Photo by my wife Yang Yi in 2026 @ Lake Louise
Embedded Files
Xuyuanda Qi
Xuyuanda Qi
Email: xq622@nyu.edu
Cells: +86-157 2138 3500 (China)
I am an Assistant Professor of Finance at NYU Shanghai. My research interests lie primarily in theoretical corporate finance, venture capital contracts, financial innovation and mechanism design.
Working Papers
with Kostas Koufopoulos and Giulio Trigilia
Presented at FTG 2025, WFA 2025, SAET 2025, CFAM 2025, FIF 2026, MFA 2026, AsianFA 2026 (Scheduled)
Abstract: In this paper, we demonstrate that demandable debt provides an effective solution to the leverage ratchet effect without requiring any additional information beyond that assumed in the existing literature. Demandable debt-holders have an option to request full repayment of debt at any time. If the firm's leverage exceeds its target debt ratio, debt-holders will exercise their option and sell this excess debt back to the firm. This mechanism efficiently disciplines the firm to maintain the target debt ratio, except under extreme negative shocks leading to inevitable bankruptcy. Furthermore, we show that as the model's time intervals shorten, the firm can asymptotically achieve the full tax shield benefits without incurring any bankruptcy risk.
with Kostas Koufopoulos
Abstract: We study liquidity provision in the canonical Diamond-Dybvig environment and show that the two functions bundled in a demand deposit -- investment and liquidity insurance -- can be separated. Agents hold their long-term investments directly, for instance through open-ended unit funds, and hedge idiosyncratic liquidity risk through a consumer credit facility that lends for consumption at a subsidized rate, funded by a non-refundable fee, while imposing a high cash-withdrawal fee to deter withdrawals for non-consumption purposes. We prove that this arrangement implements the first-best risk-sharing allocation as the unique equilibrium. Because each agent retains her own claim to long-term output, patient agents never borrow early and panic-driven runs cannot arise. Moreover, because subsidized credit is tied to consumption, agents cannot exploit it for arbitrage through financial markets. Credit cards and buy-now-pay-later arrangements share the key features of the facility, suggesting that frictions beyond liquidity provision are important for explaining the continued prevalence of demand deposits.
with Kostas Koufopoulos and Giulio Trigilia
Presneted at Dan Bernhardt's 10 years in Warwick Conference, 2025 Tsinghua Theory and Finance Workshop, CES 2026 (Scheduled)
Abstract: The Coase conjecture holds that a monopolist selling a durable good cannot sustain prices above competitive levels when she can make frequent offers. We show that this result can be overturned if the monopolist employs a smart contract that issues securities granting a third party a reward if trade occurs after the initial date. To restore dynamic commitment power in a setting in which all agents can costlessly renegotiate, the smart contract randomizes the issuance of the security to an anonymous third party while concealing from the monopolist whether it was issued. This generates an endogenous informational asymmetry that deters future trade in any Perfect Bayesian Equilibrium. Under general renegotiation protocols, the monopolist can approximately recover static monopoly profits with a sufficiently large reward. However, as the low-valuation consumer’s valuation converges to marginal cost, even an infinitesimal security reward can allow the monopolist to approximate static monopoly profits. This eliminates the classical discontinuity between the “gap” and “no-gap” cases. The mechanism can be implemented on the blockchain, which makes it robust to cyberattacks.
Cursed with Knowledge? Strategic Risks of AI-Powered Monitoring
with Yang Yi
Abstract: Why are some financial institutions slow to adopt Artificial Intelligence (AI) despite its decreasing costs, while others are quick to embrace it? This paper develops a theoretical framework to understand the tradeoff between using AI for better borrower evaluation and maintaining strict financial discipline. Although AI improves the ability to assess borrower quality during refinancing, it can also encourage financiers to refinance initially speculative borrowers, thus exacerbating adverse selection in initial lending. To avoid this risk, some institutions may optimally choose not to adopt AI. Our model predicts that lower AI costs drive adoption only when this selection risk is not severe. Moreover, AI adoption is most beneficial for institutions with a moderate-quality lending pool. These results help explain the strategic dynamics of AI adoption in credit monitoring.
with Chen Chen
Minor Revision at Operations Research
Presented at Shanghai Theory Day 2024, INFORMS Revenue Management & Pricing Conference 2025, and INFORMS Annual Meeting 2025
Abstract: We study an information design problem in which a sender seeks to allocate an indivisible good to n receivers by strategically disclosing information. The good can be allocated to at most one receiver, and each receiver decides whether to accept it based on self-interest. Relevant applications include school advisors promoting students for job placements and incubators introducing startups to potential investors. The sender can either send different signals to different receivers (i.e., private persuasion) or broadcast the same signal to all receivers (i.e., public persuasion). After receiving signals, receivers may communicate with one another in their self-interest to further reduce uncertainty about the good. We demonstrate that when the sender has a known preference among the receivers, public persuasion is optimal, regardless of how the receivers communicate. The optimal public persuasion can be derived from a first-best relaxation problem that imposes only the receivers' participation constraints. We then focus on a special case in which the good's characteristics are captured by a one-dimensional variable, and all receivers' utility functions are linear in this variable. Using a dual approach, we derive optimality conditions for persuasion mechanisms. This leads to closed-form optimal mechanisms for the two-receiver case and an explicit characterization of the set of optimal mechanisms in the general case, thereby enhancing our understanding of their structural properties.
