A sequential model of older workers' labor force transitions after a health shock.

In this paper we estimate and validate a three-period sequential model of older workers' labor force transitions following a health/disability shock, using retrospective information from Spanish cross-section data. Central to the analysis are the effects of the various disabilities and their severity. We find that the probability of remaining employed decreases both with age and the severity of the shock. Moreover, we find strong interactions between age and severity for older workers and none for prime-age workers. Suffering any kind of disability reduces the probability of being employed immediately prior to retirement age, and in such cases it is severity which is the strongest indicator. With respect to demographics, we find that female gender, having a retired spouse or being married all reduce the probabilities of both remaining in employment and returning to work following a spell of inactivity; in turn, principal breadwinner status, education and skill levels increase this likelihood.

An empirical analysis of the demand for physician services across the European Union.

This paper presents parameter estimates for physician service equations using the "European Community Household Panel" for 12 countries covering the period 1994-1996. The focus is on two specific points: (1) the identification of behavioural similarities and differences in the demand for health services across the 12 countries; (2) the variability in demand for health services represented through a joint model for all countries. We found that there are significant differences among countries, although there are also similarities in the effect of variables such as health stock, labour situation or family structure. An important fraction of the variability in the demand for health services across countries could be explained by differences in age, income and the role of general practitioners as gatekeepers in the public health system. We found some evidence of induced demand in the decision to visit a specialist and in the number of such visits.

Latent class versus two-part models in the demand for physician services across the European Union.

Using three waves of data from the European Community Household Panel, this paper estimates demand for physician services equations for 12 European countries. We focus on the selection of the most appropriate econometric specification for visits to general practitioners and to specialists among two-part and latent class models. The distinction between the demand of services from these two types of physicians allows us to distinguish cases in which two-part perform better than latent class models, evidence which is different from previous findings in the literature. The results suggest that latent class models are more appropriate than two-part models to estimate general practitioners utilisation while the opposite is found for visits to the specialists.