«Claire Smetherham School of Social Sciences, Cardiff University March 2005 Contact details: Claire Smetherham School of Social Sciences Cardiff ...»
Briefing Paper 3:
A Review of the Literature on Graduate
Employment, Underemployment and
Report for the Independent Study into the Devolution of
the Student Support System and Tuition Fee Regime
School of Social Sciences, Cardiff University
School of Social Sciences
King Edward VII Avenue
CF10 3WT SmetherhamC1@Cardiff.ac.uk Contents
1. Outline 5
2. A Review of Graduate Labour Demand and Over-Education 5 Measuring Over-Education 6 Research Evidence on Over-Qualification 7 Graduate ‘Glut’ 20 Evidence of a Skills Gap 20 The Present Balance Between Supply and Demand 21 The Future Balance Between Supply and Demand 22 Skills in Wales 22 Returns to HE: A Summary 23 Returns to HE in Wales 24
3. Gender Issues 24 Gender, Career Outcomes an Subject 28
4. Class Issues 30
5. Ethnic Minority Issues 31 Transitions to the Labour Market 32 Initial Destinations 33 Types of Employment 35 Satisfaction with Job 36 Employer Perspective 36 Under-Representation of Minority Ethnic Graduate Recruits 36 Over-Education 37
6. Educational Biographical Issues 39 7. ‘Definitional Work on Graduate’ Jobs and Over-Education 44 Defining Over-Education 44 Defining Graduate Jobs 45 Summary of Elias and Purcell’s New Classification 55 of Graduate Occupations
8. Conclusion 56
9. References 58
INDEX OF TABLES AND FIGURESParticipation by young people in Higher Education, Age Participation Index (API) Great Britain, 1975/76 to 2000/01 6 Pay and Over-Education
The purpose of this briefing paper is to review the existing research literature and findings on graduate employment, underemployment and unemployment.
It will include a discussion of class, gender, ethnic and racial differences wherever possible. In addition to this, it will also examine the conceptual confusion that currently exists around the idea of a ‘graduate job, ‘graduate glut’ and ‘new’ graduate occupations.
The briefing paper is structured in the following way:
A consideration of the demand for graduate skills and qualifications is crucial when considering the issue of over-education, as is a consideration of the types of jobs that different groups of graduate, enter.1 This section will review the existing research evidence on graduate over-education and the demand for graduates in the UK and Wales (where possible).
By international standards, the British economy is characterised by a limited distribution of skills throughout the workforce, which is matched by relatively low employer demand for skills (Brown, Green et al. 2001). In this context, there is considerable interest in learning more about the economy’s ability to absorb the increased supplies of graduates resulting from growth in HE participation (Mason 2002).
There has been a recent expansion in the literature documenting the extent to which graduate workers are being fully utilised in the workplace. This literature tends to find that a significant minority of workers are employed in jobs for which their current qualifications are not a requirement, that is, they are ‘over-educated’.
Policy discourse however assumes that there are plenty of jobs for graduates and that an increasing proportion of jobs within the division of labour actually require graduates – shown by the continued importance attached to achieving As discussed later in the paper, there continues to be much debate about what constitutes a ‘graduate job’, and what being ‘over-qualified’ actually means. Furthermore, the relationship between skill and income continues to pose significant problems for researchers.
a 50% participation target, for example, as well as labour market projections suggesting that 80% of newly created jobs will require graduates by 2010 (DfES 2003). Indeed, The Future of Higher Education White Paper (DfES 2003 : para.
5.1) states that:
“Our economy is becoming ever more knowledge-based… These trends demand a more highly skilled workforce. Forecasts by the Institute for Employment Research show that between 1999 and 2010, the number of jobs in higher level occupations – the ones most likely to be filled by those who have been through HE – will grow by over one and a half million.
That represents 80% of new jobs over the decade.” The following figure provides some data on the increase in graduate enrolments over time, illustrating the growth in HE over the last few decades of the twentieth century, and providing the context within which debates about graduate over-education have been located.
