Econometric Evaluation of Labour Market Policies

This paper exposits and relates two distinct approaches to bounding the average treatment effect. One approach, based on instrumental variables, is due to (Manski 1990, 1994), who derives sharp bounds on the average treatment effect under a mean independence form of the instrumental variables (IV) condition. The second approach, based on latent index models, is due to (Heckman and Vytlacil 1999, 2000a), who derive bounds on the average treatment effect that exploit the assumption of a nonparametric selection model with an exclusion restriction. In this paper, we study the relationship between the two sets of bounds implied by these alternative conditions. We show that: (1) the Heckman and Vytlacil bounds are sharp given their assumption of a nonparametric selection model; (2) the Man-ski IV bounds simplify to the Heckman and Vytlacil bounds under the non-parametric selection model assumption.

There has been much work on identification and inference with instrumental variables in the last decade. Researchers have investigated conditions for identification of causal effects without normality, linearity, and additivity assumptions. In this discussion, I will comment on some of the new results in this area and discuss some implications for applied researchers in the context of some specific examples, focussing on identification rather than inference. Most of the comments will be limited to the case with a binary endogenous

The assumption that the assignment to treatments is ignorable conditional on attributes plays an important role in the applied statistic and econometric evaluation literature. Another term for it is conditional independence assumption (CIA). This paper discusses identification using CIA when there are more than two types of mutually exclusive treatments. It turns out that low dimensional balancing scores, similar to the ones valid in the case of only two treatments, exist and can be used for identification of various causal effects. Therefore, a comparable reduction of the dimension of the estimation problem is achieved and the approach retains its basic simplicity. The paper also outlines a matching estimator potentially suitable in that framework.

This paper considers the use of statistical profiling to allocate persons to alternative options within government programs, or to participation or non-participation in programs Profiling has been used in the United States to allocate unemployment insurance (UI) claimants to reemployment services based on the predicted duration of their UI claim. We place profiling in the context of the choice among alternative assignment mechanisms. Different mechanisms have different costs and benefits — any one mechanism, whether profiling or something else, may not be optimal for every program. Within profiling systems, we highlight the need for clarity regarding the objective of the assignment mechanism, e.g. equity or efficiency, and we discuss situations in which equity and efficiency goals may conflict. In relation to UI profiling in the United States, we provide empirical evidence from the state of Kentucky on two important questions. First, we demonstrate that it is possible to effectively predict the duration of UI spells, but that effectively doing so requires using more covariates than many US states presently do. This finding is important because effective prediction of the profiling variable is a necessary but not sufficient condition for the success of a profiling system. Second, we show that the impact of reemployment services does not appear to vary with expected duration of the UI spell, indicating that UI profiling in Kentucky does not advance the goal of efficiency, though it may advance equity goals.

In this paper we apply the statistical framework recently proposed by (Imbens 1999) and (Lechner 1999) to identify the causal effects of multiple treatments under the conditional independence assumption. We show that under this assumption, matching with respect to the ratio of the scores allows to estimate nonparametrically the average conditional treatment effect for any pair of treatments. Consequently it is possible to estimate this effect by implementing non-parametric matching estimators, which were recently studied by Heckman, Ichimura, Smith and Todd (1998) and (Heckman, Ichimura and Todd (1998)). The application concerns the youth employment programs which were set up in France during the eighties to improve the labour market prospects of the most disadvantaged and unskilled young workers. The empirical analysis makes use of nonexperimental longitudinal micro data collected by INSEE (Institut National de la Statistique et des Etudes Economiques, Paris) from 1986 to 1988.

This paper uses an administrative dataset to analyze to what extent active labor market policies in the Slovak Republic have been beneficial for unemployed workers. The focus is on two types of temporary subsidized jobs and on training. Short-term subsidized jobs seem to be the most efficient active labor market policy. Workers that are or have been on a short-term subsidized job have a higher job finding rate than other unemployed workers have and once they find a job they have a lower job separation rate than workers that have not been on a short-term subsidized job. Long-term subsidized jobs have a negative effect on the job finding rate and no effect on the job separation rate. The positive effect of training on the job finding rate of unemployed workers may have to do with reversed causality: some workers enter a training program only after they are promised a job. Training does not seem to affect the job separation rate.

We evaluate an Italian programme which promotes youth entrepreneurship by issuing substantial benefits to candidate entrepreneurs selected through a screening process and by providing them with some training. Following previous informal analyses of this programme performance the outcome with respect to which we perform the evaluation is the firm survival time. We argue that what matters for the success of the programme is not how effective the subsidy is per se in keeping firms alive but how well selected and trained the beneficiaries are. A major implication is that the impact of the subsidy on the firms survival chances is not the parameter of interest here; rather, we need to assess whether as a result of the screening and the training process selected firms are enough good to survive on their own. After constructing a sample of spontaneous firms comparable to the subsidised ones with respect to some observable characteristics by matching and weighting we show that previous results favourable to the programme badly overstated its effectiveness. We also show that a simple criterion based on the pattern of the hazard function casts serious doubts on the programme ability to yield firms whose survival chances do not depend on the subsidy.

This paper estimates the effects of vocational rehabilitation on the probability of improved health status and the reintegration of program participants into the labour market. Bivariate probit models are used to estimate the probability of selection to a rehabilitation program, and its effect on restoring the work capacity of the participants and their rejoining the labour market. The empirical application is based on a sample of individuals residing in Western Sweden who qualified for the vocational rehabilitation programs and who were registered with long-term sickness during 1991-1994. The results show variations in the fraction of the sample selected to participate in the rehabilitation programs, in the effects of the programs, and in the importance of individual heterogeneity for the outcome of programs. Participation in vocational rehabilitation programs is found to have a positive effect on the participants' health status and on their rate of return to work. There was weak or no evidence of selection based upon the unobservable characteristics of individuals that are most likely to regain their health or individuals with a higher potential to gain re-employment. From a social point of view the social aspects and health-related needs of the participants are valued more than the economic efficiency of the vocational rehabilitation programs.

The paper presents an econometric evaluation of the effects of subsidised non-profit temporary employment agencies - a programme of the German active labour market policy - on individual labour market outcomes. The empirical analysis is based on individual data from files for computer-assisted job brokering in Rhineland-Palatinate. Econometric estimates based on matching methods suggest that the programme generated a statistically significant additional reintegration success of about 13 per centage points. Within the first four months after leaving the programme, the cumulative effect was about half a month additional employment for the participants. Sensitivity and heterogeneity analyses indicate that the group of unemployed persons with higher participation probabilities are responsible for the positive mean programme impact.