Elsevier

Economics of Education Review

Volume 30, Issue 6, December 2011, Pages 1167-1176
Economics of Education Review

What determines the return to education: An extra year or a hurdle cleared?

https://doi.org/10.1016/j.econedurev.2011.05.004Get rights and content

Abstract

The 1973 Raising of the School Leaving Age from 15 to 16 has been used to identify returns to years’ schooling. However, because the first set of “high stakes” exams are taken in the UK at age 16, the reform affected the proportion with qualifications, as well as schooling length. In order to shed light on whether the returns reflect the extra length of schooling or the increase in qualifications, we exploit another institutional rule – the Easter Leaving Rule – which we argue only affected the probability of obtaining qualifications (and not the length of schooling). We find sizeable returns to academic qualifications – increasing the probability of employment by 40% points – and our results suggest that qualifications drive most of the returns to education.

Highlights

► We examine the effect on academic qualifications of two school leaving age rules. ► The 1973 RoSLA affected both qualification attainment and years of schooling. ► The “Easter Leaving Rule” affects only qualification attainment, not schooling length. ► Much of the RoSLA effect on employment comes through qualifications. ► Policy implications for raising the minimum school leaving age to 17.

Introduction

It is well known that identifying the causal effect of education on labour market outcomes is problematic given the endogeneity of schooling choice. Changes in compulsory schooling laws, which occurred in the UK in 1947 (when the school leaving age changed from 14 to 15) and 1973 (when it increased again to 16), are natural candidates for instruments and have been widely exploited (see Devereux and Hart, 2010, Grenet, 2009, Harmon and Walker, 1995, for earnings effects and Silles, 2009, Clark and Royer, 2010, for effects on health outcomes). Typically, estimates focus on the returns to the length of schooling. However, raising the school leaving age, particularly from 15 to 16, affected not only years’ education but also the probability that people left school with qualifications since in the UK the first set of “high stakes” exams that lead to nationally recognised qualifications are typically taken at age 16. The extent to which the estimated returns to raising the school leaving age reflect the benefit of an increase in length of schooling, or the returns to gaining specific qualifications is unclear. Yet this is crucially important to policy-makers. For example, the school leaving age is planned to increase again in the UK from 16 to 17 in 2013. The issue is whether this will raise employment and wages among those affected when the second set of high stakes exams is typically taken at age 18.

The aim of this paper is to shed light on what drives returns to increased education – whether there is a benefit to increased years of schooling per se or whether qualifications are key. To do this, we exploit another institutional rule – the Easter Leaving Rule (ELR) – that determined exactly when in the school year people could leave school. Rather than being allowed to leave on the day of reaching the minimum age, children faced one of two possible leaving dates – the end of the Easter term or the end of the summer term – depending on their birthday. Specifically, those born between 1st September and 31st January could leave at Easter while those born between 1st February and 31st August had to stay until the end of the summer, exam-taking term. We show that after the school leaving age rose to 16, the age at which the first set of high stakes exams is typically taken in the UK, late leavers were significantly more likely to obtain academic qualifications. We exploit this discontinuity to identify the effect of qualifications on later labour market outcomes. We then compare these estimates of the effect of qualifications using the ELR as an instrument with estimates of the effect of qualifications using the 1973 Raising of the School Leaving Age (RoSLA) as an instrument. Since the effect of the RoSLA will also include any additional effect from increasing the length of schooling, this allows us to say something about whether what matters is qualifications or years of schooling.

The plan of the paper is as follows: the next section discusses related literature on estimating returns to education using RoSLA as well as studies that have attempted to estimate the returns to qualifications directly. Section 3 discusses the institutional rules, and our empirical strategy, in more detail. Section 4 describes the data, while Section 5 presents the main regression results. Section 6 concludes.

Section snippets

Related literature

Our paper is related to two existing literatures. First, a number of papers estimate the returns to education in the UK by exploiting changes in the school leaving age. The majority employ variants of the traditional Mincer human capital earnings function in which education is measured in terms of completed years of schooling. Very few of these studies explicitly consider the extent to which the increase in qualifications matters. However, as we show below, raising the school leaving age from

Empirical strategy

In order to disentangle the effect of an additional year of schooling from the effect of credentials gained in school, we exploit a former institutional rule in England and Wales that determined exactly when individuals could leave school – the Easter Leaving Rule.

