Sorting into tertiary education: lessons from the UK

In the UK, the number of students who graduated from a university increased considerably between 1950 and 1970. Only about 3% of a cohort went to university at the beginning of this period while about 8% did so in 1970 and about 19% in 1990. With few exceptions, only rich children had the possibility to attend college in 1950, independently of their skills. What we do not know is whether the expansion of enrollment after 1950 succeed in giving poor but smart children the opportunity to access a university? The objective of this research project is to answer this question.

The project in depth

The aim of this project is to determine whether expanding the higher education system allows a policy-maker to uncover and activate hidden talent, in particular talent from those with disadvantaged backgrounds that typically do not consider a university education. We will leverage large household surveys, containing detailed indicators of cognitive and non-cognitive skills as well as genetic information, to answer this question. The use of these data sources will allow for a clean measurement of talent as expressed by individual intelligence and personality, while the focus on a large expansion of tertiary education will assist in achieving identification of the effects of interest. Economic theory will lend structure to our investigation. The context we will consider is the under-studied massive increase of admissions to British universities following the 1963 Robbins Report. The extent to which this expansion allowed talented high-school students to obtain a university degree or just lowered the cognitive requirement to be admitted into a university is still an unanswered question.

In the UK, in 1950, overall participation in higher education was just 3.4% of young people. The Robbins Report argued that the remaining 96.6% contained “a large pool of untapped talent” that would have benefited from university education (with beneficial consequences for the British economy at large), and recommended to expand enrolment into tertiary education. The government embraced these conclusions and participation increased to 8.4% in 1970 and to 19.3% in 1990. The extent to which this expansion allowed talented high-school students to obtain a university degree or just lowered the cognitive requirement to be admitted into a university is still an unanswered question. Modelling the distribution of skills under self-selection, we aim at investigating whether the emergence of a “mass university” following a more liberal access policy has favored or hindered the provision of opportunities to talented students independently of their socioeconomic background.

We will build a model of sorting into tertiary education. An efficient sorting process matches students with certain characteristics (e.g., intelligence) with universities with certain characteristics (e.g., teaching effectiveness or quality more broadly defined) and vice versa, based on the respective preferences of students and universities over the characteristics of the counterpart. In this process, some students are selected, and some are excluded, some schools are popular, and some are not. The model will help understand how easier access to tertiary education altered this sorting process.

By resorting to a Genome-Wide Association Study to compute Polygenic Scores of intelligence and other relevant phenotypes (including educational attainment itself), we will analyze this question empirically to understand whether the emergence of a “mass university” has favored or hindered the provision of opportunities to talented students (as revealed by their genetic profile) independently of their socioeconomic background. Ancillary, administrative data from the UK USR (Universities Statistical Records) will be employed to measure precisely the variations in the supply and demand of tertiary education for the universe of enrolled students and universities in Britain.