Household income, education and the transition to third birth

Juliet A. Stone, University of Southampton
Ann M. Berrington, University of Southampton

The UK continues to show relatively high levels of progression to third birth, despite increases in childlessness. Much debate as to who has large families in the UK has taken place in the absence of robust empirical evidence. We use newly available household panel data from a large nationally representative study to explore associations between couple’s income level, receipt of child tax credits, education, and the propensity to have third births. We use the UKHLS waves 1 to 5 (2009-2014), taking woman aged 20-44 as the unit of analysis. Discrete-time event history analyses are used to model factors associated with the annual probability of experiencing a conception leading to a third birth. We also control for the woman’s age, the age of her youngest child, religion, shared children, partnership history, women’s labour market attachment and housing tenure. In bivariate analyses, all of the socioeconomic variables show a negative association with third birth, such that those with the lowest levels of income or education are the most likely to have a third birth. Logistic regression models further show that the propensity for women to have a third birth is most strongly associated with her own age and the age of her youngest child. Once these demographic variables are included, large and significant associations between income/education and third birth are no longer statistically significant. We discuss our findings in relation to economic and sociological theories regarding the inter-relationship between the timing of childbearing and family size, and how this interacts with financial, cultural and social capital.