with Bowen Lou, Jiding Zhang, Chen Jin and Liangfei Qiu
Major Revision at Information Systems Research
(Full paper available upon request)
Abstract: This study examines the market forces driving Bitcoin’s interconnected exchange and mining markets, developing a systematic framework that captures the interplay between exchange rate, liquidity, and mining activities. We model how fluctuations in the Bitcoin exchange market affect the entry and exit dynamics of the mining market, characterized by the supply and demand of mining rigs. Empirically, leveraging a unique dataset of the purchase and listing of mining rigs from a leading e-commerce platform, we find that the effects of the exchange market conditions are pronounced for investors, as potential entrants to the Bitcoin mining market. They have a significant impact on the demand for mining rigs. Bitcoin investors tend to associate a higher valuation of Bitcoin with a higher Bitcoin exchange rate and consider the liquidity of the market when making mining decisions. In addition, our empirical results indicate that the electricity consumption of mining Bitcoin moderates the effects, shaping the responsiveness of mining rig demand to the exchange market conditions. To further contextualize these findings, we develop a game-theoretic model that uncovers the micro-foundations of investor decision-making and the role of certain types of miners in market stabilization. Overall, this study elucidates the economic implications of Bitcoin exchanges on Bitcoin mining activities, which contribute to the operational performance of the Bitcoin system. It enriches our understanding of how investor activities are affected and coordinated within IT-enabled markets, while illuminating the information transmission and underlying connection among them.
Optimal Staging of Early Startups
Abstract: When early startups stage the financing of their capital investments, they are at risk of being severely diluted by venture capitalists later on. It is puzzling that early startups do so in a competitive financial market. This paper shows that staging is beneficial to an early startup with a small upside return and a high capital intensity of early development. In this case, absent staging, the entrepreneur gets a small number of shares, which provides him with weak incentives to increase the startup's value. If the financing is staged, reevaluation of the startup during the follow-on round incorporates all the nonverifiable information about interim performance. Staging provides additional incentives since the entrepreneur gets more shares when interim performance is better. Between round financing and tranched financing, the two most prevalent forms of staging, round financing generates stronger incentives for the entrepreneur, but a lower payoff to the venture capitalist, than tranched financing. As a result, with a smaller upside return and a higher capital intensity of early development, the venture capitalist is less likely to participate in round financing, and tranched financing is more likely to be used to ensure his participation. All the above results are robust in a mechanism design framework.
This figure shows how the choice of financing form depends on two empirically important parameters. The X-axis is the capital intensity of early development, and the Y-axis is the upside return of the startup. Staging is not beneficial in two cases. 1. When the upside return is huge, staging is generally detrimental, and no staging is chosen. 2. When the upside return is minimal and the capital intensity of early development is high, the startup cannot attract any financing, and staging is trivially not helpful. Otherwise, staging is feasible and beneficial. Between two prevalent forms of staging, tranched financing is more likely to be chosen with small upside return and high capital intensity of early development; round financing is more likely to be chosen otherwise.
When do incomplete contracts matter?
with Aristotelis Boukouras, Kostas Koufopoulos and Giulio Trigilia
Abstract: This paper derives conditions under which the introduction of a third-party agent solves the renegotiation-proofness problem of Moore and Repullo (1988)-type mechanisms, without introducing the potential for other agents to collude with the third-party. The key novelties of our mechanism are: (i) the introduction of a third-party agent only off-equilibrium and with some probability; (ii) the fact that both its existence and its identity are unknown to the other agents. We show that under these conditions, which are satisfied in many empirical applications, a hidden third-party agent can restore the implementation of the efficient allocation. If this agent does not observe the state of the world, we provide a sufficient condition for implementation to succeed.
Leverage Dynamics with Reputation
Presented in the FTG 2019 Summer Program
This work analyzes the classic trade-off theory of capital structure in a dynamic model where the firm does not have any commitment power. The equilibrium analyzed in this paper depends on the firm's whole history (reputation) instead of the firm's current income and debt level. This paper proves that under this non-Markov structure, the firm may repurchase its outstanding debt, which breaks the Leverage Ratchet effect discussed in Admati, DeMarzo, Hellwig, and Pfleiderer (2018). This paper also proves that under mild conditions, the equity value in a non-Markov equilibrium can be higher than the equity value in the Markov equilibrium, the one depicted in DeMarzo and He (2021). Interestingly, with some conditions, the firm can achieve the first best equity value as if the firm has commitment power. In this case, reputation is a perfect substitute for commitment power.
This figure plots the revised Leverage Target policy that maximizes the equity holders' value as if the firm has commitment power. When the firm's income is not too low, the firm maintains its leverage target. In other words, the firm issues more debt when its income rises and repurchases the outstanding debt when its income drops. This breaks the Leverage Ratchet Effect discussed in Admati, DeMarzo, Hellwig and Pfleiderer (2018). When the firm's income is too low, the firm does not issue or repurchase any debt. In this case, if the firm's income goes up later, the firm starts to maintain the leverage target again; and if the firm's income drops further, the firm defaults as predicted in Leland (1998).
Pre-Ph.D. Publication
Page updated
Google Sites
Report abuse