Three substantive chapters examine labour market dynamics among older people. Chapter 2 analyses older men and women's labour market transitions using the British Household Panel Survey (BHPS). I find large peaks in exit rates out of employment at ages 60 and 65 occurring in the exact birthday month, suggesting strong incentive effects of pension schemes. Discrete-time hazard regression analysis shows that health and potential income out of work are the most important determinants of these transitions, with effects that are larger than found in previous studies for British and US men. When modelling unobserved heterogeneity, the estimated probability of being a mover between work and non-work is twice as high for women as for men. Chapter 3 analyses how spouses in older couples react to shocks to their partner's labour income using BHPS data. After a separation, wives reduce their labour supply while husbands tend to increase theirs. If a wife becomes unemployed, it does not affect her husband's labour supply while wives whose husband becomes unemployed reduce their labour supply, too. A decline in husband's health causes the wife to reduce her labour supply while husbands tend to increase theirs when facing a decline in wife's health. Partner's death does not have statistically significant labour supply effects. Chapter 4 analyses the relationship between cognitive functioning and employment among older men and women using data from the English Longitudinal Study of Ageing. Regression analysis shows that the change in cognitive functioning over time does not have any statistically significant effects on the probability to exit or enter employment, or on working hours. These results are not sensitive to the definition of work. My findings differ from earlier research on younger age groups in Germany and the USA where some effects of cognitive functioning on labour force participation were found.

Cette thèse étudie les pensions de réversion en France en mettant l’accent sur la dimension patrimoniale « implicite» des droits à la retraite. En effet, les droits à pensions constituent une composante de la « richesse »des assurés, appréhendée comme une épargne « implicite » encore appelée équivalent patrimonial des droits à la retraite (EPDR). Après avoir analysé, dans le chapitre 1, la grande diversité des règles d’ouverture et de service de la pension de réversion entre les régimes, ainsi que les différentes logiques sous-jacentes des pensions de réversion entre secteur privé et secteur public, nous montrons, dans le chapitre 2, que l’EPDR évalué est plus important, en moyenne, dans les régimes du secteur public en raison des qualifications plus importantes des conjoints défunts dans ces régimes et des conditions de liquidation des retraites plus avantageuses. Par ailleurs,l’inégalité dans la distribution de cette « richesse de pensions » est moindre que celle généralement observée sur le patrimoine réel des ménages, mais se décompose de manière différente selon l’ancien secteur d’activité du conjoint défunt. Aussi, plusieurs facteurs, autres que la pension de réversion, expliqueraient le niveau de l’EPDR. L’analyse, dans le chapitre 3, de l’impact sur l’EPDR de l’augmentation de la durée d’assurance lors des réformes de 1993 et de 2003 montre une réduction significative de l’EPDR des pensionné(e)s dont le conjoint défunt avait validé au moins 60 trimestres de cotisation, mais était touché par la mesure. Enfin, dans le chapitre 4, nous montrons que les dispositifs de réversion permettent en moyenne aux veuves et aux veufs de maintenir leur niveau de vie antérieur au décès de leur conjoint, avec toutefois des nuances selon que le défunt était ancien cadre du privé, ancien non cadre du privé ou ancien fonctionnaire civil d’État. / This thesis examines survivors’ pensions in France by focusing on the implicit patrimonial dimension of pension rights. Indeed, pension rights are a component of the wealth of insured persons, considered as an implicit saving also called Pension wealth (PW). After examining in Chapter 1 the wide heterogeneity of rules governing the openness and service of survivors’ pensions between schemes, as well as the logic underlying survivors’ pensions between the private and public sectors, we show, in Chapter 2, that calculated Pension wealth is higher on average in public sector schemes due to the higher qualifications of the spouses deceased in these schemes and the more favorable retirement benefit conditions. Moreover, inequality in the distribution of Pension wealth is smaller than that generally observed on the real wealth of households, but Pension wealth is distributed differently according to the previous sector of activity of the deceased spouse. On the other hand, several factors, other than the survivor’s pension, would explain the level of Pension wealth. In Chapter 3, we show that the increase in the duration of insurance caused by the 1993 and 2003 reforms has significantly reduced Pension wealth for derived pensioners right whose deceased spouse had validated at least 60 quarters of contribution, but was affected by thoses measures.Finally, in Chapter 4, we show that survivors’ schemes on average allow widows and widowers to maintain their standard of living prior to the death of their spouses, although there are differences depending on whether the deceased was a private sector executive, a non-executive wage earner or a civil servant of the State.