Participation by young people in Higher Education, Age Participation Index (API) Great Britain, 1975/76 to 2000/01 API (%) /7 /7 /7 /7 /8 /8 /8 /8 /8 /8 /8 /8 /8 /8 /9 /9 /9 /9 /9 /9 /9 /9 /9 /9 /0 /0 [Source: Department for Education and Skills, Trends in Education and Skills (www.dfes.gov.uk/trends).
Measuring over-education Over-education can be measured in one of three principal ways. The first involves systematic evaluation by labour market analysts who attempt to ascertain the level and type of education required in particular occupations.
The second measure involves a worker making an assessment of his or her own job. In these two measures the extent of over-education is determined by comparing the required level of education with that actually attained. A third measure is used where there is no direct question on required education for the job. Here, it is customary to calculate the mean and/or modal levels of education, and measure over-education as a divergence from the mean or modal (Chevalier 2001).
Research evidence on over-qualification A detailed consideration of the demand for particular types of graduate labour has led several commentators to suggest that one consequence of the expansion of HE, along with changes in the occupational structure of the labour market, has lead to a situation where many (usually a significant minority of) university graduates will be unable to find a demand for their skills and knowledge within the labour market. It is suggested they will be and unable to find the kinds of job which they were expecting an HE to give them access to (Dolton and Vignoles 2000; Dolton and Silles 2001; Brynin 2002b;
Brynin 2002a; Keep and Mayhew 2004), particularly when competing in the more ‘elite’ end of the graduate labour market i.e; for professional and managerial employment with leading blue-chip companies (Brown and Hesketh 2004). A consideration of the types of jobs undertaken by university graduates, and how these have changed over time, has in turn led to continuing debate about graduate over-qualification or under-employment (Dolton and Vignoles 2000; Dolton and Silles 2001; Brynin 2002b; Brynin 2002a).
For many graduates, their initial jobs are of a temporary nature (some in jobs where their degree is not particularly relevant) to help get some immediate income to live reasonably well or pay off debts, or to get some work experience (Connor, Tyers et al. 2004). It is the nature of both the first and subsequent actual jobs, and the suitability or appropriateness of the work, that graduates undertake that is a key issue. The notion of employment success
for graduates today can be complex, bringing in aspects of:
· Job level (is it a graduate job?) · Salary (is it a well-paid job?) · Satisfaction and expectations (is it what they were seeking?) · And development and career prospects (where is it leading?).
There has been a growing economics literature on the effects of educational inflation and over-qualification (e.g. Borghans and De Grip 2000; Hartog 2000), some specifically relating to degrees (e.g. Dolton and Vignoles 2000).
Dolton and Vignoles (2000) define over-education, in the context of the UK graduate labour market, as a graduate in a job requiring sub-degree qualifications (or no qualifications at all). The data they (Dolton and Vignoles
2000) present indicates that a significant proportion of British graduates were over-educated in the 1980s: 38% of all graduates surveyed were overeducated in their first job. However this, as with much of the other data on this issue, is based on data from the 1980s and is therefore somewhat out of date.
For the UK, Sloane et al (1999) find that around 31% of British workers are ‘over-educated’ (whereby their educational attainment exceeds that required to get their job). According to Dolton and Silles’ (2001) measure of overeducation, 42% of graduates held first jobs for which a degree was not required to do the work.
In the UK, recent evidence suggests that between 29% and 47% of the work force is over-educated (Green, McIntosh et al. 1999). Sloane et al (1999) report no evidence that for over-educated workers, the quality of the match improves with a change of employer. On the contrary, over-educated workers tend to be more prone to dismissal and have more unemployment spells.
