Tertiary Education Systems and Labour Markets

Report prepared for the OECD

Stephen Machin* and Sandra McNally**

1 December 2006

*Centre for Economic Performance, LSE; Department of Economics, UCL

**Centre for Economic Performance, LSE

IssuesIn the context of rapid expansion of tertiary

education:• Is there now ‘over-supply’ of graduates?• Is there evidence of ‘over-qualification’ and skill

mismatch?• Are students studying the ‘right type’ of subjects at

tertiary level?• In particular, is there a shortage of science and

technology graduates?• Does the type of institution matter for labour

market prospects?

ContextRapid expansion of tertiary education – general

phenomenon across OECD countries. Evident for several decades

Much variation across countries

Differences in the type of tertiary education. For example, The EU lags behind the US - mainly attributable to lack of supply of general and advanced research programmes (Wasmer et al. 2006)

Percentage of population (aged 25-64) that has attained tertiary education - changes over time

0

10

20

30

40

50

60

1994

2003

Labour Market Consequences of Increasing Supply

Depends on the demand for graduates by employers as well as the supply of graduates from tertiary institutions.

Increase in supply of graduates leads to fall in the wage (everything else constant);

Increase in demand for graduates leads to a rise in the wage (everything else constant).

Labour Market Consequences of Increasing Supply (2)

What is the outcome of changes in demand and supply?

Compare the wage of tertiary graduates to the wage of closest substitutes

(upper secondary or post-secondary non-tertiary graduates). Look at changes over time.

Has the differential fallen over time on account of the increasing supply of tertiary graduates?

Wage Differentials: Tertiary versus Upper Secondary or Post Secondary ( = 100) For most countries here, relative wage in 1997 and/or 2002/03

100

110

120

130

140

150

160

170

180

190

Austra

lia

Belgium

Czech

Rep

ublic

Finlan

d

Franc

e

Germ

any

Korea

New Z

ealan

d

Norway

Portu

gal

Spain

Sweden

Switzer

land

The N

ethe

rland

s

Unite

d King

dom

Unite

d Sta

tes

19972003

Source: OECD Education at a Glance

Evidence from the literature • Method: regression analysis where ‘human

capital’ is one of several factors that determine an individual’s wage

(Mincerian wage regressions).

• In general, the wage premium to tertiary education has increased or remained stable, despite expansion of tertiary education.

Role of demand

Demand has been increasing faster than supply – the only way to rationalise stable or increasing wage premium to tertiary education

Why has demand been increasing so fast?

Weight of evidence is behind ‘skill biased technology change’ explanation: introduction of new technologies that are biased in favour of skilled workers.

What has happened to employment and unemployment?

• Compare employment and unemployment rates of graduates of tertiary education and those who achieved upper secondary or post-secondary non-tertiary education; 21 countries 1991-2003

• Tertiary graduates have much higher probability of being in employment; lower probability of being unemployed

• Some evidence of ‘catch-up’ of less educated group over time, most clearly for employment rates (some exceptions – e.g. Poland).

Evidence of ‘over-supply’?• No. Plenty of scope for tertiary education to

keep on expanding.• As more people obtain a tertiary education,

greater variation in the earnings of graduates.• Explanations:

- Variation in personal characteristics of

graduates;

- Graduates studying wider range of subjects and attending larger number of institutions.

Over-education and skill mismatch?• Can take a long time for (usually less well

performing) graduates to find a job; some are not in jobs that appear to be well matched to qualifications; shortages in certain sectors are reported

• Literature on ‘over-education’ and ‘under-education’ (terms are sometimes misused)

• Workers who are ‘over’/’under’ educated might still be well matched to jobs.

• Apparent ‘over’/ ‘under’ education might be a temporary phenomenon.

Over-education and skill mismatch? (2)

• Wasmer et al. (2006) look at these issues for several European countries. Find some evidence that ‘over-education’ is a transitory phenomenon.

• Only small wage penalty associated with ‘over-qualification’. However, ‘skill mismatch’ is a more serious issue.

• In some countries (like UK) major problems with vocational system of qualifications.

What type of education? • Trade-off between highly specialised education

and more general education programmes.• Some evidence to suggest that the latter

makes workers more adaptable to economic shocks. This affects their wages and ultimately economic growth.

• Raises a question about whole education system; not just tertiary education.

• Is early stratification into a general (academic) route and a vocational route damaging to young people and ultimately to economies?

Who provides education? • Balance between employer provided training and

publicly provided training.• General skills Vs firm specific skills. • Bassanini et al (2006) evaluate what is known about

work-based training in Europe. Many questions of policy relevance cannot yet be answered (e.g. Is there enough employer based training? Is it effective in raising productivity?)

• Governments have an important role to play in improving information about training opportunities; setting appropriate legal frameworks; ensuring portability of skills.

Are graduates studying the ‘right’ subjects?

