Publications

Targeted Wage Subsidies and Firm Performance

Joint work with Oskar Nordström Skans and Johan Vikström

Labour Economics, 2018

This paper studies how targeted wage subsidies affect the performance of the recruiting firms. Using Swedish administrative data from the period 1998–2008, we show that treated firms substantially outperform other recruiting firms after hiring through subsidies, despite identical pre-treatment performance levels and trends in a wide set of key dimensions. The pattern is less clear from 2007 onwards, after a reform removed the involvement of caseworkers from the subsidy approval process. Overall, our results suggest that targeted employment subsidies can have large positive effects on post-match outcomes of the hiring firms, at least if the policy environment allows for pre-screening by caseworkers.



Working papers

Empirical Monte Carlo Evidence on Estimation of Timing-of-Events Models

Joint work with Johan Vikström and Gerard J. van den Berg

The Timing-of-Events (ToE) model is a standard approach in dynamic treatment evaluation. This paper uses an Empirical Monte Carlo simulation design to study the estimation of ToE models. We exploit rich Swedish data on unemployed individuals with information on participation in a training program to simulate placebo treatment durations. We then estimate ToE models by omitting some of the covariates previously used to simulate the placebo treatments. This generates unobserved heterogeneity correlated across the treatment ad outcome durations. When estimating ToE models, we use a discrete support point distribution for the unobserved heterogeneity, and we compare different specifications of the model. We find that the model performs well, in particular when time-varying covariates in the form of calendar-time variation are exploited for identification. For the discrete support distribution of the unobserved heterogeneity, we find that both too many mass points and too few mass points lead to large bias. We also find that information criteria that penalize parameter abundance are a very useful way to select the number of support points.

Comparing Sequence Data Models: Prediction and Dissimilarities

Joint work with Raffaella Piccarreta and Marco Bonetti

We consider the case where individuals are observed transitioning across different states over time, and we are interested in studying the resulting trajectories as a whole rather than the occurrence of specific events. This framework applies to a variety of settings in social and biomedical studies. Model‐based approaches, such as multi-state models or Hidden Markov models, are being increasingly used to analyze trajectories, but the different assumptions underlying alternative models typically make the comparison of their predictive performance difficult. In this work we introduce a novel way to accomplish this task based on microsimulation‐based predictions. We use simulated data and propose alternative criteria to evaluate a given model and/or to compare competing models with respect to their ability to generate trajectories similar to the observed ones.