Research

WORKING PAPERS



Abstract: This paper provides novel evidence of the time-varying roles of subjective expectations in explaining stock price variations across the market and 30 industry portfolios monthly from 1976 to 2020. Cash flow expectations matter more under financial uncertainty and recessions, especially among the hardest-hit industries such as Telecommunications during the Dot-com Bubble, Financials during the Great Recession, and Healthcare during the Covid-19 pandemic. Conversely, discount rates explain more price variations during the expansionary 2010s. Finally, inflation expectations, while accounting for 60% of price fluctuations in the high inflationary environment before 2000, play a negligible role thereafter. 


(with David Hirshleifer and Kuntara Pukthuanthong)

Abstract:  Using a semi-supervised topic model on 7,000,000 New York Times articles spanning 160 years, we test whether topics of media discourse predict future stock and bond market returns to test rational and behavioral hypotheses about  market valuation of  disaster risk.  Focusing on  media discourse  addresses the challenge of sample size even when  major disasters are rare. Our methodology avoids look-ahead bias and addresses semantic shifts. War discourse positively predicts  market returns, with an out-of-sample R-squared of 1.35%,  and negatively predicts returns on short-term government and  investment-grade corporate bonds. The predictive power of war discourse increases in more recent  time periods. 


(with David Hirshleifer and Kuntara Pukthuanthong)

Abstract:  We test whether a war-related  factor model derived from textual analysis of media news explains the  cross-section of expected asset returns. The war  factor  builds from a semi-supervised topic model to extract discourse topics from 7,000,000 New York Times stories spanning 160 years. We find that a war factor helps predict the cross section of returns  across a wide range of testing assets deriving from  both traditional and machine learning construction techniques, and spanning a wide range of 138 anomalies. These findings are consistent with assets that have  poor  returns during periods of heightened war risk receiving higher risk premia, or that an asset's  sensitivity to war risk captures investor mispricing of war risk.  The return premium associated with the war  factor is incremental to factors  from prominent factor models and other measures of news-based uncertainty. Our result is robust when a war factor is constructed as factor mimicking portfolio.


(with Kuntara Pukthuanthong and Guofu Zhou)

Abstract: We analyze the daily predictability of investor sentiment across four major asset classes and compare sentiment measures based on news and social media with those based on trade information. For the majority of assets, trade-based sentiment measures outperform their text-based equivalents for both in-sample and out-of-sample predictions. This outperformance is particularly noticeable in long-term forecasts. However, real-time mean-variance investors can only achieve economic gains using Bitcoin trade sentiment, suggesting the challenge of transforming sentiment into daily profitable trading strategies.


(with Arka Bandyopadhyay and Kuntara Pukthuanthong)

Abstract: We apply advanced natural language processing to develop a dynamic dictionary of artificial intelligence (AI). Using this dictionary, we construct a real-time index of AI attention from more than three million \textit{New York Times} articles. Firms having high exposure to AI have higher returns one month ahead and lower returns five to seven months ahead, suggesting initial overreactions to AI news and subsequent corrections. The connection between AI exposure and future returns is concentrated among non-big stocks, indicating that small AI stocks are more difficult to value. A long-short AI exposure portfolio among non-big stocks generates significant alphas against benchmark multifactor models.


(with Arka Bandyopadhyay and Kuntara Pukthuanthong)

Abstract: We provide causal evidence that the narrative of diversity from the \textit{New York Times} articles has nudged the corporations to choose female CEOs to be equitable in terms of gender of firm leadership. This channel is independent of the board gender diversity, which was mandated in 2017 in California and later repealed in 2022. Surprisingly, the diversity narrative channel fails to explain the election of Indian CEOs in several HiTech firms over the last few years. We argue that the election of Indian CEOs was motivated to create a favorable image of Tech firms to Indian and Chinese labor. We conclude that providing equity in firm leadership in terms of ethnic diversity is harder compared to gender diversity. 


Abstract: We conduct empirical tests of a simplified version of the ratio habit model developed in Abel (1990), in which habit is extended beyond the preceding period. We show that change in four-year consumption growth—the measure of consumption resulting from our ratio habit preference—explains the joint equity premium–risk-free rate puzzle with a risk aversion coefficient much lower than any existing consumption measures under the standard consumption model. This outperformance of our ratio habit model over the standard model is robust across 18 non-U.S. countries. From 1928-2017, change in four-year consumption growth encompasses other consumption measures in explaining the cross-sectional variation of expected returns on various portfolios and it is the only consumption measure that passes the robust tests of the factor risk premium proposed by Kleibergen and Zhan (2020). While our measure constructed from nondurables does better at pricing the equity premium and risk-free rate, our service-based measure outperforms in explaining the cross-sectional variation of stock returns.