There are concerns that managers repurchase shares myopically to benefit themselves at the costs of shareholders and society. Due to these concerns, the SEC adopted its 2023 Share Repurchase Disclosure Modernization Rules (the ``modernized rule"). The modernized rule mandated a narrative disclosure of repurchase intent and more detailed quantitative repurchase disclosure. While the modernized rule became effective in October 2023, it was vacated in December 2023. Due to different quarter start dates, the law was effective for some firms for nearly two months; other firms were never treated. Using this quasi-exogenous variation, monthly repurchase data and a staggered DiD design, my main tests find that the modernized rule reduces share repurchases amounts, primarily by reducing the number of firms repurchasing. In cross-sectional tests, efficient (value-maximizing) repurchases decline either the same as or more than myopic (value-decreasing) repurchases. Results using event study returns and repurchase prices corroborate these effects. Overall, the findings are consistent with increased repurchase disclosure having real effects on repurchases. It reduces repurchases, but it does not have its intended effect of reducing only myopic repurchases. Efficient repurchases decline at least as much as myopic repurchases.

Policymakers increasingly use trade instruments to address national security concerns. This paper studies optimal policy for dual-use goods, items with both military and civilian applications. We begin by documenting that regulation and trade flows of dual-use goods respond to changes in the security environment over time. To put structure on the national security externality, we introduce military procurement into a trade network model and add a military contest to the national welfare function. In a simple two-country case, optimal export taxes depend on a trade-off between the good's military centrality and its distortion centrality. Military centrality is a network-adjusted sales share to the foreign military; distortion centrality reflects taxation misallocation in the domestic economy from roundabout imports. Using U.S. defense procurement data, we develop a measure of military use across goods between zero and one based on our optimal tax formulas. Our measure is associated with policy targeting and trade responses around conflicts and allows us to evaluate the U.S. security restrictions and sanctions against Russia. To quantify the macroeconomic magnitude of the consumption-security trade-off, we calibrate our model to a potential U.S.-China conflict. Our revealed preference estimate of the conflict prize amounts to 250% U.S. GDP.

I show that financial markets react asymmetrically to bad news about firm performance depending on whether it’s delivered by male or female CEOs. This asymmetry manifests in analysts’ forecasts, stock returns on earnings announcement days, and even in the tone that analysts adopt in earnings conference calls. I argue these patterns have a common origin in frictions in the communication between firm leadership and financial markets that relates to the CEO's gender. To make this case, I first document that analysts' beliefs about firm performance systematically under-react to bad news from male-led companies relative to the rational expectations benchmark, whereas they adjust their expectations rationally to similar bad news from female-led companies. Next, I show that investors also display this biased reaction to news, with stock returns reacting less negatively to negative surprises from male-led companies relative to their female-led peers on earnings announcement days. I then shed light on the underlying mechanism by constructing a text-based measure of disagreement from earnings conference calls. After negative surprises, analysts express less disagreement with the narrative conveyed by male-led firms relative to their female-led peers. This effect is entirely concentrated amongst male analysts, who represent, on average, more than 85% of the participants in these calls.

Open source software (OSS) is a form of public knowledge widely provided and relied on by the private sector. To study the effects of growing government involvement in this critical public good, I build a new empirical model where high-tech firms choose software inputs and developer labor in competitive equilibrium. For estimation, I create a new dataset of OSS and in-house investment for the global web development industry, where software choices are directly observable. I simulate counterfactuals to assess the global impact of China tightening its recent internet restrictions on cross-border OSS collaboration or increasing its financial support for domestic OSS. I find that stricter restrictions do little to boost domestic OSS investment. Instead, lost spillovers raise web development costs in China by $2 per dollar of disincentive and $7 globally. Heightened subsidies prove more effective at increasing domestic investment and cut global costs by $11 per dollar of subsidy—tripling if the US responds in kind.

Governments often restrict international migrants' job options. This paper shows that these restrictions can hurt not only migrants but also the existing residents whom they aim to protect. We study New Zealand’s ‘Essential Skills’ work visa, which was New Zealand's main work visa between 2008 and 2022. Essential Skills migrants could only work for firms which could not find New Zealanders. Loosening restrictions for a single individual has no impact on their wages: migrants who win an unrestricted resident visa through a lottery switch jobs more frequently, but receive no gain in wages. However, when the Essential Skills job restrictions were loosened for all migrants in an occupation, both job-switching and wages typically grew. These results are consistent with a wage-posting model in which each firm pays migrants and residents equally; in such a model, the wage received by each worker will not depend directly on her own outside option but rather on the distribution of outside options among her colleagues. We estimate a wage-posting model, and compare equilibrium wages under the Essential Skills job restrictions to a counterfactual simulation in which migrants’ job options are unrestricted. The restrictions decreased migrants' average wage by 8%. Although most residents were unaffected by the restrictions, 2.1% had their wage decreased by more than 2%. The restrictions increased profits, especially in firms which employed many migrants. The restrictions decreased annual welfare by $292m NZD — 30% of migrants' earnings — largely because migrants could not move to firms which they preferred for non-pecuniary reasons.

