Conference Agenda

We study the effect of a firm's new information disclosure on the information asymmetry between its informed and uninformed investors and its liquidity. To do this, we employ advanced natural language processing (NLP) methods to introduce a novel measure of firms' 10-K filing predictability that quantifies the amount of new information in these reports. Our findings show that more new information is associated with higher bid-ask spreads and lower trading volumes, indicating increased information asymmetry and reduced liquidity, respectively. Notably, institutional ownership moderates these effects, suggesting that sophisticated investors can mitigate the adverse consequences of disclosure unpredictability. An event study analysis further reveals that more new information triggers increased trading activity and abnormal returns immediately after disclosure, though these effects are short-lived.

I propose creating a small set of well-diversified high-dimensional basis portfolios such that stocks within (across) portfolios have the most (least) similar fundamentals, proxied by a large set of characteristics. If the comovement between stocks is a function of a large set of characteristics, the high-dimensional basis portfolios that are distinct in all characteristics show low comovements and high dispersion in expected returns. As a result, the optimal portfolio spanned by high-dimensional basis portfolios displays a sizeable out-of-sample Sharpe ratio of 1.78 with a monthly alpha of 1.71% (t = 11.11), without taking any extreme position on any asset.

Using detailed and comprehensive bank data, we study the impacts that children have on gender gaps in financial choices. It is well established that women are, on average, less likely to participate in risky asset markets and save less. We show that the arrival of children contributes to the gender gaps in these financial choices: At the exact point in time when women become mothers their propensity to participate in risky asset markets drops and their propensity to save through savings accounts does as well, the amounts they hold in savings accounts is reduced as well as their average monthly savings, and they draw down their private pensions. These outcomes are unaffected as men become fathers. We therefore conclude there are “child penalties” in personal finances that contribute to the gender gaps in financial choices.

Order flow internalization by wholesalers has increased in the last decade. Retail order internalization decreases liquidity in lit exchanges, but it may also favor liquidity provision to sophisticated investors. Consequently, their incentives to acquire information will be affected, and this has consequences for price informativeness and market efficiency. We measure short and long term information acquisition and find that the proportion of internalized volume is positively associated with information acquisition at both horizons. We test whether wholesalers use internalized retail order flow to provide liquidity to sophisticated investors showing that the positive effect of internalization on information acquisition is stronger for those stocks preferred by institutional investors and by retail traders. Finally, we use the retail trading boom that started in late 2019 as an exogenous shock to internalization and an instrumental variables approach to address causality and the analysis supports our previous results.

Using a novel dataset from Ant Forest, a green fintech app in Alipay, we explore how gamified social interactions influence the users’ green investment decisions. We find that the users’ green preference, measured by daily low-carbon activities, is enhanced when they engage more in gamified social interactions designed for environmental education, thereby increasing their investment proportion in green mutual funds. Our findings are stronger among male and younger users and those less involved in environmental conservation actions. Our study provides the first mechanism in which gamified social interactions facilitate green investments by enhancing individuals’ green preferences.

Data is non-rival: a firm's data can be used simultaneously by others, and information about its customers benefits other firms even across industries. How is data being shared? Using granular information on mobile app usage, functionalities, and connections with data analytics platforms, we uncover a network of inter-firm data flows. Data sharing generates comovements in operational, financial, and stock-market performances among data-connected firms, beyond what traditional economic linkages can explain, and induces strategic complementarity in firms' product-design choices. Apple’s App Tracking Transparency policy, which restricts inter-firm data flows, weakens these patterns, providing causal evidence of the role of data sharing. To explain these findings, we develop a dynamic network model of data economy, where firm growth becomes interconnected through data sharing. The model introduces a network-augmented Gordon growth formula to value data-generated cash flows, capturing direct and indirect network externalities over multiple time horizons. Our metrics of valuation centrality identify systemically important firms that disproportionately influence the data economy due to their pivotal positions within the data-sharing network.

Open banking enables small businesses to share their bank financial data with potential lenders. I examine the effect of open banking on collateralization in small business lending. For identification, I exploit institutional features of the UK’s open banking policy that creates a discontinuity in firms’ eligibility to share data. Using a novel loan-level dataset covering the entire UK secured business loan market, I document that open banking eases the pledge of assets like accounts receivable and inventory. Firms eligible to share data are more likely to pledge such assets as collateral, thereby improving their access to credit. These effects are more pronounced for firms facing greater information asymmetry and those with greater information available to share. These findings highlight the role of open banking in reducing collateral constraints by mitigating information asymmetry.

