Contributions published at Statistics and Empirical Economic Research (Rainer Winkelmann)

Much progress has been made in recent years on developing and applying a direct measure of utility using survey questions on subjective well-being. In this paper we explore whether this new type of measurement can be fruitfully applied to the study of interdependent utility in general, and altruism between parents and adult children who moved away from home in particular. We introduce an appropriate econometric methodology and, using data from the German SocioEconomic Panel for the years 2000-2004, find that the parents’ self-reported happiness depends positively on the happiness of their adult children. A one standard deviation move in the child’s happiness has the same effect as a 45 percent move in household income.

We examine the effect of single motherhood on children’s secondary school track choice using 12-year-old children drawn from the German Socio-Economic Panel. In line with previous studies for the U.S., the U.K. and Sweden, we find a negative correlation between single motherhood and children’s educational attainment. Looking for alternative explanations for this correlation, we use probit regression models to control for factors related to single motherhood such as higher educational background, lower household income and higher labor supply of the mother. Our evidence suggests that single motherhood reduces school attainment mainly because it is associated with lower resources (household income) available for the child.

I consider the problem of estimating the effect of a health care reform on the frequency of individual doctor visits when the reform effect is potentially different in different parts of the outcome distribution. Quantile regression is a powerful method for studying such heterogeneous treatment effects. Only recently has this method been extended to situations where the dependent variable is a (non-negative integer) count. An analysis of a 1997 health care reform in Germany shows that lower quantiles, such as the first quartile, fell by substantially larger amounts than what would have been predicted based on Poisson or negative binomial models.

We discuss regression models for ordered responses, such as ratings of bonds, schooling attainment, or measures of subjective well-being. Commonly used models in this context are the ordered logit and ordered probit regression models. They are based on an underlying latent model with single index function and constant thresholds. We argue that these approaches are overly restrictive and preclude a flexible estimation of the effect of regressors on the discrete outcome probabilities. For example, the signs of the marginal probability effects can only change once when moving from the smallest category to the largest one. We then discuss several alternative models that overcome these limitations. An application illustrates the benefit of these alternatives.

It is a widely held opinion that apprenticeship training represents a net investment for training firms, and that therefore firms only train if they have the possibility to recoup these investments after the training period. A recent study using a new firm-level dataset for Switzerland showed, however, that for 60 percent of the firms, the apprenticeship training itself does not result in net cost. In this context it seems important to examine the question whether the potential net cost of training (during the training period) are a major determinant for the demand for apprentices. Different count data models, in particular hurdle models, are used to estimate the effect of net cost on the demand for apprentices. The results show that the net cost have a significant impact on the training decision but no significant influence on the demand for apprentices, once the firm has decided to train. For policy purposes, these results indicate that subsidies for firms that already train apprentices would not boost the demand for apprentices.

The previous literature on the determinants of individual well-being has failed to fully account for the interdependencies in well-being at the family level. This paper develops an ordered probit model with multiple random effects that allows to identify the intra-family correlation in wellbeing. The parameters of the model can be estimated with panel data using Maximum Marginal Likelihood. The approach is illustrated in an application using data for the period 1984-1997 from the German Socio-Economic Panel in which both inter-generational and intra-marriage correlations in well-being are estimated.

The paper analyzes the effect of mother tongue on labor market outcomes of Swiss residents. This type of analysis can shed light on an important policy question. Is the Swiss labor market well integrated, or can one find instead segmentation along language borders? Improving on previous research in this area, we use a nationally representative household survey, the Swiss Household Panel 1999 and 2000, and we explicitly account for self-selection of workers into language areas. Overall, we find no evidence to suggest that the Swiss labor market is not perfectly integrated or that internal migrants are positively selected.

Empirical studies on the relationship between income and happiness commonly use standard ordered response models, the most well-known representatives being the ordered logit and the ordered probit. However, these models restrict the marginal probability effects by design, and therefore limit the analysis of distributional aspects of a change in income, that is, the study of whether the income effect depend on a person’s happiness. In this paper we pinpoint the shortcomings of standard models and propose two alternatives, namely generalized threshold and sequential models. With data of two waves of the German Socio-Economic Panel, 1984 and 1997, we show that the more general models yield different marginal probability effects than standard models.

The German health care reform of 1997 provides a natural experiment for evaluating the price sensitivity of demand for physicians’ services. As a part of the reform, co-payments for prescription drugs were increased step up to 200%. However, certain groups of people were exempted from the increase, providing a natural control group against which the changed demand for physicians’ services of the treated, those subject to increased co-payments, can be assessed. The differences-in-differences estimates indicate that increased co-payments reduced the number of doctor visits by about 10% on an average.