This paper develops a theory of how tradeoffs govern comparison complexity, and how this complexity generates systematic mistakes in choice. In our model, options are easier to compare when they involve less pronounced tradeoffs, in particular when they are 1) more similar feature-by-feature and 2) closer to dominance. These two postulates yield tractable measures of comparison complexity in the domains of multiattribute, lottery, and intertemporal choice. We then show how behavioral regularities in choice and valuation, such as context effects, preference reversals, and apparent probability weighting and hyperbolic discounting in valuations, can be understood as responses to comparison complexity. We test our model experimentally by varying the strength and nature of tradeoffs. First, we show that our complexity measures predict choice errors, choice inconsistency, and cognitive uncertainty in binary choice data across all three domains. Second, we document that manipulations of comparison complexity can reverse classic behavioral regularities, in line with the predictions of the theory.

This paper introduces Incrementality Representation Learning (IRL), a novel multitask representation learning framework that predicts heterogeneous causal effects of marketing interventions. By leveraging past experiments, IRL efficiently designs and targets personalized interventions, eliminating the need for extensive testing of numerous potential interventions. To ensure generalization to untested interventions and customers, the IRL model extracts low-dimensional representations of intervention features and customer covariates that are predictive of treatment effects and are generalizable across previously tested interventions. Unlike traditional multi-task learning methods that build separate models for each intervention, IRL uses a unified prediction model across past experiments to enhance generalizability. We empirically validate our framework in the context of promotional campaigns for consumer packaged goods. By synergizing data from 274 previously conducted experiments, our IRL model not only improves the targeting accuracy of tested interventions but also significantly outperforms existing methods in targeting untested interventions and customer segments, overcoming the generalization challenge in high-dimensional decision spaces and the cold-start problem associated with designing new interventions. Furthermore, we develop a decision framework to identify key design features and customer segments for tailoring interventions. Using our model interpretation tool, we demonstrate how companies can customize promotions to enhance profitability across different customer segments.

This study examines a critical decision faced by new ventures: whether to try to sell their existing product or invest more in product development first. I propose that adequate product development is crucial for effectively learning about product-market fit. I introduce the concept of a mediocrity trap, where ventures trying to sell less developed products tend to persist despite negative market signals, leading to delayed exit. Using a theoretical model, I argue for the distinct nature of market opportunity and product quality, exploring how their interaction affects the strategic timing of market entry and learning. An empirical analysis of 1,445 enterprise software companies supports these ideas, showing that firms trying to sell less developed products persist longer in the face of negative market signals and, ultimately, fail more often. This research enhances our understanding of learning about product-market fit, scaling new ventures, and balancing product development with sales to avoid the pitfalls of premature market entry. This study has significant managerial implications and challenges prevailing managerial approaches to experiment and “fail fast” in a resource-efficient manner.

In entering foreign markets, artists-as-entrepreneurs face a choice. Drawing from nearly 500,000 MP3 files and 15 million releases (e.g., albums, LPs), I analyze the aesthetic and collaborative choices made by Global South musicians seeking to develop their audiences beyond borders. They can produce a novel and distinctive product, improving their odds of long-term success conditional on entry. Alternately, they can choose to closely hew to existing music in the genre, increasing their likelihood of breaking into the market but undermining their appeal: higher conformity significantly raises the likelihood of exit. The paper also documents strikingly different patterns of collaboration with creative, though not technical, partners; these patterns persist even if an artist retreats from foreign markets.

The rapid growth of on-demand delivery services, particularly in the food and grocery sectors, has driven the expansion of hyperlocal fulfillment centers (FCs). This paper uses data from an on-demand grocery delivery platform in Latin America to assess how customer location influences purchasing behavior and operational efficiency. We examine how service quality varies with customer proximity to FCs and how purchasing behaviors are affected by the distance from these centers. Our analyses show that customers farther from their assigned FCs purchase less frequently and have higher service abandonment rates. We also observe that these location-dependent behaviors lead to a higher proportion of first orders than repeat orders in areas farther from the FCs, reflecting a decline in service usage due to lower service quality. Our findings highlight the critical role of customer location in determining service quality and its operational implications for retailers. If unaddressed, these dynamics could lead to a “coverage area shrinking effect,'” where geographic order density concentrates closer to FCs, which can reduce the effective coverage area size over time. We discuss the broader implications of these findings for other settings where customer location influences service quality, a scenario that is increasingly prevalent.