Empirical Economics

Boes, Stefan

doi: 10.1007/s00181-010-0354-ypmid: N/A

Treatment analyses based on average outcomes do not immediately generalize to the case of ordered responses because the expectation of an ordinally measured variable does not exist. The proposed remedy in this paper is a shift in focus to distributional effects. Assuming a threshold crossing model on both the ordered potential outcomes and the binary treatment variable, and leaving the distribution of error terms and functional forms unspecified, the paper discusses how the treatment effects can be bounded. The construction of bounds is illustrated in a simulated data example.

Farré, Lídia; Klein, Roger; Vella, Francis

doi: 10.1007/s00181-010-0376-5pmid: N/A

An innovation which bypasses the need for instruments when estimating endogenous treatment effects is identification via conditional second moments. The most general of these approaches is Klein and Vella (J Econom 154:154–164, 2010), which models the conditional variances semiparametrically. While this is attractive, as identification is not reliant on parametric assumptions for variances, the nonparametric aspect of the estimation may discourage practitioners from its use. This paper outlines how the estimator can be implemented parametrically. The use of parametric assumptions is accompanied by a large reduction in computational and programming demands. We illustrate the approach by estimating the return to education using a sample drawn from the National Longitudinal Survey of Youth 1979. Accounting for endogeneity increases the estimate of the return to education from 6.8 to 11.2%.

Klein, Tobias

doi: 10.1007/s00181-010-0355-xpmid: N/A

Recent studies debate how the unobserved dependence between the monetary return to college education and selection into college can be characterised. This paper examines this question using British data. We develop a semiparametric local instrumental variables estimator for identified features of a flexible correlated random coefficient model. These identified features are directly related to the marginal and average treatment effect in policy evaluation. Our results indicate that returns to college systematically differ between actual college graduates and actual college non-graduates. They are on average higher for college graduates and positively related to selection into college for 96% of the individuals. The dependence between selection into college and returns to college education is strongest for individuals with low math test scores at the age of 7, individuals with less educated mothers, and for working-class individuals.

Cockx, Bart; Goebel, Christian; Robin, Stéphane

doi: 10.1007/s00181-010-0357-8pmid: N/A

This article investigates whether income support for low-paid part-time workers in Belgium increases the transition from unemployment to non-subsidised, ‘regular’ employment. Our analysis uses a sample of long-term unemployed young women. Observing their labour market histories from 1998 to 2001, we implement the ‘timing of events’ method to identify the treatment effect. Our results suggest that participation in the policy has a significantly positive effect on the transition to regular employment. Participation reduced the survivor rate in unemployment by 27% points 1 year after the start of the programme. The time spent in the programme did not affect the transition to regular employment.

Zabel, Jeffrey; Schwartz, Saul; Donald, Stephen

doi: 10.1007/s00181-010-0387-2pmid: N/A

We measure the impact of the Self-Sufficiency Project (a randomized welfare-to-work experiment in Canada; henceforth, SSP) on relative wage progression. SSP provided a generous 3-year earnings supplement to treatment group members who found a full-time job within a year of the start of the experiment (take-up group). We estimate the treatment on the treated for two sub-groups of the take-up group: the incentivized and non-incentivized groups. Using an econometric model of wage determination, we find evidence of large and significant relative wage progression of approximately 9 percentage points during the 3-year supplement period for the incentivized group. The impact for the non-incentivized group is much smaller (at most 3 percentage points). There is also some limited information that the non-incentivized group in New Brunswick and the incentivized groups in both New Brunswick and British Columbia continued to work more after the 3-year supplement period ended.

Cavaco, Sandra; Fougère, Denis; Pouget, Julien

doi: 10.1007/s00181-010-0391-6pmid: N/A

In this article, we estimate the effects of a French retraining program on the re-employment rate of displaced workers by matching techniques. This program, called ‘Conventions de conversion’, was intended to improve re-employment prospects of displaced workers by proposing them retraining and job seeking assistance for a period of 6 months beginning just after the dismissal. Our empirical analysis is based upon non-experimental data collected by the French Ministry of Labour. Matching estimates show that this program succeeded in increasing the employment rate of trainees by approximately 6 points of percentage in the medium-term, namely in the 2nd and 3rd years after the date of entry into the program. This improvement is essentially due to an increase of their re-employment rate in regular jobs, namely jobs under long-term labour contracts.

Arpino, Bruno; Aassve, Arnstein

doi: 10.1007/s00181-010-0356-9pmid: N/A

This article aims to answer to what extent fertility has a causal effect on households’ economic wellbeing—an issue that has received considerable interest in development studies and policy analysis. However, only recently has this literature begun to give importance to adequate modelling for estimation of causal effects. We discuss several strategies for causal inference, stressing that their validity must be judged on the assumptions we can plausibly formulate in a given application, which in turn depends on the richness of available data. We contrast methods relying on the unconfoundedness assumption, which include regressions and propensity score matching, with instrumental variable methods. This discussion has a general importance, representing a set of guidelines that are useful for choosing an appropriate strategy of analysis. The discussion is valid for both cross-sectional or panel data.