Do Working and Parenting Trajectories Influence Retirement Timing? Evidence From Spain, Using a Sequence Analysis Approach and Focusing on Women

Abstract

This paper investigates the association between accumulated experiences in the working and parenting spheres and retirement timing among Spanish women, utilizing data from the Survey of Health, Ageing and Retirement in Europe (SHARE). The study examines whether labor market attachment or cumulative disadvantage characterizes women’s retirement and how these factors relate with the number of children. Firstly, multichannel sequence analysis and cluster analysis were employed to identify distinct work–family life courses from ages 20 to 50. Seven clusters were identified: “Full-time work with 2+ children,” “Out of the labour force and 2+ children,” “Mid-life discontinuity and 2+ children,” “Full-time work and 1 child,” “Full-time work and childless,” “Part-time work and 2+ children,” and “Part-time work or out of the labor force and 1 child.” Secondly, the study examines the association between these identified groups and retirement timing. The results indicate that Spanish women who have experienced more unstable careers characterized by extended periods of inactivity or part-time work, particularly in combination with having 2 or more children, are more likely to retire at a later age compared with women with continuous careers, with or without children. These findings underscore the ongoing challenges of balancing unpaid care work and wage work in Spain, while also revealing notable variations among women.

Housewives never retire!? Gender biases in popular sample definitions for studies on the elderly

“…For example, the retirement age or reasons are only surveyed if respondents self-identify as retirees in SHARE and therefore automatically replicate the gendered sample bias. Still, studies on retirement transitions or age use self-reported retirement status straightforwardly as a measure of being retired (e.g., Radl, 2013;Tambellini, Danielsbacka, & Rotkirch, 2023). While some studies are transparent about the sample selectivity (e.g., Ponomarenko, Leist, & Chauvel, 2019;Sohier, Van Ootegem, & Verhofstadt, 2021;Tambellini, 2023), they do not discuss the implications of the sample selection for the results.…”