Knowledge that Transforms

We use surveys, laboratory experiments and administrative data to study how heterogeneity in the perceived immorality of work and in workers' aversion to acting immorally impact labor market outcomes. Immoral work is associated with higher wages, both in administrative data and in causal experimental evidence. Individuals more willing to engage in immoral conduct find employment in firms and industries perceived as immoral less aversive and have higher employment rates in immoral work in the laboratory. These phenomena appear to be driven by impure social motives, reflecting a desire not to be involved with immoral work, rather than by consequentialist concerns.

We estimate the impact of piped water and sewers on property values in late 19th century Chicago. The cost of sewer construction depends sensitively on imperceptible variation in elevation, and such variation delays water and sewer service to part of the city. This delay provides quasi-random variation for causal estimates. We extrapolate ate estimates from our natural experiment to the area treated with water and sewer service during 1874-1880 using a new estimator. Water and sewer access increases property values by a factor of about 2.8. This suggests that benefits are large relative to: the value of the value of averted mortality, many other infrastructure projects, and construction costs.

We provide the first systematic account of summer declines in women's labor market activity. From May to July, US women's employment-to-population ratio declines 1.1pp, whereas men's rises; women's hours worked fall 9.8%, twice the decline among men. School closures for summer break provide a unifying explanation. The summer drop aligns with cross-state differences in school closure timing, is concentrated among mothers with young school-aged children, and coincides with increased time spent on childcare. Gender differences both within and across jobs explain gender gaps in summer exits from employment. Women's summer work interruptions contribute to gender gaps in pay.

Between 1967 and 1974, a bilateral treaty increased circular labor migration from Malawi to South Africa by 200%, bringing over 53 million USD in earnings into origin communities. A deadly migrant worker plane crash in 1974 ended these flows and led to migrant repatriation. We study how this shock affected local labor markets. In regions receiving more migrant capital after the crash, workers, particularly women, shifted from farming into non-farm work over thirty years. Investments in non-farm physical and human capital contribute to these sectoral changes. This natural experiment shows that temporary capital inflows can permanently reshape rural labor markets.

We develop a theory of trust in lending that distinguishes between reputation and trust. Banks emerge as more trusted lenders than non-banks. We show that trust severs the link between performance and the cost and availability of financing for lenders, but trust can be lost and is difficult to regain. Banks survive an erosion of trust better than non-banks. Banks' trust advantage arises from the lower cost of funding due to insured deposits and an endogenous belief revision channel that complements the effect of the funding cost advantage. The results have novel policy relevance for deposit insurance scope.

Competition is widely used to enhance effort and performance. However, in many domains, like education, competition could backfire, as performance is not solely reliant on individual efforts but also on collaboration endeavors among peers. Utilizing university administrative data, we examine how competition changes peer effects and peer interactions. Exploiting randomly assigned roommates, we first demonstrate that high-ability roommates have detrimental effects on the academic performance of high-ability students. More importantly, such negative peer effects significantly increase along various dimensions of competition intensity within dorm rooms. Follow-up survey findings reveal that competition hinders mutual assistance and fosters unfriendly behaviors among roommates.

We provide novel evidence on the causal impact of a child's health shock on parents' labor market outcomes. Using high-quality Finnish and Norwegian administrative data, we construct counterfactuals for treated households with families who experience the same shock in later years. We find a sharp break in mothers' earnings trajectories after the event, while we do not find significant effects for fathers. Our findings do not align with the hypothesis of household specialization explaining these adjustments. Instead, the evidence suggests that these changes are driven by increased caregiving demands, with mothers bearing the primary burden. We also document a substantial impact on parents' mental well-being.

We disprove the exchange rate macroeconomic disconnect puzzle by showing that macroeconomic news can explain most variation in exchange rates at monthly and quarterly frequencies, accounting for up to 91 percent of the quarterly exchange rate variation during US recessions and 65 percent over all periods. The main driver of the reconnect is exchange rates responding to past news—a result inconsistent with the theory of uncovered interest rate parity under full information rational expectations (UIP-FIRE). We discuss theoretical models that can explain this surprising result, including models featuring currency risk premia, regulatory or institutional frictions, or deviation from FIRE.

Economists have long suspected that firm-to-firm relationships might lower the responsiveness of prices to shocks due to the use of fixed-price contracts. Using transaction-level U.S. import data, I show that the pass-through of exchange rate shocks in fact rises as a relationship ages. Based on novel stylized facts about a relationship's life cycle, I develop a model of relationship dynamics in which a buyerseller pair accumulates relationship capital to lower production costs under limited commitment. The structurally estimated model generates countercyclical mark-ups and countercyclical pass-through of shocks through variation in the economy's rate of relationship creation, which falls in recessions.

We propose improved standard errors and an asymptotic theory for two-way clustered panels. Our theory allow for arbitrary serial dependence in the common time effects, which is excluded by existing two-way methods. Our asymptotic distribution theory is the first which allows for this level of inter-dependence. Under weak conditions, we demonstrate that OLS is asymptotically normal, our proposed variance estimator is consistent, and t-ratios are asymptotically standard normal. The results extend to two-way fixed-effect models; we argue that two-way clustering is still necessary even if two-way fixed effects are included. Simulation and empirical illustration are provided.