A Tale of Two Banking Regulations (2024)
2023 recipient of The C. Steven Cole Best Doctoral Student Paper in Financial Institutions and Markets (SWFA) (Submitted to Review of Accounting Studies)
I examine how changes to banks' regulatory burden, driven by changes surrounding size-based regulatory thresholds, impact liquidity creation. Using amendments in 2005 to the FDIC Improvement Act (FDICIA) and the Community Reinvestment Act (CRA), I document that treating each regulatory change as a separate event leads to confounding results. When I ignore the overlap of the two regulations, I find that liquidity creation increases following decreased regulatory burden. After disentangling the effects of the two regulatory changes, I show that banks increase their liquidity creation in response to the amendment to the CRA, but banks do not modify their liquidity creation in response to the change in the FDICIA. Finally, I find evidence that small business lending increases following the CRA regulatory change and find no evidence that the regulatory change had outcomes contradictory to the purpose of the CRA.
Beyond Benefits: Uncertainty and Sticky Information Costs (2024)
with Harrison Ham, Wang Renxuan, and Zhongjin Lu
Motivated by the ambiguous predictions of existing information choice theories, we propose and test the "Sticky Information Cost" (SIC) hypothesis to understand how investors acquire information in uncertain financial markets. SIC asserts that information processing costs for investors are influenced by a firm's slow-changing information environment, closely linked to its fundamental uncertainty. Using direct measures for information processing costs and the return predictability of analysts' biases as a proxy for information acquisition, we find opposite relationships between uncertainty and information acquisition when comparing across firms and over time. These results hold across various uncertainty measures and other earnings-related anomalies, supporting the SIC hypothesis while challenging existing theories. Incorporating the SIC into the existing information choice theories provides a new perspective on return anomalies.
Expectations Matter: When (not) to Use Machine Learning Earning (2024)
with John Campbell, Harrison Ham, and Zhongjin Lu (Revise and Resubmit at Management Science)
We comprehensively examine whether machine learning technology can meaningfully improve earnings forecasts, and if so, whether market expectations appear to reflect those superior forecasts. First, we use a consistent methodology to evaluate a comprehensive list of machine learning forecasts from 1990 to 2020. We find evidence that the best machine learning forecast outperforms analysts’ forecasts, but the improvement declines over time and is small when analysts face stronger incentives to be accurate. Second, using earnings response coefficient (ERC) tests, we infer that investors’ expectations put more weight on analysts’ forecasts than prescribed by the best machine forecast. Investors’ overweighting becomes statistically insignificant among large-cap firms with more sophisticated investors. Third, our time-series analyses suggest that analyst and machine forecasts are converging over time and that analysts’ information production remains critical, blurring the line between human and machine forecasts. Overall, our study provides an updated and comprehensive take on the most accurate earnings forecast and the best proxy for investors’ earnings expectations.
Beyond Green Labels: Assessing Mutual Funds’ ESG Commitments through Large Language Models (2024)
with Hieu Pham and Chaehyun Pyun (Submitted to Finance Research Letters)
This paper investigates whether mutual funds that adopt ESG-related names genuinely follow through on the implied increase in ESG commitments caused by this action. Utilizing Large Language Models (LLMs) to analyze mutual fund prospectuses, we find that funds significantly increase their ESG focus post-name change, leading to higher traditional ESG scores. However, the marginal benefit of additional ESG content in prospectuses diminishes after the name change. Our results suggest that while LLMs can be a cost-effective tool for assessing a fund's ESG commitment, investors should continue to exercise due diligence, particularly after an ESG-related name change.
Behind the Scenes of Activism Mergers
with Arjun Goel and Daniel Rettl
One Shot IPOs vs. Withdrawn IPOs
Banking and Social Capital
with Raymond Kim
Components of VIX
with Ansley Chua