Do income and wealth mediate associations between fertility histories and later life health?
Principal Investigator: Emily Grundy (PATHWAYS)
Co-investigators: Mike Kenward (PATHWAYS), George Ploubidis (PATHWAYS), Sanna Read (PATHWAYS), Monica Costas Dias (PEPA), James Banks (Institute for Fiscal Studies)
One of the substantive projects with the PATHWAYS node seeks to address the question "To what extent does stress, social support and health related behaviour mediate the effect of fertility history and childhood circumstances on later life health?" using data from the English Longitudinal Study of Ageing. It is also well recognised that childbearing and rearing have substantial economic implications and that both income and wealth are associated with differentials in health at older ages.
This collaborative project builds on the PEPA node’s knowledge and expertise on the measurement of income and wealth and previous track record of research on estimation of economic effects of childbearing with the PATHWAYS team’s interest and previous work on biosocial effects of fertility histories. The collaboration draws on previous work by both PATHWAYS and PEPA/IFS on transfers from older parents to adult children. The project team also work on identifying exogenous sources of variation in fertility.
The aim of this collaboration is principally to assess the extent to which associations between particular fertility trajectories and later life health may be mediated by fertility effects on income and wealth. Additional aims include producing cohort specific estimates of direct and indirect costs of childrearing, under various assumptions, and estimating effects of childbearing/rearing histories on wealth accumulation and depletion.
The project team use data from the English Longitudinal Study on Ageing, including accessing linked income data in a secure setting. The first stage of the work involves estimating direct costs of children for various cohorts/educational strata. The more complex estimation of indirect costs involves taking account of changing tax and benefit regimes and the fact that educational and career plans of women (and to some extent men) may be influenced by planned fertility (meaning that opportunity costs may be underestimated). Estimating effects on wealth is even more complex and involve examining further data on material transfers of parents to children. In the final sets of analyses, we will adopt a structural equation modelling approach to estimate the main pathways, including latent variables to deal with measurement error when feasible, and with appropriate Monte Carlo sensitivity analyses to account for model misspecification.