We offer a novel resolution to several asset pricing puzzles by investigating how complexity affects pricing errors when rational, risk-averse agents have imperfect knowledge of the data-generating process. Our theoretical framework yields three key implications as complexity increases: (1) equilibrium pricing errors grow systematically larger, (2) the optimal portfolio increasingly exploits estimation error components rather than fundamental risk, and (3) multiple strategies achieve higher Sharpe ratios while maintaining low cross-correlations. Our model explains the limited pricing power of parsimonious factor models, the weak relationship between betas and average returns, and the proliferation of anomalies. Empirically, we document substantial complexity in return predictability and covariance structures. Analyzing sophisticated quantitative strategies, we find remarkably low correlations, with an average $R^2$ below 1\% among systematic hedge funds' active positions, consistent with our model's prediction that different strategies exploit distinct dimensions of estimation error in complex markets.

We construct managed portfolios based on textual analysis of firms' earnings call transcripts and investigate their asset pricing implications. Loadings on the text-managed portfolios can explain a comparable amount of stock-level return variation as those on the conventional characteristics-based factors. Combining the earnings call text and firm characteristics enhances the conditional mean-variance efficiency of factor portfolios but not the predictive power for stock returns. Drawing on the insights of Kozak and Nagel (2024), our evidence suggests that the earnings call text contains information about the return covariances that are missing from characteristics. Text-managed portfolio returns correlate with investor sentiments and forecast macroeconomic outcomes.

This paper proposes a novel measure of investment companies’ reliance on generative AI, focusing on its implications for hedge funds. We document a sharp increase in generative AI usage by hedge funds after ChatGPT’s 2022 launch. A difference-in-differences test shows that hedge funds adopting generative AI earn 3-5% higher annualized abnormal returns than non-adopters. We further identify this effect by exploiting ChatGPT outages as exogenous shocks. The outperformance originates from funds’ AI talent and ChatGPT’s strength in analyzing firm-specific information. Non-hedge funds yield no significant returns from AI adoption, suggesting generative AI may widen existing disparities among investors.

This paper shows that new technologies can help asset managers attract capital inflows. Exploiting information from their websites’ codes, I track when fund managers start collecting and analyzing customers’ data using tools like Google Analytics. Funds adopting such technologies attract 1.5% higher annual flows, with the effect being concentrated in retail share classes. Additionally, they expand product offerings and charge higher fees. The effects decrease with competition as more funds within the same category adopt similar technologies. Overall, these results highlight that technological innovation in asset management extends beyond portfolio allocation decisions to impact how funds attract and retain capital. This evidence underscores the economic importance of managers learning from customers’ data.

We analyze a model of a monopolistic informed investor who receives private information sequentially and faces a post-trading disclosure requirement. We show that this trading model can be transformed into a fictitious consumption-saving model with a borrowing constraint. Hence, insights from the consumption-saving literature can be adapted for the trading model. For example, analogous to the insights from the permanent income hypothesis, the informed investor ``saves'' more of his current information when expecting less future information advantage (``saving for rainy days'') or more uncertainty about it (``precautionary saving'') and smooths his information ``usage'' over time (``consumption smoothing'').

Conventional wisdom holds that informed investors benefit from colluding in their trading. However, we show that this may not hold when investors face uncertainty about other traders’ behavior. In a Kyle (1985) framework, we compare trading profits under collusive and competitive equilibria when informed investors face uncertainty about liquidity trading volatility. While low uncertainty favors collusion, we show that the expected profit of an individual investor under competition can be higher than the total profits for all investors under collusion when uncertainty is sufficiently high. This finding cautions against relying solely on profits to detect collusive behavior.

In this study, I examine the value of information from sell-side analysts by analyzing a large corpus of their written reports. Using embeddings from state-of-the-art large language models, I show that textual information in analyst reports explains 10.19% of contemporaneous stock returns out-of-sample, a value that is economically more significant than quantitative forecasts. I then perform a Shapley value decomposition to assess how much each topic within the reports contributes to explaining stock returns. The results show that analysts' income statement analyses account for more than half of the reports' explanatory power. Expressing these findings in economic terms, I estimate that early acquisition of analysts' reports can yield significant profits. Analysts' information value peeks in the first week following earnings announcements, highlighting their vital role in interpreting new financial data.