• Few academic studies estimate returns to higher education by subject of degree, especially if we want to compare countries and consider changes over time.

• Machin and Puhani (2006) estimate returns by degree subject in Britain, France, Germany and the US using a consistent framework

Estimated Return to Degree Subject – Men

Britain

France Germany United States

1993 2000 1993 2000 1993 2000 1993 2003 Arts

-- -- -- -- -- -- -- --

Science/Engineering /Technology

0.18 0.25 0.31 0.20 0.19 0.25 0.34 0.35

Social Science

0.12 0.21 0.37 0.18 0.15 0.20 0.30 0.34

Rest/Combined (incl. Medicine, education)

0.17 0.17 0.40 0.24 0.18 0.20 0.17 0.16

Source: Machin and Puhani (2006)

Estimated Return to Degree Subject – Women

Britain

France Germany United States

1993 2000 1993 2000 1993 2000 1993 2003 Arts

-- -- -- -- -- -- -- --

Science/Engineering /Technology

0.02 0.16 0.21 0.12 0.09 0.09 0.21 0.18

Social Science

0.02 0.10 0.22 0.08 0.07 0.05 0.17 0.14

Rest/Combined (incl. Medicine, education)

0.08 0.18 0.20 0.14 0.14 0.07 0.06 0.02

Source: Machin and Puhani (2006)

Returns by field of study

• In the four countries considered, returns to a university degree are lowest for Arts subjects whereas they are higher for other subjects – often highest (at least for men) in Science/Engineering/Technology

• Requires much further research for other countries – especially for making comparisons over time and across countries.

Shortage of Science and Technology Graduates?

• High wage return; reports of shortages: e.g. Belgium, Australia, New Zealand, UK.

• Variation between countries in the proportion of S&T graduates

Science and Engineering degrees as a % of all first university degrees, 2000

Asia 32%

European Union 28%

EFTA 19%

Central/Eastern Europe

29%

North America 18%

South America 22%

Oceania 22%

Proportion of first university degrees classified as 'Science and Technology'

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1970 1975 1980 1985 1990 1995 2000 2005

China

Japan

S.Korea

US

UK

Science and Technology researchers per 1000 labour force (1999)

0

2

4

6

8

10

12

Shortage of Science and Technology Graduates?

• Countries differ in relative numbers graduating with an S&T degree; working in the labour market – e.g. US and Europe

• Research on perceived ‘shortage’ in the US and Europe

• Explanation related to international mobility of highly skilled professionals, scientists and engineers.

• Concern about ‘brain drain’ and implications for economic growth

How does type of institution matter?• Countries differ along a number of dimensions: unitary

vs binary systems; importance of private sector; extent of decentralisation; extent of change over time.

• Graduate outcomes are often correlated with institutional characteristics

• Difficult to separate the effect of institutional type from the fact that students with very different characteristics may choose to attend different types of institution

How does type of institution matter? (2)• Has higher numbers going to institutes of

tertiary education led to a decline in the quality of tertiary education? Are new institutes providing as good an education as longer established institutes?

• One would expect wage returns to vary by quality of institute attended – to the extent this is perceived by employers

• Little empirical evidence outside the US

US research on institutional quality

• Difficult issues: how to measure quality; how to take account of the other attributes of college entrants (e.g. ability).

• In general, evidence for a positive effect of measures of ‘college quality’ on the subsequent wages of graduates

Conclusions and implications(1) Further expansion

• Despite very rapid expansion of tertiary education, no evidence of ‘over-supply’: the average wage gap between graduates and non-graduates remains high. Furthermore, it has been stable/increasing in most countries in recent years.

• High positive wage returns to tertiary education and the positive relationship between tertiary education and economic growth are good reasons for further expansion.

Further expansion

What barriers are there to tertiary education and what should be the policy response?

• Capacity constraints?: provide more places. • Credit constraints?: student bursaries

(especially for those from poor socio-economic backgrounds)

• Is sufficient information available to potential students?

Conclusions and implications(2) Field of Study

• Much more evidence needed, especially for comparisons over time and between countries.

• Available evidence suggests great variation in returns to tertiary education conditional on field of study.

• Policy response: provision of good information to potential students; Argument for permitting fees to vary by subject of degree and/or to provide bursaries which are differentiated by subject area

Conclusions and implications(3) Skill shortage and ‘mismatch’

• Problems with graduates not always having the skills required by employers

• Policy response: - examine the content and accreditation system of vocational courses. Is it appropriate?- what is the balance between employer-provided training and that which is publicly provided?- In the public system, what is the balance between general education and vocational education?

Conclusions and implications

• Well founded concerns about international mobility among science and technology graduates and potential implications for R&D and productivity. How can conditions of employment be made better?

• Quality of tertiary education institutions: little good evidence outside the US for how this affects labour market outcomes; Priority for data collection and analysis in other countries.