瑞士国家银行

This study considers the practically important case of nonparametrically estimating heterogeneous average treatment effects that vary with a limited number of discrete and continuous covariates in a selection-on-observables framework where the number of possible confounders is very large. We propose a two-step estimator for which the first step is estimated by machine learning. We show that this estimator has desirable statistical properties such as consistency, asymptotic normality, and rate double robustness. In particular, we derive the coupled convergence conditions between the nonparametric and the machine-learning steps. We also show that estimating population average treatment effects by averaging the estimated heterogeneous effects is semiparametrically efficient. The resulting estimators are compared to other suggestions in the literature in Monte Carlo experiments that are inspired by real data. They are found to perform relatively better in most settings. The new estimators are applied to the empirical example of the effects of mothers' smoking during pregnancy on the birthweight of their babies.

Double/debiased machine learning (DML) uses for estimating an average treatment effect (ATE) a double-robust score function that relies on the prediction of nuisance functions, such as the propensity score, which is the probability of treatment assignment given covariates. Estimators relying on double-robust score functions are highly sensitive to errors in propensity score predictions. Machine learning algorithms have been found to produce models that often overestimate or underestimate these probabilities. Several calibration approaches have been proposed to improve probabilistic forecasts of machine learners. This paper explores their integration into the DML framework, showing via simulations that using calibrated propensity scores significantly reduces the root mean squared error of ATE estimates in finite samples while preserving DML's asymptotic properties.

This paper examines the impact of Ethiopia's Road Sector Development Program (RSDP) from 1997-2016 on land use and economic activity, using spatial variation in road upgrades and satellite imagery. We use three approaches to triangulate the effect of the RSDP: difference-in-differences examining economic activity near upgraded roads, long-difference instrumental variables comparing areas connected to upgraded versus non-upgraded roads, and market access analysis. For RSDP phases I-III, upgrades increase local economic activity, with effects varying by baseline activity. Areas with medium-to-high baseline activity showed positive effects, while areas with the lowest activity showed minimal effects. Results are weaker for RSDP IV.

Bank of England governor Mark Carney warned after the 2016 Brexit vote that the UK is reliant on the “kindness of strangers” to fund its increasing current account deficit. In this paper, we examine whether firms in Switzerland attenuated their investments in the UK following the Brexit vote or whether British-controlled firms in Switzerland repatriated their foreign assets to the UK. Three empirical findings in the bilateral context suggest that Carney’s warning was overly cautious. First, Carney focused strictly on the foreign willingness to invest in the UK; however, the alternative channel of repatriating British assets abroad is equally important. Second, capital inflows and outflows are positively correlated not only in the aggregate, but also across a range of subgroupings at the firm level. Third, the nonuniform firm response to the Brexit vote suggests that understanding aggregate capital waves is more complicated at the firm level.

We present new empirical evidence on the role of granularity in the current account (CA), using a unique and comprehensive firm-level dataset for Switzerland. We show that idiosyncratic shocks to large firms account for almost two thirds of the fluctuations in the headline CA and are the primary source of CA volatility. The granular effect is present across goods, services, and income components and persists over both short- and medium-term horizons. In addition to their direct impact, idiosyncratic shocks propagate through inter-firm linkages via input–output relationships and cross-product connections associated with multinational enterprise activity. Our findings complement standard macroeconomic models by providing a micro-founded perspective and demonstrating the importance of granularity in explaining fluctuations in external balances.