Job Market Paper, single-authored
Won 2025 Dyckman Research Grant Awards
Abstract: This study examines the effects of international harmonization of accounting standards—specifically, the Expected Credit Loss (ECL) model—on credit market dynamics in informal economies. Motivated by global initiatives to improve financial reporting transparency, the shift to the ECL framework altered lenders' risk-assessment practices. Using the staggered rollout of ECL in Ghana, I find that adopting banks operating in more-informal communities increase loan loss provisions by about 3% and reduce lending by about 15% relative to adopting banks in less-informal areas. The contraction in lending is driven by reduced non-collateralized loans to households and small enterprises alongside higher interest rates and compliance costs. Importantly, these adjustments are not explained by realized losses, which decline, but by greater documentation frictions under the new standards. These shifts correspond to reduced economic activity in more informal areas. My findings show that global accounting harmonization, while intended to improve transparency, can unintentionally exacerbate financial exclusion where information frictions are most severe.
Presentations: Emerging Scholars in Accounting Conference at Frankfurt School of Management (Oct 2025), Conference on The Role of Accounting and Information Frictions in Microenterprises (Rice University-Sep 2025)
Under Review at The Accounting Review, with Mary Adenale (UT Austin), Tendai Masaya (Penn State), and David Park (CUHK, Shenzhen)
Abstract: Section 1502 of the Dodd-Frank Act requires SEC filers to disclose their use of conflict minerals in their products to ensure supply chain transparency on minerals that could contribute to financing armed groups. Motivated by the continuous decline in conflict minerals disclosures (CMD) filings, we examine the effect of private pressure on the disclosures. Specifically, we investigate whether and how exerting private pressure on firms through their grievance mechanisms, in the form of an anonymous person informing a firm that one of its suppliers is, in fact, sanctioned, affects its subsequent CMD choices. Using a randomized field experiment, we find that informing firms of the potential sanction violation causes them to remove the sanctioned entity from, or explain why it remains in, their CMD. Providing a disclosure example explaining why a firm may have a sanctioned entity in its CMD increases the focal firm’s likelihood of explaining rather than removing the entity from its disclosures. We also analyze firms' engagement levels with our inquiries and find that many firms actively investigated the concerns we highlighted via their grievance mechanisms. Overall, we find some evidence that private pressure through grievance mechanisms affects corporate disclosure choices.
Presentations: HARC 2025, AAA 2025, 2024 Midwestern Accounting Conference
Working Paper, with John Donovan (Notre Dame) and Arthur Morris (HKUST)
Abstract: We study how loan pricing uncertainty affects the design of private debt contracts. The transition from LIBOR to SOFR provides a shock to loan pricing uncertainty that is unrelated to borrower fundamentals or creditworthiness. Using a loan-path panel of syndicated loans, we document that switching from LIBOR to SOFR is associated with a temporary increase in the use of financial covenants. This effect is concentrated in performance covenants, suggesting that lenders rely more heavily on accounting-based triggers to manage benchmark uncertainty until it can be resolved ex-post. We also find that the increase in covenant use is amplified when lenders have greater bargaining power. Our results highlight that covenants are not only mechanisms for resolving agency conflicts, but also flexible tools that lenders employ to manage periods of heightened loan pricing uncertainty.
Working Paper, with Albert Mensah (HEC Paris)
Status: Data analysis, with Samuel Chang (Chicago Booth), Hans Christensen (Chicago Booth), and Donald N’Gatta (MDE Business School)
Status: Data collection, single-authored