Research

My research asks how financial contracts and market mechanisms shape information production, risk-taking, and real corporate decisions, and how capital structure and financial distress affect firms' operations, employment, and aggregate welfare. My recent work extends this agenda to automated market making, prediction markets, decentralized finance, and AI-driven financial markets.

Research statement (PDF)

Working papers

1. Prediction Markets as Data Marketplaces: The Price of Information under Automated Market Making (draft available upon request)

   With Zhan Pang

   We view a prediction market as a crowdsourced data marketplace. Traders supply information through their trades, and the resulting market prices provide forecasts for a broad range of data users. We characterize optimal liquidity for an automated market maker using the logarithmic market scoring rule (LMSR) to elicit costly information from a risk-neutral trader. The trader's expected trading profit equals liquidity times the reduction in the forecast's expected log-loss, making liquidity the price paid per unit of information. Greater liquidity encourages information acquisition but raises payments on every unit supplied. The market maker therefore acts as a monopsonist, setting liquidity below its marginal valuation of forecast accuracy. The market opens only when this valuation covers the full cost per unit of information.

2. Rational Expectations Equilibrium with an Automated Market Maker (draft available upon request)

   With Zhan Pang

   We solve the Grossman–Stiglitz rational expectations equilibrium with costly information acquisition for a market whose counterparty is an automated market maker. The market maker posts the logarithmic market scoring rule (LMSR), pools all orders into one batch, and prices the batch with the rule. Uninformed traders learn from the price they pay. At any fraction of informed traders a continuous equilibrium exists with a unique price map: the price reveals the aggregate order flow of informed and noise traders, and the market maker's liquidity only rescales the price. Noise trading shelters informed traders only up to a point: at a fixed positive share of informed traders, the value of information vanishes at both extremes of noise-trading volatility. Deep markets attract informed traders: once liquidity is large, more liquidity and more volatile noise trading both raise the share of informed traders. The market maker earns a profit in a shallow market and incurs a loss in a deep one. The market maker weighs forecast accuracy against expected loss and capital when posting its liquidity.

3. Automated Market Making and Liquidity Provision in Prediction Markets (draft available upon request)

   With Zhan Pang

   2026 Wabash River Finance Conference

   We model an automated market maker's liquidity as the price of information. Under the logarithmic market scoring rule the market maker pays that liquidity, in expectation, for each unit of entropy a trader resolves. Bearing that subsidy and valuing the accuracy the trade delivers, the market maker prices as a monopsonist: a deeper liquidity buys more information but raises the bill on every unit, so it marks the liquidity down below the marginal value it places on accuracy. Under sequential arrival, free entry decides how much the market learns, and need not reward early arrival when the market opens where the information acquisition barely covers its cost through trading.

4. Put Credit Rating Agency's Money Where Its Mouth Is

   With Zhan Pang and Alexei Tchistyi

   Revise and Resubmit, Management Science

   2022 Finance Theory Group Spring Meeting; 2023 American Finance Association Annual Meeting; 2023 American Real Estate and Urban Economics Association International Conference, Cambridge

5. Rare Disaster Information Paradox

   With Peter DeMarzo and Alexei Tchistyi

   2022 American Finance Association Annual Meeting; 2022 SFS Cavalcade North America

6. A Model of Capital Structure Under Labor Market Search

   2018 SFS Cavalcade North America; 2018 Wabash River Finance Conference; 2019 Finance Theory Group Summer School; 2020 Midwest Finance Association Annual Meeting

7. Credit Risk in General Equilibrium: The Case of Equity-Maximizing Capital Structure

   With Tim Johnson and Chelsea Yu

   Previously titled: The Private and Social Value of Capital Structure Commitment

   2018 UBC Winter Finance Conference; 2018 SFS Cavalcade North America; 2018 Western Finance Association Annual Meeting; 2018 Northern Finance Association Annual Meeting

8. Executive Pay-for-Performance Sensitivity and Stochastic Volatility

   With Shuaiyu Chen and Yan Liu

   2022 SFS Cavalcade Asia

9. The Contract Year Phenomenon in the Corner Office: An Analysis of Firm Behavior During CEO Contract Renewals

   With Yuhai Xuan

   Minnesota Corporate Finance Conference; Drexel 8th Annual Academic Conference on Corporate Governance; 2015 Brigham Young University Red Rock Finance Conference; 2016 American Finance Association Annual Meeting

Publications, forthcoming and conditionally accepted papers

1. Corporate Resiliency and the Choice between Financial and Operational Hedging

   With Viral V. Acharya, Heitor Almeida, and Yakov Amihud

   Previously titled: Efficiency or Resiliency? Corporate Choice between Operational and Financial Hedging

   Conditionally accepted, Journal of Financial and Quantitative Analysis

2. How Does Health Insurance Affect Firm Employment and Performance? Evidence from Obamacare

   With Heitor Almeida, Yuhai Xuan, and Ruidi Huang

   Previously titled: The Impact of Obamacare on Firm Employment and Performance: Theory and Evidence

   Forthcoming, Management Science

3. The Client Is King: Do Mutual Fund Relationships Bias Analyst Recommendations?

   With Michael Firth, Chen Lin, and Yuhai Xuan

   Journal of Accounting Research 51, 165–200

4. Inside the Black Box: Bank Credit Allocation in China's Private Sector

   With Michael Firth, Chen Lin, and Sonia Wong

   Journal of Banking and Finance 33, 1144–1155