Maison de l'économie et de la gestion d'Aix
424 chemin du viaduc
In this paper, we propose a new variance reduction method for quantile regressions with endogeneity problems, for alpha-mixing or m-dependent covariates and error terms. First, we derive the asymptotic distribution of two-stage quantile estimators based on the fitted-value approach under very general conditions. Second, we exhibit an inconsistency transmission property derived from the asymptotic representation of our estimator. Third, using a reformulation of the dependent variable, we improve the efficiency of the two-stage quantile estimators by exploiting a tradeoff between an inconsistency confined to the intercept estimator and a reduction of the variance of the slope estimator. Monte Carlo simulation results show the fine performance of our approach. In particular, by combining quantile regressions with first-stage trimmed least-squares estimators, we obtain more accurate slope estimates than 2SLS, 2SLAD and other estimators for a broad set of distributions. Finally, we apply our method to food demand equations in Egypt.
The main two methods of endogeneity correction for linear quantile regressions with their advantages and drawbacks are reviewed and compared. Then, we discuss opportunities of alleviating the constant effect restriction of the fitted-value approach by relaxing identification conditions.
Owing to recent concerns about the negative externalities of traditional fuel use on the environment and health, the issue of the household fuel transition in developing countries, from dirty fuels towards clean fuels, is receiving growing research attention. This paper provides an up-to-date survey of the economic literature on household fuel use in these countries. We first present the conceptual and theoretical frameworks. Then we discuss the empirical results that show the wide range of factors that drive the household fuel transition and can be understood better by linking them with theory.
Heterogeneity in how some independent variables affect a dependent variable is pervasive in many phenomena. In this respect, this paper addresses the question of constant versus nonconstant effect through quantile regression modelling. For linear quantile regression under endogeneity, it is often believed that the fitted-value setting (i.e., replacing endogenous regressors with their exogenous fitted-values) implies constant effect (that is: the coefficients of the covariates do not depend on the considered quantile, except for the intercept). Here, it is shown that, under a weakened instrumental variable restriction, the fitted-value setting can allow for nonconstant effect, even though only the constant-effect coefficients of the model can be identified. An application to food demand estimation in 2012 Egypt shows the practical potential of this approach.
In this paper, we propose a robust test of exogeneity. The test statistics is constructed from quantile regression estimators, which are robust to heavy tails of errors. We derive the asymptotic distribution of the test statistic under the null hypothesis of exogeneity at a given quantile. The finite sample properties of the test are investigated through Monte Carlo simulations that exhibit not only good size and power properties, but also good robustness to outliers.
Training costs may hamper intra-firm human capital accumulation. As a consequence, firms may be tempted to have workers pay for their on-the-job training (OJT). In this paper, we analyse the links of OJT and worker remuneration in the suburb of Tunis, using case study data for eight firms. We find that the duration of former OJT negatively influences starting wages, while there is no anticipated effect of future training on wages at the firm entry. In contrast, current wages are positively affected by former OJT but negatively affected by ongoing OJT. These results provide very rare empirical support in Less Developed Countries (LDCs) for classical human capital theories and cost sharing theories applied to OJT.
In this paper, I test and reject the separability of production and consumption decisions of agricultural households in Ethiopia, using data from a rural household survey conducted in 1994 and an estimated labor demand equation. I also elicit socio-demographic and asset variables that are positively linked with agricultural labor demand. These results reflect the limited development of fully organized labor markets in rural Ethiopia. They also imply that price subsidies, taxes and other purely market-driven agricultural policies may have only limited or perverse impacts. They should be complemented by policies directly affecting household decisions, such as food aid, technology transfer, free supply of fertilizers and so on.
We conduct a case study of the linkages of task organization, human capital accumulation and wages in Morocco, using matched worker-firm data for Electrical-mechanical and Textile-clothing industries. In order to integrate task organization into the interacting processes of workers’ training and remunerations, we assume a recursive model, which is not rejected by our estimates: task organization influences on-the-job training that affects wages. Beyond sector and gender determinants, assignment of workers to tasks and on-the-job training is found to depend on former education and work experience in a broad sense. Meanwhile, participation in on-the-job training is stimulated by being assigned to a team, especially of textile sector and for well-educated workers. Finally, task organization and on-the-job training are found to effect wages.
We argue that the economic evaluation of health care (cost–benefit analysis) should respect individual preferences and should incorporate distributional considerations. Relying on individual preferences does not imply subjective welfarism. We propose a particular non-welfarist approach, based on the concept of equivalent income, and show how it helps to define distributional weights. We illustrate the feasibility of our approach with empirical results from a pilot survey.
We provide a unified treatment of the two approaches pioneered by Atkinson and Bourguignon (1982, 1987) [3,4] by resorting to compensation principles in the bivariate case. We treat the attributes of individual utility asymmetrically by assuming that one attribute can be used to compensate another. Our main result consists of two sufficient second-order stochastic dominance conditions. In the case where the compensated variable has a discrete distribution, the distribution of the compensating variable must satisfy a condition which degenerates to the Sequential Generalized Lorenz test for identical marginal distributions of the compensated variable. Furthermore, the distributions of the compensated variable must satisfy the Generalized Lorenz test.