Knowledge that Transforms

Small firms struggle to grow beyond a few employees. We introduce monitoring devices into commuter minibuses in Kenya and randomize which minibus owners have access to the data using a novel mobile app. We find that treated vehicle owners modify the terms of the contract to induce higher effort and lower risk taking from their drivers. This reduces firm costs and increases firm profitability. There is suggestive evidence that some firms expand. These results suggest that small firms may be able to utilize monitoring technologies to overcome problems of moral hazard and enhance their profitability. (JEL D22, D24, D82, J41, L25, L92, O14)

Developing countries have increasingly decentralized power to local governments. This paper studies the implications of a central element of decentralization (polity size) using population-based discontinuities that determine local government boundaries for over 100,000 Indian villages. Over the short and long run, individuals allocated into local governments with smaller populations have better access to public goods. We provide suggestive evidence that these results are related to heightened civic engagement and stronger political incentives, but not to other mechanisms such as elite capture. (JEL D72, H41, H75, H76, O17, O18, R50)

We study a dynamic contribution game where investors seek private benefits offered in exchange for contributions, and a single, publicly minded donor values project success. We show that donor contributions serve as costly signals that encourage socially productive contributions by investors who face a coordination problem. Investors and the donor prefer different equilibria, but all benefit in expectation from the donor’s ability to dynamically signal his valuation. We explore various contexts in which our model can be applied and delve empirically into the case of Kickstarter. We calibrate our model and quantify the coordination benefits of dynamic signaling in counterfactuals. (JEL C73, D26, D82, G32, L26, M13)

We study how performance metrics affect the allocation of talent by exploiting the introduction of the first citation database in science. For technical reasons, it only covered citations from certain journals and years, creating quasi-random variation: some citations became visible, while others remained invisible. We identify the effects of citation metrics by comparing the predictiveness of visible to invisible citations. Citation metrics increased assortative matching between scientists and departments by reducing information frictions over geographic and intellectual distance. Highly cited scientists from lower-ranked departments (“hidden stars”) and from minorities benefited more. Citation metrics also affected promotions and NSF grants, suggesting Matthew effects. (JEL A14, I23, J44)

We provide evidence that classic lottery anomalies like probability weighting and loss aversion are not special phenomena of risk. They also arise (and often with equal strength) when subjects evaluate deterministic, positive monetary payments that have been disaggregated to resemble lotteries. Thus, we find, e.g., apparent probability weighting in settings without probabilities and loss aversion in settings without scope for loss. Across subjects, anomalies in these deterministic tasks strongly predict the same anomalies in lotteries. These findings suggest that much of the behavior motivating our most important behavioral theories of risk derive from complexity-driven mistakes rather than true risk preferences. (JEL C91, D44, D81, D91)

We propose a model of optimal decision making subject to a memory constraint. The constraint is a limit on the complexity of memory measured using Shannon's mutual information, as in models of rational inattention; but our theory differs from that of Sims (2003) in not assuming costless memory of past cognitive states. We show that the model implies that both forecasts and actions will exhibit idiosyncratic random variation; that average beliefs will also differ from rational-expectations beliefs, with a bias that fluctuates forever with a variance that does not fall to zero even in the long run; and that more recent news will be given disproportionate weight in forecasts. We solve the model under a variety of assumptions about the degree of persistence of the variable to be forecasted and the horizon over which it must be forecasted, and examine how the nature of forecast biases depends on these parameters. The model provides a simple explanation for a number of features of reported expectations in laboratory and field settings, notably the evidence of over-reaction in elicited forecasts documented by Afrouzi et al. (2020) and Bordalo et al. (2020a).Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

Do coordination failures constrain financial technology adoption? Exploiting the Mexican government’s rollout of 1 million debit cards to poor households from 2009 to 2012, I examine responses on both sides of the market and find important spillovers and distributional impacts. On the supply side, small retail firms adopted point-of-sale terminals to accept card payments. On the demand side, this led to a 21 percent increase in other consumers’ card adoption. The supply-side technology adoption response had positive effects on both richer consumers and small retail firms: richer consumers shifted 13 percent of their supermarket consumption to small retailers, whose sales and profits increased. (JEL E42, L25, L81, O14, O33)

We study how misperceptions of others’ tastes influence beliefs, demand, and prices in markets with observational learning. Consumers infer a good’s quality from the quantity demanded and price paid by others. When consumers exaggerate the similarity between their and others’ tastes, such “taste projection” generates discrepant quality perceptions, which are decreasing in a projector’s taste and increasing in the observed price. These biased inferences produce an excessively elastic market demand. We also analyze dynamic monopoly pricing with short-lived taste-projecting consumers. Optimal pricing follows a declining path: a high initial price inflates future buyers’ perceptions, and lower subsequent prices induce overadoption. (JEL D42, D83, D91, L15)

We propose a parsimonious mechanism for generating premature deindustrialization (PD). In the baseline model, the Baumol effect drives the hump-shaped path of the manufacturing share. Countries follow different paths due to the difference in the sector-specific adoption lags. The condition for PD under which countries differ only in technology gap implies that the cross-country productivity dispersion is the largest in agriculture. Moreover, when calibrated to match Rodrik’s (2016) findings, it is the smallest in manufacturing. In three extensions, we add the Engel effect, international trade, and catching up by late industrializers, to demonstrate the robustness of the mechanism. (JEL F11, L16, L60, O14, O33)

We document large gender disparities within a government program that entitles 46 million poor individuals to free hospital care. We show that care is not free in practice and higher costs are associated with larger disparities. Lowering care costs increases female utilization but does not reduce gender disparities because marginal beneficiaries are as likely to be male as inframarginals. Long-term exposure to local female leaders reduces disparities by addressing factors lowering female care. In the presence of gender bias, subsidizing social services may fail to address gender inequalities without actions that specifically target females. (JEL H51, I12, I13, I14, J16, O15)