The phase of retirement has been steadily growing since the 1950s. Rules and regulations related to retirement have been changing. Recent cohorts have reached this phase in better health than previous ones. Until recently, retirement has been rather void of societal expectations, leaving room for individual decisions regarding amount and type of activity. Thus, investigating activity patterns displayed during this life phase, their predictors and outcomes seems overdue. The study addresses three questions: (a) Which distinct clusters of productive activities among retirees can be identified in Germany? (b) Do activity patterns of clusters follow complementary or substitutive composition rules? (c) Which are the most important predictors of cluster membership? Using probability-based sample data (N = 2,141) from the survey ‘Transitions and Old Age Potential’ (TOP), this study investigated clusters of productive activities among retirees aged 60–70 years in Germany. The activities examined included paid work after retirement, formal and informal volunteering, child care and care-giving. Results showed a four-cluster structure. The clusters (Multiple Engagers, Volunteers, Family Helpers and Family Disengagers) differed with regard to the composition and the intensity of productive activities. Both complementary and substitutive relations were identified within clusters. Individual, familial and economic resources were predictors of cluster membership. Results are discussed with regard to role theory, cumulative inequality theory and the ongoing debate about old-age potential.

This study investigates the variations by older people's socio-economic status (SES) (i.e. educational level and social class) in the use of informal and formal help from outside the household in Great Britain, Italy, Belgium and The Netherlands. In all these countries, it was older people in low SES groups who mostly used such help. Multinomial logistic regression analyses showed that, in each country and for both types of help, there were SES gradients in the utilisation of both formal and informal care, and that differences in age, health and marital status largely accounted for the former but not the latter. Cross-national differences in the use of both informal and formal help remained when variations in sex, age, SES, health, marital status, home ownership and the use of privately-paid help were taken into account. Significant interaction effects were found, which indicated that older people in low SES groups in Great Britain and The Netherlands had higher odds of using informal help from outside the household than their counterparts in Italy, and similarly that those in The Netherlands were more likely to use formal help than their Italian peers. The results are discussed in relation to the cultural differences and variations in the availability of formal services among the countries.