Uncertainty-Induced Reallocations, Innovation, and Growth (with Ravi Bansal, Mariano Massimiliano Croce, and Wenxi Liao)
R&R, Review of Economic Studies
This paper highlights the effect of volatility on sectoral reallocation and its endogenous impact on total factor productivity (TFP). We provide empirical evidence that volatility shocks reduce TFP, private R&D investments, and innovation (patents) in the economy. Using micro data, we document that increased volatility leads to capital reallocation away from riskier high-R&D-intensive firms and toward safer, low-R&D-intensive firms. In our multisector DSGE model with recursive preferences and endogenous growth, when volatility increases, a volatility-averse investor demands higher risk premia and reallocates resources toward safer but less innovation-intensive sectors. This reallocation endogenously leads to lower TFP and growth.
SONOMA: a Small Open ecoNOmy for MAcrofinance (with Mariano Massimiliano Croce and Mohammad Jahan-Parvar)
R&R, Journal of Finance
Winner: China Finance Review International Research Excellence Award, 2025 MRS International Risk Conference
Data Visualization Website
We develop a new small open economy model (SONOMA) in which domestic corporate debt and equities coexist (Jermann and Quadrini, 2012) and are affected by shocks to both external credit and equity markets. In a novel empirical analysis of several small-but-developed economies, we show that both external debt and equity shocks are important determinants of domestic economic fluctuations, corporate leverage, and net foreign asset positions. SONOMA replicates our empirical facts about asset prices, financial flows, and economic activity.
Preference-Driven Herding in Primary Markets: Evidence from Non-Fungible Tokens (with Sebeom Oh and Anthony Lee Zhang)
Summarized in the Chicago Booth Review Here and Here
Previously circulated under the titles "Investor Experience Matters: Evidence from Generative Art Collections on the Blockchain" and "Digital Veblen Goods"
Social goods are difficult to study because of selection bias, as available data typically focus on successful products. The non-fungible token (NFT) market offers a rare exception, as the blockchain records every launch, success, and failure. Using comprehensive NFT data from 2021–2024, we study preference-driven herding in primary markets. Launch outcomes are sharply bimodal, demand accelerates as sellouts approach, and early participation shocks persist for months. These patterns are consistent with preference-driven coordination rather than purely informational herding. We provide further evidence consistent with social-goods models, including substantial primary-market underpricing that is exploited by scalpers.
Environmental Deregulation, Uncertainty, and Financial Statement Verification in Bank Lending (with Sudipta Basu, Yujie Ma, and Wei Wang)
We examine how U.S. banks adjusted their information acquisition practices after a deregulatory shift in federal environmental and energy policy beginning in 2017. Although these policy changes reduced compliance costs, they also increased uncertainty about environmental liabilities for private oil and gas firms. We test whether banks responded by tightening documentation requirements, specifically by increasing their demand for highly assured financial statements used in credit evaluation. Using proprietary data, we compare assurance levels for oil and gas borrowers with those in unaffected industries in a difference-in-differences design. We find that banks increasingly required unqualified audits after 2017, with the strongest effects among large firms and in regions where Environmental Protection Agency enforcement had been historically strict or declined most after deregulation. Taken together, these patterns suggest that when deregulation increases uncertainty about borrower outcomes, lenders respond by increasing their demand for highly assured financial reporting.
Valuation Precision and Economic Outcomes (with Christophe Spaenjers)
Markets aggregate agents’ valuations through prices and trades, yet agents may differ in how precisely they form and express those valuations. We study heterogeneity in valuation precision and its implications for bargaining behavior and economic outcomes. Using transaction-level data from decentralized non-fungible token markets without mechanical pricing constraints, we show that round pricing provides a stable agent-level proxy for differences in valuation precision and pricing granularity. Rounding behavior predicts systematic variation in listing choices, counteroffers, price adjustments, and realized returns. Decomposition analyses show that investor heterogeneity explains a substantial share of rounding behavior beyond price-level focal points and asset-specific conditions.
Direct Lenders and Financial Stability (with Young Soo Jang)
Direct lenders have gained substantial market share in lending to middle-market firms. We assess their implications for financial stability across three channels: (1) contagion via funding chains and fire sales, (2) weaker monitoring due to close ties with private equity sponsors, and (3) reduced lending in distress. The first two risks are largely mitigated by institutional factors. The third, lending pullbacks during downturns, remains uncertain due to limited empirical evidence. Overall, our analysis suggests that direct lenders do not currently pose greater systemic risk than banks, though this assessment may evolve as the sector grows and new vulnerabilities emerge.
Competition and Relationship Lending in Small Business Credit (with Yaming Gong and Simin Xie)
Previously circulated under the title "The Impact of Lender Competition on Small Business Loan Pricing: Evidence from the SBA 7(a) Program"
Classic theories predict that informational frictions in small business credit markets can generate nonstandard competitive dynamics, weakening or reversing the link between competition and prices. Using the universe of SBA 7(a) loans from 2009 to 2022 as a laboratory, we document that relationship-intensive segments exhibit a flatter, and sometimes negative, relationship between concentration and interest rates, while more transactional segments display standard market-power effects. We then provide causal evidence using a merger-based design that isolates changes in competition within the SBA market. The results confirm that relationship-lending incentives govern how market structure translates into small business loan pricing.
Mission-driven Lenders (with Tlian Tang)
The U.S. government created the Community Development Financial Institution (CDFI) certification to promote greater credit access in distressed communities. In this paper, we provide a systematic analysis of CDFIs and provide insights into why CDFIs are growing and how they are different from other lenders. Consistent with their mission-driven requirement, we document that CDFIs have expanded in counties with a greater reliance on government-subsidized business lending, higher unemployment rates, and a larger minority population. Within the universe of depository institutions, credit unions and minority depository institutions (MDIs) are more likely to become certified CDFIs as well as institutions with relatively low levels of cash and high leverage. After becoming certified, CDFIs tend to grow faster and lend more, which suggests that the resources available to CDFIs alleviate institution-level financial constraints. In our final analysis, we analyze the cost of CDFI lending using a novel loan-level dataset.
The Transmission of Federal Credit Through Financial Intermediaries: Evidence from CDFIs (with Tilan Tang)
To understand how federal credit transmits through intermediaries, we study the Community Development Financial Institution (CDFI) program, which channels support through certification and competitive capital awards. We decompose participation into certification (eligibility), application (demand), and award receipt (allocation). We show that these margins reflect distinct forms of selection. Certification is driven by persistent institutional characteristics, while application behavior among credit unions is associated with financial conditions. Conditional on application, award allocation is primarily related to profitability, indicating that while more constrained institutions seek funding, awards are directed toward those best positioned to deploy capital. Following award receipt, credit unions expand assets, lending, and non-deposit funding, while banks exhibit more limited responses. Certification also responds to anticipated local investment demand during the 2017 Opportunity Zone nomination phase. Our findings show that intermediary-targeted federal credit operates through distinct selection and allocation margins shaped by institutional form.
