GRIGNON Michel

This doctoral thesis focuses on the economics of aging. It examines in greater detail the determinants of individual decisions in terms of retirement, health prevention and the purchase of long-term care insurance. Chapter 1 starts from the observation that financial incentives to encourage individuals to delay their retirement have relatively limited and heterogeneous effects. It explores the role of another potential determinant of the retirement decision: the time inconsistency bias. Using French survey data, the econometric analysis shows that individuals exhibiting time inconsistency are much less likely to delay their retirement in order to benefit from a pension bonus (surcote). Chapter 2 examines the causal effect of retirement on preventive and risky behaviors within couples. This article is the first to take into account coordination between spouses and the effects of asymmetrical externalities of behaviour in this type of evaluation. The econometric results on European survey data reveal heterogeneous effects depending on the behaviors studied, but also on the status of the pre-retirement job and the order of retirement within the couple. Chapter 3 examines the nature of the relationship between prevention effort and the demand for LTC insurance, in the context of the introduction of a prevention program by an insurer. The equilibrium predictions of a theoretical insurance model show that the prevention program does not lead to a reduction in self-protection effort, but does have an ambiguous effect on the choice of insurance coverage rate. The results of the associated econometric analysis on French survey data show a positive effect of the intention to participate in the programme on the prevention effort and on the insurance decision, thus rejecting the existence of an eviction effect. They also indicate the existence of advantageous selection in the French LTC insurance market.

This thesis contributes to the understanding of human capital investment decisions of young adults in France. Young adults are in a decisive period in terms of their development and expression of preferences. Particular attention is paid to preferences, as well as to the resulting inequalities. The first chapter focuses on the roles of anticipatory treatment and multivariate preferences in the health insurance decision. Predictions from a theoretical model are tested on data collected in an experimental laboratory. Results show that higher health preference leads to more intensive treatment demand and that being correlation averse leads to more than full coverage. In the second chapter, using survey data, it is shown that differences in the use of care are primarily associated with need, followed by circumstances, reflecting inequalities of opportunity, and effort, reflecting fair inequalities. The third chapter focuses on the effect of students' psychological fragilities on their control beliefs. The instrumental variable strategy shows that more psychological frailties lead to increased control beliefs, which is consistent with the psychological literature that depressed and anxious individuals blame themselves more.