Knowledge that Transforms

We conduct a comprehensive study on momentum spillovers in the Chinese stock market using various types of economic linkages, with particular attention to momentum spillover via news co-mention linkages. We utilize millions of Chinese business news articles and develop a flexible and innovative algorithm to identify linkages among listed firms. We find that news co-mention momentum spillover is stronger than others, unifying various forms of momentum spillover effects in the Chinese market and replacing the role of analyst co-coverage in the U.S. News co-mention identifies a wide range of economically important linkages, particularly recovering more cross-industry linkages than other link identification methods, which contributes to its strong performance.

We investigate whether air pollution affects household stock market participation. Using representative household survey data in China, we find that households significantly reduce their financial risk taking when exposed to heavy air pollution. This finding is robust to an instrumental variable analysis and a regression discontinuity test based on the Huai River policy. To explore the underlying mechanisms, we find that air pollution significantly impairs health, increases unemployment risk, impedes cognitive performance, and dampens mood. We further show that these channels explain a non-trivial proportion of the relationship between air pollution and stock market participation. Our study sheds light on the far-reaching influence of air pollution on households’ well-being from the perspective of portfolio choice.

Does factor momentum drive stock price momentum? We examine this relationship across 51 countries. Factor momentum proves strong across many markets and international portfolios, independent of typical return predictability drivers. However, its ability to capture stock momentum profits depends on methodological and dataset choices. Empirical factor momentum cannot entirely subsume stock or industry momentum globally. Conversely, price momentum often better explains its factor counterpart than vice versa. Notably, factor momentum based on principal components is more robust, capturing a major share of price momentum gains in developed and emerging markets. Our findings challenge the view that momentum merely times other factors rather than constituting a distinct anomaly.

We examine whether analyst coverage and mutual fund trades are complements or substitutes in the course of information incorporation into stock prices. Our empirical evidence indicates that they are complementary. Clustered trades in stocks with low analyst coverage is associated with a subsequent return reversal, which is more pronounced among less actively managed mutual funds. Mutual fund herding under low analyst coverage also amplifies future stock price crash risk through decreased corporate disclosure quality. These negative effects of mutual fund herding are not apparent for stocks with high analyst coverage. To address potential endogeneity concerns, we conduct additional tests using brokerage firm mergers and closures as exogenous shocks to analyst coverage and find consistent results. Our findings highlight the role of analysts in mitigating price-destabilizing herding behavior of mutual funds.

This paper examines complexity in CEO compensation contracts. We develop a measure of compensation complexity and provide empirical evidence that complexity has increased substantially over time. We document that complexity results not only from factors reflecting efficient contracting, but also from external pressures from compensation consultants, institutional investors, proxy advisors, and attempts to benchmark to peers, with these external factors having greater impact in more recent years. Examining consequences of contract complexity, we find an association with lower future firm performance that is related to the influence of external factors on compensation design. We further find this relation is partially mitigated when a contract's performance metrics are more highly correlated, consistent with information processing costs hampering decision-making. Collectively, these findings confirm concerns raised by investors and the media regarding compensation complexity and can inform boards in their design of CEO pay packages.

We offer empirical evidence on the role of accounting information in shaping the incentives to produce fake news. We document that fake news authors strategically (1) publish their articles near earnings announcements, leveraging the widespread market attention these events attract, and (2) within the near-announcement window, avoid publishing post-announcement when investors are less susceptible to fake news due to the disclosure of accounting information. In extending our analyses to the broader accounting information environment, we find that fake news authors are less likely to target firms with more robust accounting information and elicit lower market reactions when doing so. These results highlight both ex-ante and ex-post roles that accounting information plays in safeguarding firms from financial disinformation.

We use textual analysis of mandatory accounting filings to develop firm-level, time-varying measures of exposure to individual government agencies including the Securities Exchange Commission (SEC) and Internal Revenue Service (IRS). The measures vary predictably across industries and with agency-specific events such as the Sarbanes Oxley Act at the SEC and budget cuts at the IRS. The measures positively relate to undisclosed agency investigations and financial statement downloads. Firms' total exposure across government agencies negatively relates to their profitability, consistent with exposure to government agencies imposing net costs. Consistent with a causal interpretation of these results, the positive stock market reaction to the surprise election of Donald Trump, who promised to reduce the power of government agencies, positively varies with firms' exposure to government agencies. As initial applications of our measures, we demonstrate that expanded SEC oversight increases firms' stock liquidity and reduced IRS oversight decreases firms’ effective tax rates.

We use data from TransUnion, a large U.S. credit bureau covering millions of individual consumer loans, to examine the transition to the Current Expected Credit Loss (CECL) accounting standard and to provide novel evidence about the impact that raising reserve requirements has on banks’ pricing and lending decisions in the U.S. consumer lending market. We find that greater reserve requirements following the adoption of CECL induce a statistically significant but economically moderate increase in loan interest rates. The effects are more pronounced for weakly-capitalized banks and even more so for underprivileged individuals borrowing from weakly-capitalized banks. Our evidence informs the ongoing policy debate between standard setters and members of the financial industry about the potential effects of CECL on credit markets.

We predict that institutional investors in loan syndicates charge bellwether firms lower loan spreads as compensation for having access to private information that can help identify trading opportunities in other firms' public market securities. Consistent with this prediction, when lending to bellwether firms, institutional investors charge a loan premium that is between 17 and 25 bps lower relative to non-bellwether firms, and earn annualized excess returns of 1.5–2.2% from trading in other firms' securities. Findings from cross-sectional analyses reveal that the reduction in loan spread is larger when the value of private information from bellwether firms is higher. Additionally, institutional lenders' excess returns are lower when lending to more transparent bellwether borrowers and when they pay a lower price—as reflected in loan spreads—in exchange for the private information, supporting the notion that the value of private information relates to institutional investors’ trading performance.

This study explores the relationship between executive compensation and the preference dynamics of managers and shareholders. Our analysis centers on the theoretical prediction that changes in firms' asset value can differentially affect the risk-taking preferences of the two groups, potentially influencing the optimal compensation policy. Utilizing local real estate price changes to identify variations in firms’ asset value, we find that a decrease in asset value leads to more risk-taking incentives in compensation, and this effect is more pronounced in firms that are more likely to be influenced by the hypothesized mechanisms. In the second empirical setting, we provide corroborating evidence using a natural experiment involving disaster-induced negative shocks to the firm fundamentals. Collectively, our findings suggest that the design of compensation contracts facilitates incentive alignment by incorporating the dynamic preferences of the contracting parties.