Nationwide health interventions are difficult to evaluate as contemporaneous control groups do not exist and before-after approaches are usually infeasible. We propose an alternative semi-parametric estimator that is based on the assumption that the intervention has no direct effect on the...

Randomized experiments provide policy relevant treatment effects if there are no spillovers between participants and nonparticipants. We show that this assumption is violated for a Danish activation program for unemployed workers. Using a difference-in-difference model we show that the...

This paper presents and describes a new database of major minimum wage and collective bargaining reforms covering 26 advanced economies over the period 1970-2020. The main advantage of this dataset is the precise identification of the nature and date of major reforms, which is valuable in many...

Many countries support business start-ups to spur economic growth and reduce unemployment with different programmes. Evaluation studies of such programmes commonly rely on the conditional independence assumption (CIA), allowing a causal interpretation of the results only if all relevant...

We show that a transfer targeting a minority of the population is sustained by majority voting, however small the minority targeted, when the probability to receive the transfer is decreasing and concave in income. We apply our framework to the French social housing program and obtain that...

In this paper we estimate labor force participation responses for married women in Sweden using population-wide register data and detailed information about individuals' budget sets. For identification we exploit a reform in the system for housing allowances in 1997 which affected participation...

This paper investigates the robustness of determinants of economic growth in the presence of model uncertainty, parameter heterogeneity and outliers. The robust model averaging approach introduced in the paper uses a flexible and parsimonious mixture modeling that allows for fat-tailed errors...

We argue that in modelling cross-country growth models one should first identify so-called outlying observations. For the data set of Sala-i-Martin, we use the least median of squares (LMS) estimator to identify outliers. As LMS is not suited for inference, we then use reweighted least squares...