Motivated by recent concerns surrounding the use of third-party pricing algorithms by competing firms, we study repeated Bertrand competition where market demand or the cost of serving the market is observed by an intermediary (or "algorithm") that selectively discloses demand or cost information to maximize firms' collusive profit. We show that an upper censorship disclosure policy is optimal, which leads to price rigidity and supra-monopoly prices in some states. Improving the algorithm's accuracy reduces expected consumer surplus whenever it does so under monopoly pricing. When the state is positively correlated over time, the algorithm discloses more information when recent demand was lower or costs were higher. The analysis extends to a generalized model that accommodates product differentiation and capacity constraints. We relate our findings to recent antitrust cases.
This article shows that a CEO's trust enhances innovation within a firm, providing a novel micro-foundation for the well-known trust-growth relationship. I build a new matched CEO-firm-patent data set covering 5,753 CEOs in 3,598 U.S. public firms and 700,000 patents during 2000-2011. I exploit variation in generalized trust across CEO ethnic origins, inferred from their last names using deanonymized historical censuses. Following CEO turnovers, a one standard deviation increase in a CEO's generalized trust is associated with 6% more future patents and 4%-6% higher average patent quality, driven entirely by higher-quality patents. Text analysis of employee reviews shows that the CEO's trust enhances a firm's trust culture. These results are consistent with insights from qualitative interviews suggesting that the CEO's trust and the firm's trust culture encourage researchers to undertake high-risk explorative research and development. In addition, changes in the CEO's bilateral trust toward inventors in different countries have comparable effects on inventors' patenting, controlling for CEO and other fixed effects.
Empirical studies find that the pass-through of input cost changes to prices is incomplete: a 10% increase in costs causes downstream prices to rise less than 10%, even at long horizons. Using microdata from gas stations, food products, and manufacturing industries, I find that incomplete pass-through in percentages often disguises complete pass-through in levels: a $1/unit increase in input costs leads to $1/unit higher downstream prices. Pass-through appears incomplete in percentages due to a gap between prices and costs. Complete pass-through in levels contrasts with workhorse macroeconomic models that feature homothetic industry demand systems. I identify an alternative class of demand systems that yields pass-through in levels and highlight four implications. First, measuring pass-through in percentages can lead to spurious evidence of asymmetry and size dependence. Second, pass-through in levels leads to systematic fluctuations in relative price and markup dispersion that are not associated with changes in allocative efficiency. Third, pass-through in levels can explain dynamics of industry gross margins, operating profits, and entry in the data that are at odds with workhorse models. Finally, incorporating pass-through in levels into an input-output model of the U.S. economy better matches the volatility of consumer price inflation and the response of inflation to identified shocks.
Does information about the long-run financial costs of reduced labor supply increase mothers' working hours? We document descriptively that long-term financial factors are not top of mind when mothers decide on their employment level. Moreover, a substantial share of women holds overly optimistic expectations about pension receipt and wage growth under part-time work. In a large-scale field experiment in Switzerland, we randomly assign mothers working part-time as teachers to receive objective information about the long-run costs of reduced labor supply. The treatment increases both demand for financial information and future labor supply plans, in particular among women who underestimate the costs of part-time work. Leveraging linked employer administrative data one year post-intervention, we find that this group of mothers increases working hours by 7%. These findings underscore that policies reducing information frictions in labor supply decisions may help address remaining gender gaps in the labor market.
We study the effect of conflict-induced displacement on human capital and occupational shifts, focusing on the Mozambican civil war (1977-1992), during which millions of civilians were forced to flee to the countryside, cities, and neighboring countries. Reconstructing the wartime mobility histories of the surviving population, we examine the consequences of multiple displacement trajectories in a unified framework. First, we characterize the education and sectoral employment of the universe of (non-)displaced. Second, we exploit differences in relocation trajectories among extended kin members during their schooling years. Displacement is associated with significant gains in education. Third, using a movers design, we show that minors displaced earlier to better districts experienced an increase in educational attainment. Focusing on moves during the intensification of the war and when comparing members of the same household, regional childhood exposure effects remain strong, whereas spatial sorting becomes negligible. Fourth, we jointly estimate place-based, spatial sorting, and uprootedness effects, showing that all forces are at play. Fifth, a small survey in Mozambique's largest northern city reveals long-term effects: internally displaced people report higher education than their siblings who stayed behind but lower social capital and worse mental health relative to locals. Our findings demonstrate that displacement shocks can foster human capital accumulation, even in very low-income settings, albeit at the cost of enduring social and psychological traumas.
