Working Papers

We explore the welfare implications of data-tracking technologies that enable firms to collect consumer data and use it for price discrimination. The model we develop centers around two features: competition between firms and consumers’ level of…

We study social learning and information transmission in a sender-receiver game wherein senders may be attacked (“cancelled”) for challenging the status-quo beliefs. We find that cancellations (and self-censorship) don’t arise unless there is a…

We design and implement an adaptive experiment (a “contextual bandit”) to learn a targeted treatment assignment policy, where the goal is to use a participant’s survey responses to determine which charity to expose them to in a donation…

Two leading hypotheses for why individuals unintentionally share misinformation are that 1) they are unable to recognize that a post contains misinformation, and 2) they make impulsive, emotional sharing decisions without thinking about whether a…

Our efforts here represent an exploratory attempt to make some progress on developing a fuller understanding of gender and organizational culture. In doing so, we report on a wide-ranging survey we administered to a sample of full-time workers (…

Before the era of large central bank balance sheets, banks relied on incoming payments to fund outgoing payments in order to conserve scarce liquidity. Even in the era of large central bank balance sheets, rather than funding payments with…

Does going public affect the amount and type of corporate tax planning? Using a panel of U.S. corporate tax return data from 1994 to 2018, we show that IPO completion is associated with the implementation of multinational income shifting…

Learning optimal policies from historical data enables the gains from personalization to be realized in a wide variety of applications. The growing policy learning literature focuses on a setting where the treatment assignment policy does not…

Online platforms often face challenges being both fair (i.e., non-discriminatory) and efficient (i.e., maximizing revenue). Using computer vision algorithms and observational data from a microlending marketplace, we find that choices made by…

Multi-armed bandit (MAB) algorithms are efficient approaches to reduce the opportunity cost of online experimentation and are used by companies to find the best product from periodically refreshed product catalogs. However, these algorithms face…

Decisions take time, and the time taken to reach a decision is likely to be informative about the cost of more precise judgments. We formalize this insight using a dynamic model of optimal evidence accumulation. We provide conditions under which…

Labor economists regularly analyze employment data by fitting predictive models to small, carefully constructed longitudinal survey datasets. Although modern machine learning methods offer promise for such problems, these survey datasets are too…

Under what conditions can members of poor disenfranchised communities survive and even foster entrepreneurship in environments where violence is cheap? How do such conditions alter ethnic identities and political institutions? In this paper, we…

Modern digital advertising platforms allow ads to be targeted in a variety of ways, and generally aim to match the ad being shown with either the user or the content being shown. In this study, we examine the effect of matching in emotional…

How does a failure in a retailer’s mobile app impact shoppers’ purchases in its online (website and app) and offline (brick-and-mortar store) channels? Our main hypothesis is that an app failure has a negative effect on offline purchases and no…

For individuals who take on the responsibility of caring for another person due to illness, disability, or declining abilities, it can often be challenging, lonely, costly, and exhausting. As the U.S. continues to address the impact of the…

We revisit the classic result on the (non-)existence of pure-strategy Nash equilibria in the Generalized First-Price Auction for sponsored search advertising and show that the conclusion may be reversed when ads are ranked based on the product of…

We present a data-driven algorithm that advertisers can use to automate their digital ad-campaigns at online publishers. The algorithm enables the advertiser to search across available target audiences and ad-media to find the best possible…

“Buy Now Pay Later” (BNPL) is a largely unregulated FinTech innovation that provides consumers with easy access to credit for specific retail purchases. The BNPL market is projected to reach $1 trillion by 2025, but we know little about the…

This study finds that voluntary non-earnings disclosures substitute for redacted proprietary contract information. When firms redact contract information, they provide more voluntary disclosures and have higher information uncertainty and…