Are storks striking for a contract renewal? Childbirth under changing employment, family and welfare arrangements: Full Research Report ESRC End of Award Report

This study uses nationally representative longitudinal data –specifically, the European Community Household Panel (ECHP, 1994-2001), the British Household Panel study (BHPS, up to wave 15) and the Italian Panel Study (ILFI, up to wave 4)– to systematically examine the socio-demographic and job characteristics that help predict the timing of entry into -and exit from- the labour market as well as childbirth, both at the individual and the household level. It goes beyond traditional event history models by adding a focus on how some unobservable traits, preferences or characteristics that make individuals different from one another (unobserved heterogeneity) might influence these careers; and on how employment and fertility choices might be to some extent jointly determined (endogeneity). Disregarding these aspects produces biased estimates of the effects of predictors of interest on the risk of experiencing such events at the population level. This report presents findings from the analyses of employment sequences, event history analysis models and simultaneous equations modelling and has three main objectives: • to identify how uncertainty in employment relationships may influence individuals’ childbirth decisions; • to establish if employment participation and the decision to have a child are interrelated processes; and • to explore if and how men’s and women’s employment decisions are related to their joint fertility choices. Key findings Uncertainty in employment relations (temporary contracts) strongly influences employment exits but not fertility in the UK and very little so in Italy • Household fertility choices are influenced by both observed and unobservable partners’ individual characteristics and by their employment decisions. Temporary employment, however, does not seem to directly impact on childbirth decisions either in the analyses with ECHP data (for Denmark, Germany, Italy or the UK) or in those with national panel data for the United Kingdom and Italy (where a very small effect is found). Employment and fertility careers are jointly shaped only for women • Once other factors, observed and unobservable, are accounted for a higher fertility is associated with a lessened attachment to employment for women but not for men in both Italy and the UK. • The interrelation between fertility and employment is stronger for women than for men: in those households were women tend to exit employment more easily, fertility tends to be higher (both in UK and Italy); in the UK, those households where women tend to enter more frequently (or earlier) into employment after interruptions have a lower fertility; Partner’s careers are interdependent • Although there is no direct interdependence between men’s employment participation decisions and the couple’s fertility, men’s employment is endogenously related to their partners’ employment decision (which in turn is linked to fertility) • Men’s and women’s employment participation (and interruption) choices are to some extent jointly shaped. • Some unobserved traits, in both countries, point to a polarisation between households’ behaviour in face of men’s employment exits: in households where men are more likely to exit employment, women tend to either exit more themselves (either due to homogamy or maximising on welfare benefits) or to enter more (compensating for income loss). This second household type is more exposed to the risk of falling into a poverty trap. • Simultaneous equation modelling is able to reduce the bias in estimates that traditional models bring about by ignoring individual unobservable traits and preferences (unobserved heterogeneity); it also accounts for the partial joint determination (endogeneity) of the fertility and employment careers, and that of both partners’ employment trajectories. These finding, by providing an answer to the research questions, show how all the objectives of the project have been fully met.