Published and Forthcoming Papers:
Optimal Integration: Human, Machine, and Generative AI, Accepted at Management Science
I study the optimal integration of humans and technologies in multi-layered decision-making processes. Each layer can correct existing errors but may also introduce new ones. A one-dimensional quality metric – a decision-maker’s error correction capability normalized by its new errors – determines the optimal rule: deploying higher-quality technologies in later stages. Interestingly, the final decision-making layer may not achieve the greatest error reduction; instead, its role hinges on minimizing new errors. Human effort varies asymmetrically across layers—early stages prioritize error correction with lower effort, while later stages emphasize avoiding new errors with higher effort. Applying the model to artificial intelligence (AI) reveals that AI's generative capabilities make it more likely to serve as the final decision-maker, reducing the need for costly human input, but underscoring the risks of AI hallucination. The theoretical framework also extends to applications including repeated delegation, automation design, loan screening, tenure review, and other multi-layer decision-making scenarios.
Dynamic Runs and Bankruptcy Regulations, with Zhen Zhou, Accepted at Review of Financial Studies
We build a dynamic model of creditor coordination and study bankruptcy regulations. Ex post coordination around bankruptcy may make it more difficult for creditors to exit distressed firms, thereby worsening their ex ante incentive to stay invested. This trade-off shapes creditors' decisions to run, firms' choice to declare bankruptcy, and optimal bankruptcy regulations. Intriguingly, filing for bankruptcy early, thereby preserving more assets for latecomers, can prolong firm life. Furthermore, regulators' clawbacks on pre-bankruptcy repayments can be superior to firms' commitment to early bankruptcy filing. Our analysis generates implications for automatic stay, avoidable preference, banking regulations, and creditor seniority.
Bigger pie, bigger slice: liquidity, value gain, and underpricing in IPOs, with Yang Guo and Yuanzhi Li, Journal of Financial Market, in press
Since investor participation is essential for successful IPOs, we hypothesize that issuers share value gain from IPOs with IPO investors, resulting in IPO underpricing. We test the positive relation between value gain and underpricing from the liquidity angle, as improved liquidity via IPO increases firm value. We find supporting evidence that underpricing is positively related to the expected post-IPO liquidity of the issuer. Using two regulation changes as exogenous shocks to share liquidity before and after an IPO, we show that underpricing is more pronounced with better expected post-IPO liquidity or lower pre-IPO liquidity.
Equity Issuance Methods and Dilution, with Mike Burkart, Review of Corporate Finance Studies, 12(1), 78–130
We analyze rights and public offerings when informed shareholders strategically choose to subscribe. Absent wealth constraints, rights offerings achieve the full information outcome and dominate public offerings. When some shareholders are wealth constrained, rights offerings lead to more dilution of their stakes and lower payoffs, despite the income from selling these rights. In both rights and public offerings, there is a trade-off between investment efficiency and wealth transfers among shareholders. When firms can choose the flotation method, either all firms choose the same offer method or high and low types opt for rights offerings, while intermediate types select public offerings.
A Dynamic Model of Optimal Creditor Dispersion, Journal of Finance, 76(1), 267-316.
Borrowing from multiple creditors exposes firms to rollover risk due to coordination problems among creditors, but it also improves firms' repayment incentives, thereby increasing pledgeability. Based on this trade-off, I develop a dynamic debt rollover model to analyze the evolution of creditor dispersion. Consistent with empirical evidence, I find that firms optimally increase creditor dispersion after poor performance. In contrast, cross-sectionally higher-growth firms can support more dispersed creditors. Frequent debt renegotiation limits firms' ability to increase pledgeability by having more creditors. Finally, holding a cash balance while borrowing from multiple creditors improves firms' repayment incentives uniformly across all future states.
A Theory of Multiperiod Debt Structure, with Chong Huang and Martin Oehmke, Review of Financial Studies, 32(11), 4447-4500
We develop a theory of multiperiod debt structure. A simple trade-off between the termination threat required to make debt repayments incentive compatible and the desire to avoid early liquidation determines the number of repayments, their timing, and amounts. As firms increase their borrowing, they add periodic risky repayments from the back of the maturity structure, with the time between repayments increasing in cash-flow risk. Cash-flow growth or a significant risk-free cash-flow component limits the number of risky repayments. Firms with a significant risk-free cash-flow component choose dispersed maturity profiles with smaller, relatively safe repayments every period, rather than riskier periodic repayments.
