Social origin and the financial feasibility of going to university: the role of wage penalties and availability of funding

ABSTRACT

The evidence on why students from lower social origin are persistently underrepresented in higher education (HE) suggests social, educational and economic factors all play a role. We concentrate on the influence of monetary costs/benefits and how these are influenced by social origin. In particular, we consider the effect of a class-based wage penalty in the labour market and, using evidence from a large-scale survey of Scottish students, we show how the greater financial constraints facing working-class students affects the incentive to participate in HE. Using a simple model of human capital investment, the low rate of working-class participation in HE is shown to be consistent with rational behaviour, i.e. weighing the monetary costs and benefits, participating in HE is a less attractive investment proposition for some students. We conduct simulations which suggest this could be mitigated by generous income-contingent support.

KEYWORDS:

• Higher education
• participation rates
• social origin
• investment
• wage penalty
• returns to education

JEL CLASSIFICATION:

• CG12
• H52
• I22
• I23
• I24
• J08

Acknowledgements

The authors would like to thank Alan McGregor and Patricia Findlay for valuable comments on earlier versions of this paper. We would also like to the thank participants in the Oxford Symposium on the Financing of Education for their feedback, as well as seminar participants at the University of Glasgow and the University of Hull.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Kristinn Hermannsson http://orcid.org/0000-0001-9957-3914

Notes

1 For instance, in the UK many secondary schools offer young people the opportunity to attend FE colleges to take vocational qualifications as part of their school timetable. A young person who had already decided, for whatever reason, against proceeding to HE, may decide to follow a vocational route and thereby attain fewer academic qualifications.

2 In summary, out of a total population of 160,000, 9181 completed the screening survey and, of that, 5314 (58%) completed the main survey which was further reduced to 4965 after the removal of outliers and part-time FE students, of whom only 36 completed the main survey. Calculating non-response rates was complicated by the fact that the precise number of students contacted in the FE colleges is not known. At the time, not all students in FE colleges were contactable by email and this was the method used to distribute the screening and main survey.

3 See https://www.ons.gov.uk/methodology/classificationsandstandards/standardoccupationalclassificationsoc

4 We do not disaggregate the analysis further, such as for different qualifications, gender, ethnicity or disability status, but the framework is sufficiently general to allow this, provided sufficiently detailed data are available. We should note that implicit in the class-wage penalty calculated in Section 5.5 is the aggregate difference between subjects and institutions attended by working-class and middle-class students respectively.

5 The implicit cost of such funding is influenced by age, race, education and social class, from 10% to the most privileged to 19% for the least privileged. However, it is difficult to map these findings onto social classes in the UK therefore for simplicity we adopt the baseline estimate for all groups.

6 For an overview of the market, see for example: https://www.moneyguru.com/compare/credit-cards/

7 This is the first wave of the LFS that includes information on social origin. Unfortunately this does no't coincide with the data year of the income/expenditure data. However, we are not using the LFS to inform levels of wages, but ratios between different groups so this should not be affected by price level changes. Similarly, it is reasonable to assume that a class wage penalty was already in effect in 2007/2008 as contemporary wage premia analysis had already identified a polarisation of graduate outcomes, e.g. Walker and Zhu (2008).