European Center for Advanced Research in Economics and Statistics

Title : How do you know you won’t like it if you’ve never triedit? Preference discovery and strategic bundling

Abstract : Consumers discover their preferences through experience, yet the sequence and composition of those experiences are often designed—by platforms, algorithms, or policymakers.Adopting a “data-design” approach to preference discovery, we study how the structureof consumption data affect learning. When experiences are bundled, the co-occurrence ofgoods creates a network of correlation externalities that shapes how consumption surprisepropagates across products. Using tools from OLS estimation and spectral analysis, wecharacterize the conditions that favor learning and preference discovery and those thatsustain bias. The framework introduces a new way to think about learning from designeddata, providing a foundation for new theoretical and applied work on preference discoveryand belief manipulation. A toy empirical example from the movie industry illustrates themodel’s key mechanisms.

Title : Working time reductions and monopsonypower

Abstract : I use newly digitized micro datasets to evaluate one of the first-ever labor regulations in Belgium: a maximum 9-hour workday in coal mines in 1910. On average, hourly wages and employment increased. However, these effects were sizedependent: hourly wages decreased in small firms, while all employment gainswere concentrated in large firms. I argue that these results are inconsistent withthree common assumptions: (i) workers choose hours, (ii) competitive wages, and(iii) a Cobb-Douglas technology in jobs and hours. I rationalize these results in adirected search model where firms with heterogeneous TFP post vacancies, hours,and wages while internalizing workers’ leisure preferences. Monopsony power leadsto long hours. When hours are capped, a firm either increases wages to substitutelost hours with new hires, or cuts wages to restore markdowns. Which mechanismdominates depends on firm size, even holding monopsony power constant. Welfareanalysis with sufficient statistics suggests that the 1910 reform improved workers’welfare.

Title : Transparency and competition for influence

Abstract : We study the impact of mandatory disclosure of contributions paid by interestedthird parties to decision makers such as doctors, politicians, or financial advisors. Whiletransparency is commonly viewed as a means of reducing potential conflicts of interest, our analysis reveals less benign outcomes when multiple third parties attempt toinfluence decision makers in opposing directions. We argue specifically that in thiscase transparency enables competing third parties to establish separate spheres of influence, where their ascendancy is not attenuated by the opposing efforts by rivals.Consequently, decision makers’ choices become more polarized. We apply this theoryto the market for anticoagulants, using data on prescriptions and payments made bypharmaceutical companies to doctors in the United States before and after the Physician Sunshine Act of 2010 that mandated payment disclosure. The empirical analysissupports our theory

Title : Export Subsidies as Industrial Policy:the Case of the 19th Century Sugar Industry

Abstract : Can export subsidies spur technological upgrading? What are their welfare implications?We build a model in which firms select into exporting and can pay to upgrade technology.An export subsidy can induce upgrading by increasing exporters’ scale, making upgrading worthwhile. Despite this, we show that an export subsidy is not generally welfareimproving. Absent spillovers – whereby social benefits from upgrading exceed private –adoption is efficient. With spillovers, welfare effect comprises: (i) a positive term frommoving the adopter share toward the social optimum, and (ii) an allocative term whosesign depends on whether the subsidy alleviates or aggravates pre-existing misallocation.We test the model using a natural experiment in the nineteenth-century Austro-Hungarianbeet-sugar industry: an 1864 consumption-tax reform unintentionally created a 10% exportsubsidy as the industry transitioned to new sugar-extraction technology. Newly digitizedfactory-level data show that the subsidy reallocated production toward high-export-accessregions, whose factories adopted frontier technology earlier. Complementary evidence fromFrance documents positive local spillovers. Calibrating the model to Austro-Hungariandata, we find the subsidy increased adoption but exceeded the optimum, reducing welfare.In turn, by worsening misallocation caused by pre-existing protective tariff any positiveexport subsidy would have been counterproductive.

Title : Hegemonic Competition with Carrots and Sticks

Abstract : Hegemonic powers use economic tools to influence the geopolitical alignmentof third countries. We develop a bargaining model in which two competing hegemons use direct payments (carrots) and economic threats (sticks) to influence thirdcountries. Guided by the model, we measure carrots and sticks using historicaldata from the Cold War. We use our measures to empirically estimate the effectsof carrots and sticks on geopolitical alignment. These tools create political alignment, but are expensive. We combine the model with the empirical estimates tocompute the geopolitical benefit-cost ratio of foreign aid and evaluate the c

Title : Strategic Transportation Investment andCoordinative Policies:Evidence from the U.S. Highway Network

Abstract : Transportation networks are often built by multiple jurisdictions whose investment decisions may not fully account for cross-border benefits. This paper builds aframework to evaluate the welfare implications of limited regional cooperation andcentral subsidies in transportation investments, accounting for goods trade, commuting, and governments’ budget constraints. In this framework, regional governmentschoose investments in their segments of the network to serve constituents’ interests,taking as given the investment choices of other regions and cost-sharing subsidiesfrom a central government. Through the lens of the model, decentralized investmentin the U.S. highway network leads to 15% underinvestment relative to the nationaloptimum and generates a welfare loss equivalent to 30% of current highway spending. While raising the federal subsidy rate increases total spending, it fails to achievea cost-effective allocation of investment across the network.