Essays on the role of cognitive ability in midlife returns to education

This thesis comprises three papers exploring the role of cognitive ability in returns to education.

Chapter 2 investigates the selection effect of childhood cognitive ability on three sequential educational decisions after compulsory education, comparing the effects of pre-school cognitive ability (at age 5) and post-compulsory school cognitive ability (at age 16). We construct a measurement model of latent abilities. A structural equation modelling approach is used together with maximum likelihood estimation. Findings show that both early cognitive abilities have positive selection effects on encouraging people to continue education, while post-compulsory school cognitive ability has a longer-lasting influence than pre-school cognitive ability.

Chapter 3 analyses the treatment effect of sequential educational decisions after compulsory education on midlife cognitive ability (at age 46), controlling for early cognitive abilities and socioeconomic factors. All latent abilities are estimated through a measurement model. The model specification is an extension of the framework presented in chapter 2 and is adopted a structural equation modelling approach. We find that completing postgraduate education has positive treatment effects on midlife cognitive ability. However, completing post-compulsory education and undergraduate education has a limited impact.

Chapter 4 explores the mediation effect of midlife cognitive ability on midlife returns to educational decisions, considering midlife earnings, physical health and mental health as outcomes. We use the structural equation modelling approach and extend the model in chapter 3 to rule out selection bias caused by early cognition. Results indicate that midlife cognitive ability mediates the effects of education on midlife physical and mental health. The impact of education on earnings is dominated by the direct impact, while the mediation impact is minimal.