The identification of average causal effects of a treatment in observational studies is typically based either on the unconfoundedness assumption or on the availability of an instrument. When available, instruments may also be used to test for the unconfoundedness assumption (exogeneity of the...

This paper discusses the evaluation problem using observational data when the timing of treatment is an outcome of a stochastic process. We show that, without additional assumptions, it is not possible to estimate the average treatment effect and treatment on the treated. It is, however,...

We test if social work norms are important for work absence due to self-perceived sick-ness. To this end, we use a randomized social experiment designed to estimate the effect of monitoring on work absence. The treated were exposed to less monitoring of their eligibility to use sickness...

In many cases assignment to a treatment may affect concomitant variables. I show how a concomitant variable can be used to corroborate evidence from an observational study. In the observational study two types of training programs are compared. One program is part of regular Swedish labor market...

We propose a general test for exogeneity that is robust against distributional misspecification. The test can also be used to identify other types of misspecifications, such as the presence of a random coefficient. The idea is to sort the data with respect to a variable (a sorting score) and...

The treatment effect of a Swedish employment subsidy is estimated using exact covariate-matching and instrumental variables methods. Our estimates suggest that the programme had a positive treatment effect for the participants. <p> We also show how non-parametric methods can be used to estimate the...</p>

Using both register and survey data, two types of labour market training programs are compared. One program is part of the regular Swedish labour market training and the other, Swit, was initiated as an experiment during a two-year period, in an attempt to solve a bottleneck problem with people...

This paper discusses the evaluation problem using observational data when the timing of treatment is an outcome of a stochastic process. We show that, without additional assumptions, it is not possible to estimate the average treatment effect and treatment on the treated. It is, however,...