The Relationship between Formal and Informal Care in Europe and Its Implications for the Number of Caregiving Hours. ENEPRI Research Report No. 100, November 2011

Jiménez-Martín, Sergi and Sánchez, Raquel Vegas and Prieto, Cristina Vilaplana (2011) The Relationship between Formal and Informal Care in Europe and Its Implications for the Number of Caregiving Hours. ENEPRI Research Report No. 100, November 2011. UNSPECIFIED.

Preview

PDF - Published Version Download (552Kb) | Preview

Abstract

Understanding the factors that determine the type and amount of formal care is important for assessing the need for care in European nations and developing consistent long-term policies. In this report, the provision of care in terms of its extensive (choice of care) and intensive qualities (the number of hours of care received) is analysed. Following the methodology proposed in Bourguignon et al. (2007) and using SHARE data, we estimate a sample selection model with the particularities that the first step is a multinomial logit model and the second step is a standard regression equation. The analysis is carried out for representative countries as well as for the pooled sample of countries available in SHARE. At the country level, the results obtained vary depending on the country considered: the Bourguignon model is a valid model for Spain and Italy, given that the task-specific approach and the complementarity paradigm prevail in these countries. On the other hand, the selectivity terms are not significant in the Netherlands or in Germany. The latter results, however, are highly conditioned by the small number of observations we have by country. Finally, we have analysed the pooled sample of countries grouped under three different criteria: by geographical clusters, by the generosity and by the characteristics of their long-term care systems. In the first-stage multinomial logit results, we find that the taskspecific/ complementarity model cannot be rejected in all cases. Furthermore, the second-stage selectivity terms are found to be significant in all cases. This implies that the bias of the LS estimates for the hours equations can be sizeable.

Actions (login required)