Since the Education Act of 1870 a September 1st cut-off has determined which school cohort a child belongs to in England and Wales: thus school cohort t comprises all children born between 1st September in year t and 31st August in

Data and descriptives

Our data come from the Quarterly Labour Force Survey (LFS), pooled from 1993 quarter one to 2010 quarter two inclusive. The LFS is the largest regular household survey in Great Britain and is designed to be representative of the population living in private households, with approximately 60,000 households responding each quarter. The survey is a rotating panel with each household interviewed in five successive quarters and is designed such that, in each quarter, one fifth of the households are

Effect of RoSLA and ELR on academic qualifications and labour market outcomes

Table 2 presents reduced form estimates of the effects of each leaving rule on both the probability of attaining academic qualifications (linear probability model) and on the labour market outcomes themselves: log wages in panel (a), employment (linear probability model) in panel (b). Each reported coefficient comes from a separate regression and in all regressions the full set of age, cohort, region, ethnicity and survey quarter*year controls are included.

“Late leaver” is an indicator that

Concluding remarks

In this paper, we have used one institutional rule – the Easter Leaving Rule (ELR) – to shed light on what drives the effect on employment outcomes of another institutional rule – the 1973 RoSLA from 15 to 16. The RoSLA reform had two effects on educational outcomes that could potentially impact employment: it both increased the length of time that children spent in school by up to one year and made it more likely that they would leave school with some academic qualifications since high stakes

References (16)

There are more references available in the full text version of this article.

Cited by (32)

  • Does education improve financial behaviors? Quasi-experimental evidence from Britain

    2021, Journal of Economic Behavior and Organization
    Citation Excerpt :

    The line corresponding to 16 year olds serves as a valid comparison: their amount of schooling has been unaffected by these reforms. This result is consistent with previous studies, for example, Chevalier et al. (2004) and Dickson and Smith (2011), in that the educational reforms did not stimulate an increased proportion of individuals to study beyond the compulsory school leaving age. One of the most important contributions of our paper is the use of two reforms separately to identify the treatment effect.

  • Educational expansion and the economic value of education in Vietnam: An instrument-free analysis

    2021, International Journal of Educational Research Open
    Citation Excerpt :

    This research agenda has produced mixed results that have led to a debate on the topic. Some researchers argue that the return to education decreases as education expands because education has the status of a scarce good, and reduction of scarcity reduces its premium (Van de Werfhorst, 2011a; Dickson & Smith, 2011). However, evidence from other studies suggests that the relative premium of education remains unchanged even with an expansion, due to people with low educational levels being penalized by the labor market (Bills, 2016; Smyth & McCoy, 2011).

  • From high school to the high chair: Education and fertility timing

    2019, Economics of Education Review
    Citation Excerpt :

    Second, the education expansion led to an improvement in labour market opportunities which subsequently led the affected cohorts of women to postpone having a family and child-rearing activities. The improvement in labour force participation, employment and earnings of the affected cohorts has been well documented in the previous literature (Anderberg and Zhu, 2014; Dickson and Smith, 2011; Del Bono & Galindo-Rueda, 2007). While we do not show evidence on the effect of education on the overall fertility, a paper by Fort et al. (2016), which exploits the variation in education coming from the compulsory schooling reforms in England and Continental Europe, might be informative in this respect.

  • Returning to returns: Revisiting the British education evidence

    2017, Labour Economics
    Citation Excerpt :

    Devereux and Hart (2010) do not find any significant returns to compulsory schooling for the pooled sample, whereas they find IV estimates for men that are much closer to the conventional OLS estimates. Dickson and Smith (2011) use the Quarterly Labour Force Survey (LFS) data to exploit separately the 1963 reform and the 1972 reform to investigate whether an additional year of schooling is beneficial per se or whether qualifications are key. They focus on men, and they restrict their analysis to the cohorts born between September 1947 and August 1967, i.e., ten years before and after the 1972 reform.

View all citing articles on Scopus

We would like to thank participants at the CMPO conference, The Economic Return to Education, for helpful comments on an early version of this paper. Matt Dickson acknowledges financial support from the Economic and Social Research Council. Data from the Special Licence Quarterly Labour Force Survey were made available by the Economic and Social Data Service with the permission of the Office for National Statistics. We acknowledge the assistance of ESDS and ONS with thanks. Any remaining errors are our own.

View full text