We uncover financial professionals’ mental models—the narratives they use to explain their subjective beliefs. Using 82,000 equity reports, we prompt large language models (LLMs) to extract 3.5 million narratives, each combining a topic, valuation channel, sentiment, and time outlook. To validate the reliability of our output, we introduce a multi-step LLM-based approach and new diagnostic tools. We establish three sets of findings. First, narratives are centered around a limited set of topics, primarily focused on top-line items, with variation in topic focus over time and across industries. Narratives are mostly forward-looking, with three times as many arguments focusing on the future as on the past. Second, differences in topic focus, sentiment, and time outlook across forecasters strongly predict the extent of disagreement in their subjective quantitative forecasts. Lastly, time-series variation in the average narrative’s sentiment and in the average narrative’s focus on top-line items closely track Shiller’s CAPE ratio (ρ = 0.84 and ρ = 0.42), and the cross-sectional variation in narratives predicts key asset pricing patterns. Narratives associated with growth stocks are more optimistic and forward-looking than those for value stocks, consistent with forecasters (mis)perceiving growth stocks as having above-average growth potential. Overall, this paper helps bridge the gap between ‘what forecasters believe’ and ‘why they believe it.’

We provide a comprehensive analysis of trading by major trader groups, both inside and outside firms, around the onset of cyberattacks and the associated market dynamics. Besides hackers, no one should know about the incoming cyberattacks. However, we find a perplexing pattern: abnormal shortselling, including the more likely information-motivated retail short selling, significantly increases in the weeks preceding data breaches, more so for stocks with a diverse pool of share lenders. No similar patterns are found for the detection and public announcement of cyberattacks, pseudo-cyber events, or among industry peers. Insiders neither can possess advanced knowledge of attacks nor engage in trading around them, and neither do institutions. Retail investors, despite being the least sophisticated, presciently sell attacked stocks alongside short sellers. This activity coincides with spikes in Google searches for keywords such as “cyberattacks.” Following cyberattacks, attacked stocks experience low returns, implying $400 million wealth transfer between informed and uninformed traders, while short-selling fees, effective spreads, and price impacts increase significantly. Our results suggest that cyberattacks, potentially representing the tip of the iceberg of information events originating outside firms, shake the traditional paradigm of information asymmetry premised on insiders’ informational advantage over outsiders.

We examine the impact of financial influencers (finfluencers) by analyzing real equity and derivative investments across four Nordic countries, with an instrument that randomly assigns influencers to followers. We find that (1) investors subscribe to influencers with high Sharpe ratios, frequent trades, shared country of residence or language, and male gender; (2) Influencers affect followers' portfolios and trading behaviors, especially if they have many followers, central network positions, or engage in many group discussions. Their impact is higher among female investors and those exposed to fewer influencers, or when trading passive funds; (3) Some influencers seem to monetize their influence.

This paper investigates the impact of Socially Responsible Investment (SRI) capital on the polluting practices of industrial Multinational Enterprises (MNEs) across all their facilities. We leverage the inverse relation between local pollution and high-frequency satellite-based measurements of local vegetation health through the normalized difference vegetation index (NDVI). Our empirical analysis encompasses a comprehensive dataset of 911 parent companies and 52,806 establishments worldwide. We estimate how the within cell panel variation in NDVI relates to changes in SRI ownership and document an overall positive association between SRI ownership of a company and the NDVI around the company’s establishments. However, this improvement is limited to facilities located within OECD countries. These heterogeneous findings underscore the importance of considering the global nature of MNEs when evaluating sustainability efforts.

This paper examines the benefits and costs of greenwashing in mutual funds. We first identify greenwashing funds by using large language models (LLM) to analyze fund reports along with sustainability ratings. Exploring the economic incentives behind greenwashing, we find that such funds charge higher expenses, attract greater fund flows, and face lower investor sensitivity to poor performance. However, greenwashing funds are also more likely to receive ESG-related comment letters from the SEC, which subsequently lead to net fund outflows. Furthermore, there exists heterogeneity in how institutional and retail investors respond to greenwashing.

