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 β than comparable private market strategies. Accounting for β, impact funds underperform the public market, though not necessarily more so than comparable strategies. We consider alternative pricing models, accounting for sustainability and emerging markets risk. We show investors’ wealth portfolios and taste change the perceived financial merit of impact investing.

We study financial event studies—empirical settings where asset market returns assess the impact of information or policy changes. We show that abnormal return estimators are sensitive to factor model misspecification, making them inconsistent estimators of causal effects. We demonstrate that staggered event timing can mitigate this issue in short-horizon studies but not in long-horizon analyses, where misspecification bias accumulates over time. Synthetic control methods emerge as a solution that avoids these issues by directly modeling counterfactual security paths without requiring correct specification of the factor structure. Our empirical applications to political connections (Acemoglu et al., 2016) and S&P 500 index inclusion (Greenwood & Sammon, 2025) demonstrate the practical implications of these methodological insights, particularly in settings that lack both short horizons and random event timing.

We study how the lender relationship with venture capital (VC) investors shapes the life cycle of venture debt. We develop a model that highlights two competing mechanisms of the relationship, an information channel capturing certification benefits and a market power channel reflecting rent extraction. Using a comprehensive dataset on global venture debt, we test these channels across the life cycle of venture debt. At entry, the relationship mitigates asymmetric information and increases the likelihood of obtaining venture debt. In the investment stage, lenders with a relationship reduce hard restrictions while charging higher spreads. Post deal, relationship-backed startups are more likely to secure subsequent VC and successful exits by reallocating innovation toward commercially salient and safer projects. Our findings highlight that VC-lender relationships reduce information frictions and enable rent extraction while facilitating value creation in high-growth ventures.

This paper examines the biases in managerial expectations as a function of firms' financial conditions. Using microdata from the Duke CFO Survey and managerial guidance, I document pervasive overreaction to news across forecasts of firm profitability, operations, and financing decisions, and even the aggregate market returns. Forecast errors are predictable from past realizations, consistent with extrapolative belief formation. I demonstrate that these biases are context-dependent: overreaction is stronger among financially constrained firms. Using exogenous shocks to earnings and financial conditions, I show that constrained managers overreact more strongly to negative earnings news, while managers with improved financial conditions exhibit weaker overreaction to positive earnings news. Evidence from moderate earthquakes as exogenous emotional shocks supports an affect-based mechanism: negative emotions distort managerial expectations, and these expectations, in turn, influence managers’ future investment decisions. Overall, the results uncover novel affect-driven channels through which managerial expectations amplify distortions in firm investment.

This paper investigates whether an acquirer's public visibility influences antitrust enforcement in mergers and the subsequent real effects. We compile a novel dataset connecting outcomes of regulatory inspections conducted by the Federal Trade Commission (FTC) and the Department of Justice (DOJ) with news coverage data of merger parties. Using within-industry and within-firm variations in visibility under a common political environment, we show that a 10 percent increase in the acquirer's share of industry news coverage of the acquirer raises the likelihood of being flagged by 1.5 to 3 percent. We establish causality using geographical proximity to media branches as an exogenous source of variation in news coverage. We then document that both under-enforcement and over-enforcement generate inefficiencies. First, overlooked mergers enable firms to accumulate profits and market share, allowing them to pursue riskier investments despite weak growth opportunities. Second, firms that were flagged but ultimately approved expand market power and undertake inefficient expansions, even under close oversight. These findings underscore how public visibility shapes enforcement discretion and creates persistent inefficiencies in antitrust policy.

We document a relationship between memory-based models of beliefs and a general class of kernel methods from the statistics and machine learning literature. Motivated by this relationship, we propose a new form of memory-based beliefs which aligns more closely with the state of the art in the machine learning literature. We explore this approach empirically by introducing a measure of “narrative memory”– similarity between states of the world based on similarity in narrative representations of those states. Using textual embeddings extracted from conference call transcripts, we show that our estimates of memory-based beliefs explain variation in errors in long-term growth forecasts of IBES analysts. We conclude by discussing implications of this relationship for the literature on memory-based models of beliefs.