Never again? (Why) rich parents enjoy children less

Marco Le Moglie, University of Turin
Letizia Mencarini, Università Bocconi - Dondena Centre for Research on Social Dynamics and Public Policy
Chiara Rapallini, Università di Firenze

In this paper we investigate the empirical relationship between subjective well-being (SWB) and fertility considering household and individuals' earned income and the gender of parents as mediators, with the German Socioeconomic Panel Survey data. The focus on income and gender is justified being the increase in women's earnings interpreted as a key driver behind fertility decline, while recent evidence suggests that the relationship between socio-economic development and fertility is turning positive at the highest level of development. With this paper we contribute to the open debate about the relationship between income and fertility at micro level, through the lens of SWB, studying the German case (country with one of the highest GDP per capita in the world, but with very low fertility). In particular, we estimate the parental SWB trajectories around the first childbirth, i.e. the onset of parenthood, because of the bidirectional nature of the relationship between SWB and fertility. If the parental SWB is affected by the first child, and subsequent parental SWB positively predicts the birth of the second as recent studies have shown, then income as a mediator of parental SWB around this first childbirth affects the probability of a second parity – the lack of which is the main responsible for German fertility decline. We found that, not only the gender, but also household and individual income affect differently SWB, with the latter strongly shaping mothers SWB’s trajectories. Women in the second and third tertile of income distribution show a negative effect in the years following the childbirth with respect not only their pre-childbearing SWB treadmill level, but also those women in the first tertile of the distribution. Upper income men have a negative effect in terms of treadmill, but no difference emerges if they are compared with those in the first tertile.