Buying High and Selling Low: Stock Repurchases and Persistent Asymmetric Information, with Philip Bond, Review of Financial Studies, 29(6), pp.1409-1452.
Share prices generally fall when a firm announces a seasoned equity offering (SEO). A standard explanation is that an SEO communicates negative information to investors. We show that if repeated capital market transactions are possible, this same asymmetry of information between firms and investors implies that some firms also repurchase shares in equilibrium. A subset of these firms directly profit from repurchases, while other firms repurchase in order to improve the terms of a subsequent SEO. The possibility of repurchases reduces both SEOs and investment. Overall, our analysis highlights the importance of analyzing SEOs and repurchases in a unified framework.
Working Papers:
Share Issues versus Share Repurchases, with Philip Bond and Yue Yuan, Resubmission requested at Management Science
Almost all firms repurchase shares through open market repurchase (OMR) programs. In contrast, issue methods are more diverse: both at-the-market offerings, analogous to OMR programs, and SEOs, analogous to the rarely-used tender-offer repurchases are used by significant fractions of firms. Moreover, average SEOs are larger than at-the-market offerings. We show that this asymmetry in the diversity of transaction methods in issuances and repurchases and the size-method relation in issuances are natural consequences of the single informational friction of a firm having superior information to investors. Finally, repurchasing firms are likely maximizing long-term shareholders' payoffs rather than boosting short-term share prices.
A Theory of Shareholder Class Action
Shareholders sometimes participate in class action lawsuits against their own firms for failing to disclose adverse information. Paradoxically, these lawsuits typically result in a wealth transfer from innocent shareholders to the plaintiff class, while shareholders who benefited from the nondisclosure by selling shares before the bad news became public remain unpenalized. Although some argue that class actions create ex-ante incentives for disclosure, I show that this channel is largely ineffective. In many cases, nondisclosure is the dominant strategy for initial shareholders, as they do not bear the full cost of compensating the class. However, firms may still disclose moderately bad news strategically—both to influence litigation decisions by lawyers and to shield more severe nondisclosure from legal scrutiny. Increasing ex-post litigation penalty may further reduce disclosure incentive. Finally, I show that a higher presence of small, short-term shareholders increases both the likelihood of litigation and managerial disclosure, revealing a novel channel of corporate governance.
Market Structure of Intermediation
Finding good products often requires costly screening effort. A skilled intermediary can represent multiple consumers, generating economies of scale in effort costs. However, consumers still need to find skilled intermediaries, re-creating the original screening friction. I provide a model of intermediary market structure, focusing on the number of rounds, sector size, skill distribution, and effort choices along the intermediation chain. Interestingly, even when the intermediary sector is less skilled than consumers, efficient intermediation can still occur. Furthermore, a two-round intermediation chain makes it possible for intermediaries with different skill levels to self-select into different rounds, approximating the first-best outcome: consumers almost certainly receiving good products with negligible screening costs. Finally, even when intermediaries' skill types are unknown, thereby rendering their self-selection ineffective, a sufficiently long chain can restore first-best efficiency. These results shed light on the structure of social media influencers and the asset management industry.
Brown Spinning, with Yang Guo and Jianing Yuan (draft coming soon)
We analyze firms' spinoff choices concerning pollutive subsidiaries when shareholders, potential buyers, and managers exhibit heterogenous environmental preferences. Stronger environmental preferences among managers and shareholders result in more spinoffs. However, the subsidiary is typically sold to prospective buyers who exhibit a weaker environmental preference, leading to escalated pollution production after acquisition. Despite the more pollutive outcome, the ex-post spinoff threat from shareholders induces managers to adopt greener production levels proactively, even if shareholders have lower environmental concern. Notably, spinoffs diminish when agents experience disutility from economy-wide pollution rather than self-generated pollution.