Working Papers

This study constructs a novel measure that aims to capture face-to-face private communications between firm managers and sell-side analysts by mapping detailed, large-volume taxi trip records from New York City to the GPS coordinates of companies and brokerages. Consistent with earnings releases prompting needs for private communications, we observe that daily taxi ride volumes between companies and brokerages increase significantly around earnings announcement dates (EAD) and reach their peak on EAD. After controlling for an extensive set of fixed effects (firm, analyst, year, and firm-broker) and other potential confounding factors, we find that increases in ride volumes around EAD are negatively associated with analysts' earnings forecast errors in periods after EAD and positively associated with the profitability of recommendations issued after EAD (but these effects dissipate over longer horizons). Taken together, our results suggest that analysts may obtain a private source of information orthogonal to their pre-existing information from these in-person meetings, which may help them better understand the implications of current earnings signals for future earnings.

Prior research finds that online social media usage may lower self-control and encourage indulgent behavior in laboratory subjects. We find that corporate CEOs show similar tendencies: CEOs with online social media presence are more likely to succumb to lower self-control and abuse their information advantage to profit from unethical insider trades. Specifically, CEOs' social media presence strongly predicts their insider trading activity in terms of incidence, intensity (amount and frequency), and profitability. We further find that the effect is driven by insider buys (not by sells) and is more pronounced for opportunistic buys which tend to contain more material non-public information.

This paper examines how labor frictions affect investment rate and new firm entry. Using matched employee-employer data from LinkedIn, I first show that increases in the enforceability of non-compete agreements lead to widespread declines in employee departures across seniority levels, driven by workers in knowledge-intensive occupations. Investment rates at existing firms increase, especially for firms that employ more skilled workers. This comes at the expense of new firm entry, which declines substantially in knowledge-intensive sectors. The results suggest that labor frictions play an important role in investment decisions, and that NCs may factor into slowing business dynamism.

We analyze worker reactions to firms' credit deterioration. Using weekly anonymized networking activity on LinkedIn, we show workers initiate more connections immediately following a negative credit event, even at firms far from bankruptcy. Our results suggest that workers are driven by concerns about both unemployment and future prospects at their firm. Heightened networking activity is associated with contemporaneous and future departures, especially at highly-rated firms. Other negative events like missed earnings and equity sell recommendations do not trigger similar reactions. Overall, our results indicate that the latent build-up of connections triggered by credit deterioration represents a source of fragility for firms.

We provide the first analysis of the risk exposure and risk-adjusted performance of impact investing funds, private market funds with dual financial and social goals. We introduce a dataset of impact fund cash flows and exploit distortions in VC performance measures to characterize risk profiles. Impact funds have a lower market than comparable private market strategies. Accounting for , impact funds underperform the public market, though not more so than comparable strategies. Adding a public sustainability factor to our pricing model helps explain impact returns, though the correlation of fund cash flows with this factor is not necessarily positive.

Does investors' political ideology shape international capital allocation? We provide evidence from two settings-syndicated corporate loans and equity mutual funds-to show ideological alignment with foreign governments affects the cross-border capital allocation by U.S. institutional investors. Ideological alignment on both economic and social issues plays a role. Our empirical strategy ensures direct economic effects of foreign elections or government ties between countries are not driving the result. Ideological distance between countries also explains variation in bilateral investment. Combined, our findings imply ideological alignment is an important, omitted factor in models of international capital allocation.

While there is no apparent reason for loan spreads to cluster at certain numbers, we find that around 70% of loans have round-yard spreads (i.e., multiples of 25 basis points). We hypothesize that dominant banks implicitly collude by using the round-yards as focal pricing points when negotiating with their borrowers. The tacit collusion leads to higher spreads and total costs of the round-yard-priced loans than non-round-yard-priced loans. Consistent with our tacit collusion hypothesis, dominant banks round up rather than round down loan spreads to the multiples of yards. Moreover, round-yard pricing is more prevalent among lower-quality and non-repeat borrowers. Overall, we provide the first evidence that dominant banks use round-yard pricing as an effective tool for tacit collusion in the loan market.

The first twenty years of the 21st century were a period of transformation and change in the development models of the financial market. One of the strongest in the history world financial crises of 2007-2008 ended the post-deregulation model. Transition to the new financial market model turned out to be largely unpredictable, complex, and spontaneous, unlike the previous periods, without the purposeful participation of state regulators. An objective but random reason for this course of events was the COVID-19 pandemic. Pandemic has distorted the effect of the loose monetary policy, which caused building the grounds for a new financial market model. The post-pandemic model of the financial market is still in the early stages of formation, and it is too early to talk about all its properties and elements. However, as seen from current events and processes, the essential factor of the new financial market model formation is a gamification of investors' behavior. The author believes this behavioral model requires much more attention of researches than that in nowadays scientific literature.

Hedge funds can subsequently amend their originally reported 13F quarterly holdings using restatements. We conduct the first systematic analysis of such filings, which are as common as confidential filings (used by funds to delay holdings disclosures), but affect four times as many stocks. Restated holdings are associated with significant abnormal returns, suggesting that some original holdings are strategically misreported to hide funds trading intentions. We construct a return gap measure to gauge the value added by such restatements and find that it predicts future fund performance. Finally, commonly used databases such as Thomson Reuters do not fully adjust for restatements.

We exploit variation in the ancestries of U.S. equity mutual fund managers and show that ancestry affects portfolio decisions. Controlling for fund firm location, we find that funds overweight stocks from their managers' ancestral home countries in their non-U.S. portfolio by 132 bps or 20.34% compared with their peers. Similarly, funds overweight industries that are comparatively large in their manager's ancestral home countries. The documented ancestral biases are pervasive across fund styles and across different manager ancestries. The effect is more pronounced for funds that are less resource-constrained and for managers whose connection to their ancestral home country is more recent. Stocks linked to managers' ancestry do not outperform stocks in the same countries and industries but held by managers of other ancestry, confirming that ancestry-linked investments are not informed.

Forecasters predicting how people change their behavior in response to a treatment or intervention often consider a set of alternatives. In contrast, those who are treated are typically exposed to only one of the treatment alternatives. For example, managers selecting a wage schedule consider a set of alternative wages while employees are hired at a given rate. We show that forecasts made in joint-prediction mode - which considers a set of alternatives -generate predictions that expect substantially larger behavioral responses than those made in separate-prediction mode - which considers the response to only one treatment realization in isolation. Results show the latter to be more accurate in matching people's actual responses to interventions and treatment changes. We present applications to managerial decision-making and forecasting of scientific results.

This paper uses machine learning to infer nonprofessional social media investment analysts' (SMAs) beliefs from their opinions on individual stocks. SMAs' average beliefs predict future abnormal returns and earnings surprises. However, there exists substantial heterogeneity in SMAs' ability to form beliefs that yield investment value. Some 13% high-skilled SMAs form beliefs that yield a sizeable one-week three-factor alpha of 61 bps, while the remaining 87% low-skilled SMAs generate only 6 bps. Firm and industry specializations are the most distinctive characteristics of high-skilled SMAs. When forming beliefs, SMAs extrapolate from past returns and herd on the consensus view of their peers. However, these seemingly behavioral biases do not result in systematically wrong beliefs.