Are cryptocurrencies mere speculative bubbles, or do their prices reflect the underlying technology? We address this question by examining flaws in the underlying technology that could lead to cryptocurrency theft and their effect on returns. Most cryptocurrencies are open-source projects hosted on GitHub with public source code, whose cybersecurity flaws are documented to be fixed later. We find that one flaw predicts a 5 basis points decrease in the coin's daily return. A portfolio that longs no-flaw coins and shorts high-flaw ones, which we term the “cybersecurity factor,” earns 30 basis points daily. Our results demonstrate that cryptocurrency prices reflect the quality of its underlying technology.

We investigate the demand for safety within the crypto ecosystem, where investors cannot frictionlessly resort to traditional asset classes. To enter and exit crypto markets, investors pay substantial fees associated with fiat deposits and withdrawals, while facing increased scrutiny from tax authorities over capital gains. Coupled with regulatory uncertainty around crypto trading, large institutions may leave significant resources “trapped” in this market. Our analysis shows that stablecoins cater to investors’ safety demand, as measured by the spread between the T-bill yield and the overnight index swap rate (“money premium”). An increase in the demand for safety leads to higher deposit volumes and lower interest rates in stablecoin lending pools than any other assets in the crypto space. During periods of market stress, stablecoin lending pools no longer respond to measures of demand for safety, whereas in normal times, these pools are consistently regarded as a safe alternative, without distinction among tokens and blockchains. Moreover, the demand for safe assets gravitates toward DeFi protocols with greater market size. Overall, the evidence suggests that investors’ demand for safety in DeFi mimics established patterns in traditional finance.

Do banks price their borrowers' exposure to climate transition risk? I find that in the E.U., firms negatively exposed to climate transition risk face higher lending rates by banks specialized in their borrowers' industry. However, I also find evidence of lower lending rates to more exposed firms after an oil supply news shock relevant for energy-intensive firms, especially during periods of high aggregate financial stress. Interpreting bank specialization as a source of heterogeneity in costs of private information acquisition, I develop a bank lending model with competitive lending, costly information acquisition, and non-Bayesian belief updating. Because of screening, specialized banks can better distinguish between borrowers' risk exposure, resulting in relatively higher lending rates to more exposed firms. However, this interest rate differential decreases in favor of more exposed borrowers when banks underreact to relevant public information. This effect is more pronounced during periods of poor borrower quality or increased financial stress. These results imply that lowering banks' cost of acquiring firm-level transition-risk exposure information is crucial to reduce green firms' financing costs, even when there is high quality public information and communication about decarbonization.

Does the private information banks generate about their corporate borrowers deepen and

change in nature over time, and if so, how? Exploiting the comprehensive Federal

Reserve’s supervisory dataset, we distinguish two dimensions to the private information

embedded in internal credit ratings: depth and direction (better or worse), which we

confirm to correlate with loan terms. Longer firm-bank relationships deepen private

information in both directions, with effects often strongly nonlinear and peaking at about

five years. Learning effects are particularly salient for smaller and leveraged firms,

smaller, leveraged, and illiquid banks, at longer firm-bank distances, and during non-

COVID times.

The Securities and Exchange Commission's investigative process remains opaque and challenging to study due to limited observability. Leveraging de-identified smartphone geolocation data, we provide new insights into the SEC's monitoring practices by tracking SEC-associated devices that visit firm headquarters. We document that SEC visits frequently occur outside formal investigations, target larger firms and those with prior enforcement actions, and cluster within industries, with substantial cross-regional monitoring. These visits are economically meaningful events, associated with significant negative stock price reactions even when no formal investigation follows. We also find that while insiders are generally less likely to sell around visits, those who do sell avoid substantial losses. Our results reveal previously unobservable aspects of SEC oversight and suggest important information flows occur outside of formal proceedings.

We find that 80% of companies that lobby Congress also lobby executive agencies. Although executive agencies are not beholden to companies for campaign contributions, the agencies are nevertheless influenced by lobbying: companies’ lobbying leads to more favorable rules, more special exemptions, more government contracts, and more favorable decisions on enforcement actions. Agencies’ bestowment of favors appears to be motivated by opportunities within the private sector: lobbying is significantly greater among agencies that have stronger revolving door relations with the private sector. Following a negative exogenous shock to agency power, the Supreme Court’s Chevron decision, firms engaged in agency lobbying experienced negative abnormal returns, underscoring the strategic value of lobbying agencies.