Does education increase the productivity of workers – as in Gary Becker’s Nobel Prize-winning
work on ‘human capital’ theory? Or is it simply that more productive workers invest more in
their education in the first place – as suggested by Michael Spence’s Nobel Prize-winning work
on ‘signalling’ theory?
New research by Dr Arnaud Chevalier and colleagues looks at how raising the school leaving
age in 1972 affected a large nationally representative selection of individuals in England and
Wales. The results, published in the November 2004 Economic Journal, show that the change
only raised the educational achievements of early school leavers – which is strong evidence
for the human capital story that education does increase productivity.
This question of whether human capital or signalling theory is the closest representation of
reality is of more than academic interest. Human capital theory implies that education can be
used as a policy instrument – for example, subsidising tuition costs will raise education and
hence productivity and earnings. If education only signals workers’ ability, then any subsidy to
education has no return.
Like many important questions, the answer is elusive – both theories imply a positive
relationship between education and earnings, so it’s hard to discriminate between them. This
study tries to exploit the different motivations people might have for investing in education: to
increase their absolute productivity in the case of human capital theory; or to increase their
relative ability in the case of signalling theory.
In 1972, the minimum school leaving age was raised from 15 to 16 in England and Wales.
According to signalling theory, this should have raised education at all levels as all individuals
increase their educational investment to maintain their relative positions. According to human
capital theory, the reform will only increase the education level of people directly affected by it,
that is, early school leavers.
The research uses a large nationally representative dataset of individuals from England and
Wales to test this hypothesis. The authors report tests that show no statistically significant
change at higher levels of education. They conclude that education actually does raise
productivity.
The test consists of predicting the highest qualification attained for cohorts of individuals in
school around the time of the reforms. There is a sharp change in the proportion of pupils
gaining CSE and, for women only, O-levels, after the reform.
But there are no significant variations in the achievement of higher level qualifications – in
other words, the proportion of individuals obtaining A-levels or university degree is the same
before and after the reform.
ENDS
Notes for editors: ‘Does Education Raise Productivity, or Just Reflect it?’ by Arnaud
Chevalier, Colm Harmon, Ian Walker and Yu Zhu is published in the November 2004 issue of
the Economic Journal.
Chevalier and Zhu are at the University of Kent, Canterbury; Harmon is at University College
Dublin; Walker is at the University of Warwick and the Institute of Fiscal Studies (IFS).
For further information: contact Dr Arnaud Chevalier on 01227-827412 (email:
a.chevalier@kent.ac.uk); or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or
07768-661095 (email: romesh@compuserve.com).