The previous literature on the determinants of individual well-being has failed to fully account for the interdependencies in well-being at the family level. This paper develops an ordered probit model with multiple random effects that allows to identify the intrafamily correlation in well-being. The parameters of the model can be estimated with panel data using Maximum Marginal Likelihood. The approach is illustrated in an application using panel data for the period 1984-1997 from the German Socio-Economic Panel in which both inter-generational and intra-marriage correlations in well-being are estimated.

The apprenticeship system is the most important source of formal post-secondary training in Germany. Using German register data – the IAB Employment Sample – it is found that apprentices staying with their training firm after graduation have a longer first-job durations but not higher wages than apprentices leaving the training firm. Retention rates, first job durations, and post-apprenticeship wages are all increasing functions of training intensity. Some implications for the ongoing debate as to why firms are willing to invest in general training are discussed.

This paper evaluates the German health care reform of 1997, using the individual number of doctor visits as outcome measure and data from the German Socio-Economic Panel for the years 1995–1999. A number of modified count data models allow us to estimate the effect of the reform in different parts of the distribution. The overall effect of the reform was a 10% reduction in the number of doctor visits. The effect was much larger in the lower part of the distribution than in the upper part.

I consider the problem of evaluating the effect of a health care reform on the demand for doctor visits when the effect is potentially different in different parts of the outcome distribution. Quantile regression is a useful technique for studying such heterogeneous treatment effects. Recent progree has been made to extend such methods to applications with a count dependent variable. An analysis of a 1997 health care reform in Germany shows the benefit of the approach: lower quantiles, such as the 25 percent quantile, fell by substantially larger amounts than what would have been predicted based on Poisson or negative binomial models.

The paper analyses the effect of mother tongue on labor market outcomes of Swiss residents. This type of analysis can shed light on an important policy question. Is the Swiss labor market well integrated, or can one find instead segmentation along language borders? Improving on previous research in this area, we use a nationally representative household survey, the Swiss Household Panel 1999 and 2000, and we explicitly account for self-selection of workers into language areas. Overall, we find no evidence to suggest that the Swiss labor market is not perfectly integrated or that internal migrants are positively selected.

This paper reports on a re-evaluation of the German health care reform of 1997. A previous evaluation found a limited effect of a 4.4 percent reduction of the number of doctor visits in a sample of pharmacy customers. The re-evaluation based on a representative household survey, the German Socio-Economic Panel, yields a much larger effect. The paper uses this case study to discuss the methods and benefits of modern techniques of program evaluation.

This paper uses recent data for Germany and a new outcome variable to assess the consequences of parental separation on the well-being of youths. In particular, it is considered how subjective well-being, elicited from an ordinal 11-point general life satisfaction question, differs between youths living in intact and non-intact families, holding many other potential determinants of well-being constant using ordered probit regressions. The main finding of this study is that living in a non-intact family has not the hypothesised large negative effect on child well-being.

The apprenticeship system is the most important source of formal post-secondary training in Germany. Using German register data - the IAB Employment Sample - we find that apprentices staying with their training firm after graduation have longer first-job durations but not higher wages than apprentices leaving the training firm. Retention rates, first job durations, and post-apprenticeship wages are all increasing functions of training intensity. Some implications for the ongoing debate as to why firms are willing to invest in general training are discussed.

The German health care reform of 1997 provides a natural experiment for evaluating the price sensitivity of demand for physicians’ services. As part of the reform, copayments for prescription drugs were increased by up to 200 percent. However, certain groups of people were exempted from the increase, providing a natural control group against which the changed demand for physicians’ services of the treated, those subject to increased co-payments, can be assessed. The differences-in-differences estimates indicate that increased co-payments reduced the number of doctor visits by about 10 percent on average.

We analyse post-war Dutch migration to New Zealand. We document that history, reflect on analytical and econometric modelling and then combine a sample of Dutch migrants in New Zealand with a representative sample of Dutch in The Netherlands to estimate wage equations and the determinants of the migration decision. We use the results for ex post evaluation of the migration decision.

The paper evaluates the German health care reform of 1997, using the individual number of doctor visits as outcome measure and data from the German Socio- Economic Panel for the years 1995-1999. A number of modified count data models allow to estimate the effect of the reform in different parts of the distribution. The overall effect of the reform was a 10 percent reduction in the number of doctor visits. The effect was much larger in the lower part of the distribution than in the upper part.