Presentation: Bayes/Bristol Accounting PhD conference, 2024; EAA Doctoral Colloquium, 2024; EAA Annual Congress, 2024; Emerging Scholars in Accounting Conference, 2024; ESSEC, 2024; SSE, 2024; Yale University, 2024; 

Abstract: This study examines the substantial growth in female professional representation within the certified public accountant (CPA) profession during the 1970s. It investigates whether this increase in female participation can be attributed to Title IX, a policy enacted in 1972 to combat gender discrimination in education. My analysis shows that the entry rates for female CPAs increased more than those for male CPAs, following the passage of Title IX. Consistent with the mechanism of reduced educational barriers, this effect is driven by states that require specific accounting education as a prerequisite for CPA certification. The increase in female CPA entries is also more pronounced in states with significant reliance on federal funding from the Office of Education and in states with more historically male-only business fraternities. Overall, these results suggest that improvements in gender equality in education play a crucial role in enhancing female participation in professional labor markets.

Presentations: Yale University, 2024; FARS Midyear, 2025; ASE, 2025; Peking University, 2025

R & R at Journal of Accounting and Economics 

Abstract: This study examines whether the adoption of CPA firm mobility influences local audit markets. While this policy expands client access to audit services by allowing out-of-state CPA firms to enter local markets without registration and fees, there are concerns about low-quality entrants and compliance costs. Employing a stacked difference-in-differences design, we find that clients in states adopting CPA firm mobility experience lower audit fees and higher audit quality, particularly for clients of smaller audit firms. Additionally, the number of CPA offices increases post-adoption. Collectively, our findings inform the current regulatory debate on CPA firm mobility and highlight a positive effect of audit market competition on the quality of audit services. 

Presentations: SSE, 2025

R & R at Journal of Accounting and Economics

Abstract: This paper examines the effects of algorithmic bias audits on workforce diversity by exploiting New York City's (NYC) pioneering Bias Audit Law (Local Law 144). The law requires third-party audits for NYC employers that use algorithm-driven hiring tools to examine instances of recruiting bias. Using a difference-in-differences analysis, we find that bias audit requirements lead to a two percentage-point reduction in the share of male hiring, while having no significant effects on white employee hiring. The effects of decreased male hires are particularly pronounced among firms with greater pre-audit gender imbalances and higher AI intensity. We find that this decline is also concentrated in lower to mid-tier positions, and applies to both public and private companies. We also document costs in the form of longer time to fill vacancies. Overall, our results provide the first large-scale empirical evidence on the effectiveness of mandated algorithmic bias audits, highlighting a trade-off between levelling the playing field for job seekers and reduced recruitment efficiency. They highlight the value of audits beyond financial statements and show that audits can help address AI discrimination in employment practices.

Presentation: Accounting and Finance Association of Australia and New Zealand Conference, 2024 

R & R at Accounting Horizon 

Abstract: We investigate the temporal dynamics of R&D excess returns of U.S. firms over five decades. Unlike the commonly held implicit assumption that R&D excess returns are persistent across time, we document substantial time-series variations. Initially modest but significant in the 1970s and 1980s, R&D excess returns increased significantly during the 1990s, peaking in the early 2000s. From 2005, R&D excess returns declined sharply and were rarely significant. Focusing on a 22-year period (1983 to 2004) during which R&D excess returns were consistently significant, we find that the temporal variations in excess returns were more consistent with risk than mispricing explanations. We also find that value/growth and operating leverage anomalies mitigate R&D excess returns in the 1980s, but their impact is limited to that period and does not fully subsume R&D excess returns. We also find that temporal variations in R&D excess returns mimic macro factors associated with overall market sentiment and total business R&D spending, especially during the run-up to the dotcom bubble in the 1990s. We then address R&D nondisclosure by firms and find it significantly attenuates R&D excess returns. Finally, structural break tests confirm that R&D excess returns exhibit a significant systemic shift in 2004, following which excess returns are rarely significant. However, none of the examined variables are likely to explain this shift. Overall, we provide new insights into the existence of and temporal variations in R&D excess returns and underscore the importance of considering the dynamic interplay among firm-specific, market-wide, and disclosure factors in future studies.

Presentations: 5th Analyst Research Conference, 2024

Based on my first-year summer paper at the ESSEC

Abstract: We examine the impact of the closure of local newspapers, an important supplier of local information, on financial analysts’ ability to forecast the performance of firms in the regions of the closures. Analyst forecast accuracy declines for firms near a closed newspaper after the closure event, and the impact is more pronounced for firms with a worse ex-ante information environment and for newspapers with greater relative importance before the closure. This adverse effect fades away after the two most immediate forecasts issued after the closure event. In addition, we find a negative effect of local newspaper closures on nearby firms’ bid-ask spread, but such an impact is muted for those with analyst coverage. Overall, our results suggest that local newspapers serve as a provider of relevant information about nearby firms for outside stakeholders, but their informational role is unlikely irreplaceable, and analysts’ research efforts could mitigate the informational loss arising from local newspaper closures.