Chevalier (2001) argues that previous studies on over-education implicitly assumed homogeneity of works, and thus overestimated the true extent of the phenomenon; some graduates are not in fact endowed with the skills required to obtain a graduate job. Most of the literature has ignored the issue of educational heterogeneity by defining over-education as departing from a norm and assuming the homogeneity in skills of all workers with an identical qualification, therefore overestimating the extent of over-education. Chevalier (2001) therefore divides over-educated, workers by their skill level, relying on a proxy variable concerning the satisfaction on the match between education and the job. He assumes that over-educated workers who are satisfied with this match are only apparently over-educated while those who are dissatisfied are genuinely over-educated. This idea is explained in more detail in Chevalier (2001).
Grouping over-educated workers together, Chevalier (2001) estimates a 14% pay penalty, which is consistent with previous estimates for UK graduates (Battu, Belfield et al 1999; Dolton and Vignoles 2000). He argues that the over-education of graduates in the UK appears to be mostly ‘mis-education’;
some graduates do not possess the skills required to obtain a graduate job.
However, the data that Chevalier uses is based on Belfield et al’s (1997) data, which rely on a relatively small sample of graduates from the University of Birmingham. Furthermore, the data on which he bases his analyses stem from two cohorts of graduates from the mid, 1980s and 1990s? and is therefore somewhat out-of-date – a problem with many published studies on over-qualification.
On average, graduates in jobs without a formal requirement were found to still earn a premium relative to other people doing the same type of job – although less than non-overeducated graduates.
Dissatisfaction with the match between qualifications and job amongst employed graduates is much lower (and expressed by around 10% of graduates) (Chevalier 2001):
Pay and Over-Education
[Source: Chevalier (2001)] However, is over-qualification simply a temporary phenomenon that declines later on in a career? As well as considering the types of jobs undertaken by different groups of graduates on entry to the labour market, it is also important to consider if and how job quality changes over time for different groups of graduates who are in work.
Some research evidence suggests that over-qualification is not simply a temporary phenomenon that declines later on in a career and with labour market ‘maturity’, as policy discourses suggest. Dolton and Vignoles (2000), for example, found that a large proportion of graduates in their sample (30%) remained over-educated six years after graduation. In fact,15% of their sample required no qualifications at all in their first jobs and even six years on, this figure was only down to 11%. Dolton and Vignoles’ evidence is an examination of the 1980 National Survey of Graduates and Diplomates, a one in six random sample postal survey of UK graduates undertaken six years after graduation. Over-education, they argue, appears to be less likely if an individual has more human capital (e.g. a better grade of degree), does not work in public administration, and has an occupationally relevant degree (e.g.
engineering). Again, this evidence is now somewhat out-of-date.
Similar analysis of the survey data obtained by Smetherham (2005) leads to
comparable conclusions. For example:
Whether a Degree was a Formal Entry Requirement for Respondent’s Current Job by Degree Classification (1997 cohort)
After five years in the labour market, looking at the 1997 cohort, those with First Class degrees in Smetherham’s (2005) sample were significantly more likely to be in a job for which a degree was a formal entry requirement, although even a large minority of those with Firsts (24%) were in jobs for which a degree was not a formal entry requirement. Forty three % of Smetherham’s sample with 2.2 degrees, by contrast, were in jobs for which a degree was not a formal entry requirement - even after five years in the labour market.
Whether a Degree was a Formal Entry Requirement for Respondent’s Current Job by Area of Degree Study (1997 cohort; 1sts)
[Source: Smetherham (2005)] Considering the responses of her 1997 cohort with Firsts, Smetherham (2005) shows that although after five years in the labour market, the majority of graduates with Firsts were in jobs for which a degree was a formal entry requirement, well over a quarter (36%) of those from arts based disciplines with Firsts were still in a job for which a degree was not a formal entry requirement. Similarly, 30% of those from social science disciplines were in such a job, compared to just over a fifth (21%) of those from natural science courses, 19% of those from vocational courses and 14% of those from maths/computing disciplines.
Green et al. (2000) argue that an estimated 30% of UK graduates have more education than their job requires six years after graduating. Sloane et al.