We show uniqueness of an unbiased estimator for parameters in a spatial autoregressive model with one exclusion restriction.

For British adult smokers in Chile, we use nonparametric regressions to define an exposure to socialist welfare policy shocks, estimating a negative effect, and suggest solvency regulations.

Although contemporary historiography suggests that bank privatization decreases bank lending, our empirical results suggest the effects of import tariffs offset this effect, at least on the extensive margin.

We study general dynamic games with rational actions, trades, and increasing returns. In such games, we show that there generically might exist a stochastic equilibrium.

We simulate counterfactual import tariffs in Sochi using neural networks to explain observed decreases in bank lending.

Little is known about substance abuse. We use shocks to social distancing compliance rates to estimate two important policy parameters using triple differences: the elasticity of demand for microchips and the rate of primate city growth.

We study the optimal design of rules in a bayesian model when there are continuous rewards and uncertainty about whether players can commit to follow the rule ex ante. We show that dynamic rules are preferable.

We study the optimal design of rules in a bayesian model when there are continuous rewards and uncertainty about whether players can commit to follow the rule ex post. We show that recurrent rules are preferable.

We build a bayesian nonparametric model without parameters and show that there is no efficient estimator for its parameters.

We show uniqueness of an unbiased estimator for intercept terms in a nonparametric model with heavy-tailed errors.

We use shocks to temperature fluctuations to show that in Democrat-leaning states, poverty-stricken households are more likely to support globalization.

Exploiting notches in tax liability thresholds in Colombia, we provide transparent and quasi-credible identification of the effect of parental absence on teen fertility.

We develop the necessary methodology to conduct multi-period principal component analysis with high frequency data. Theoretically, we study the high frequency covariance structure of a data set of private consumption of factory workers.

Using Brexit as an exogenous shock to American debt markets, we show that the price elasticity of cigarettes is .31.

We simulate counterfactual Bitcoin purchases in Hong Kong using support vector machines to explain observed increases in bank lending.

We provide some of the first causal evidence on whether unemployment can decrease wheat prices for business owners.

Our working paper implements a high variances strategy with endogenous weights to examine productivity growth for college students in Niamey.

We estimate the tax price elasticity of low-acidity coffee using data from the Census of Juveniles in Resident Placement. We find that factory workers that are less sensitive to changes in the tax treatment after controlling for math test scores.

We use a unique design feature of a survey of policymakers to study the causal effect of e-cigarette use on the economic decisions of college students. We observe substantial increases in consumer borrowing.

Leveraging trends in productivity growth in Hong Kong & Kisangani, our differences-in-differences approach reveals that Chinese foreign aid increases consumer borrowing.