Welcome to my homepage!

I am Professor of NUS Business School. I joined the National University of Singapore in July 2016. Before that, I was an Assistant Professor of Accountancy at the City University of Hong Kong. My works have been published in Accounting, Finance and Psychology journals, including The Accounting Review, Journal of Accounting Research, Journal of Accounting and Economics, Review of Accounting Studies, Contemporary Accounting Research, Journal of Financial and Quantitative Analysis, Management Science, Organization Science, Journal of Applied Psychology, and Nature: Human Behaviour. 

I was born in a small town on the Leizhou Peninsula in Guangdong Province and studied for six years at Mathematical Olympiad School (within the Affiliated High School of South China Normal University) in Guangzhou, where my early interest in mathematics and physics took shape. I received my bachelor’s degree in economics from Sun Yat-sen University and Ph.D. in finance from the National University of Singapore. My intellectual interests lie in banking, the role of information in insecure times, and Zen (禅).  

A book is but one among the myriad things of the world. My words are no more than a quiet page or two within the endless records of existence. Their meaning lies in the chance of finding readers whose hearts echo in sympathy. And so we wait, each of us, for the readers who are truly ours. 

🎉🎉🎉The paper I like most at the moment 

Voluntary Disclosure, Misinformation, and AI Information Processing: Theory and Evidence (with Jeremy Bertomeu, Yibin Liu, and Zhenghui Ni) 

Key Message: When humans are seduced by the convenience of AI and come to rely on it excessively, they may gradually lose control over information, ultimately drifting into the kind of "digital dystopia" that Huxley depicted in his novel Brave New World. 

ABSTRACT: This paper investigates the impact of generative AI on firms’ voluntary disclosure choices. Our theoretical model highlights a trade-off between AI’s improved ability to process disclosed information and its potential for misinformation, modeled as a random “hallucination” unrelated to the firms’ fundamentals. We predict that increased AI processing leads to more strategic non-disclosure due to two related economic forces. First, hallucinations provide additional camouflage after strategic non-disclosure. Second, because users consider the risk of misinformation, they discount observed marginal disclosures, further reducing the benefit of disclosure. To test our predictions, we leverage OpenAI’s launch of ChatGPT in November 2022 as a shock to AI processing. Consistent with the theory, firms with more AI processing reduce their voluntary disclosures. Further, the introduction of ChatGPT reduces information processing failures, which manifests in increased information processing speed. Combining the crowding-out effect on information supply and the positive impact on information processing speed, we do not find evidence of a net increase in information quality. 

🎉🎉🎉Latest Publications:

a) Benefits of Bank-CLO Relationships: Evidence from Bankruptcy and Restructuring Outcomes of CLO-held Loans. conditional accept The Accounting Review

Key Message: Despite persistent academic and regulatory concerns, Collateralized Loan Obligations (CLOs) have consistently shown resiliency over the past two decades. We document that firms whose loans are initially purchased by CLOs are less likely to experience negative credit events within the following 12 months, particularly when CLOs exhibit institutional relationships with originating banks (via repeated transactions) or when there is evidence of individual-level relationships (via personnel flows between the lead arrangers and CLOs). The effects of these relationships are more pronounced when originating banks possess superior private information about borrowers. We further observe that conditional upon filing for bankruptcy, borrowers whose loans are held by CLOs with institutional/individual level relationships are more likely to successfully restructure under Chapter 11. Collectively, our findings highlight a dual information channel, institutional and personnel-based, that mitigates information frictions in the leveraged loan market.

Methods: stylized facts + OLS

b) A Tale of Two Market Disciplines: How does Bank Financial Misconduct Affect Peer Banks in the Local Deposit Market (2026) Journal Accounting Research.

Key Message: We show local peer banks exhibit divergent deposit responses, contingent on how the misconduct is perceived by information recipients in different economic contexts. During normal periods, depositors receiving a negative signal about bank misconduct reallocate their funds from misconduct banks to local peers—a local reallocation effect that decreases deposit spreads and increases deposit inflows for peer banks. During financial crisis periods, however, bank misconduct leads to withdrawals from both misconduct banks and their peer banks – a local contagion effect whereby local peer banks face increased deposit spreads and deposit outflows following the misconduct.

Methods: DiD with high order fixed effect using high dimensional data.

c) The 2003 U.S. Dividend Tax Cut, Small Business Loan Supply, and the Real Economy (2026) The Accounting Review

Key Message: An important complement of Yagan (2015) that there is a credit supply-side effect of the U.S. 2003 dividend tax cut on the real economy through the banking sector. C-corporation banks (treatment group), particularly those capital-constrained, increase the supply of small business loans more than S-subchapter banks (control group) following the tax cut. The areas with a greater presence of C-corporation banks exhibit more small business formations, employment, and innovations. 

Methods: DiD (C versus S banks)

d) The Impact of Generative AI on Information Processing: Evidence from the Ban of ChatGPT in Italy (2025). Journal of Accounting and Economics. 

Key Message: The ban discourages Tech Analysts located in Italy from using ChatGPT to produce firm-specific information (It is all about identifying who in which time window uses what informational tool to produce what type of information).

Methods: DiD + Lab Exp + Textual analysis