits rather than interest rates. To understand lenders’ credit limit choices, I use novel statement-level data on the near-universe of UK credit cards active between 2010–2015 to estimate a structural model of the credit card market. The model explains differences in lenders’ credit limit distributions through a screening technology that provides lenders with a noisy signal of customers’ risk. I identify model parameters using a novel cost shock that results from the April 2011 case in the England and Wales High Court concerning the mis-selling of payment protection insurance. I use the estimated model to evaluate a counterfactual scenario in which lenders can freely individualize interest rates and credit limits, which the existing environment precludes. As a result, interest rates and credit limits are individualized, and profits increase. Risk-based interest rate discrimination emerges, resulting in large reductions in consumer surplus for the riskiest individuals. I conclude with potential explanations for the puzzling absence of risk-based pricing in the UK credit card market.