<html><head /> <body> <META http-equiv="Content-Type" content="text/html; charset=UTF-16"><title>Overeducation: A Tough Nut to Crack</title><meta name="keywords" content="economy,overeducation,skill,anna,culture,degree,earnings,economics,education,francis,green,jobs,literacy,mcintosh,ncds,numeracy,overskilled,qualifications,requirements,steven,survey,undereducation,vignoles,knowledge,"><table style="font-family:Verdana; font-size:larger; " align="center" border="0" width="50%"><tbody><tr> <td style="background-color:silver; border-color:white; border-left-style:none; border-style:none; " width="730"><span style="font-family:Verdana; font-size:larger; ">Overeducation: A Tough Nut to Crack</span></td> </tr> <tr> </tr> </tbody></table><br><table style="font-family:Verdana; font-size:medium; " align="center" bgcolor="white" border="0" width="50%"><tbody><tr> <td height="131" width="669"><span style="font-family:Verdana; font-size:x-small;"><strong>Editors Introduction</strong></span><span style="font-family:Verdana; font-size:x-small; "><IMG SRC="auth_vignoles.jpg" WIDTH="90" HEIGHT="104" ALT="vignoles" VSPACE="0" HSPACE="0" BORDER="0" ALIGN="right"> For all the talk about the knowledge economy, it appears as if we might be an overeducated society. There is evidence that many highly qualified people are less able, on average, than those more appropriately educated for their job. The overeducated have jobs appropriate for lower skill levels. Francis Green, Steven McIntosh and Anna Vignoles, members of the Centre for Economic Performance at the London School of Economics and Political Science, investigate the proposition that "overeducation" is a problem for Britain.</span><br> <br> <span style="font-family:Verdana; font-size:x-small; ">It's a bizarre concept. At a time when there is almost universal agreement on the importance of education, both for individual well-being and for national economic prosperity, how on earth can we think of people as overeducated? To compete successfully in the global economy nations must provide high quality goods and services, produced by a highly-skilled workforce. To survive in today's knowledge-based society, an individual must be well-educated, and capable of continually updating his or her skills in a process of lifelong learning. For more than a decade, the complaint in Britain has been of insufficient investment in education and training. So how could anyone argue we are investing too much? Of course they're not--or at least not in the way you might think. But there is an argument for saying that "overeducation" is a serious problem in the UK, and that this phenomenon should lead to a reassessment of the way resources are used for education and training.</span><br><br><span style="font-size:x-small;"><strong>Is overeducation a real problem?</strong></span><br><span style="font-family:Verdana; font-size:x-small; ">As most people know, there's been a rapid and sharp increase in the provision of higher education in Britain. <br> <A HREF="1948_table1.htm" TARGET="_blank"> Table 1 </A> shows that in 1997 3 percent of the working-age population had a higher degree, more than double the proportion 12 years earlier; over the same period the proportion of people with a first degree went up by almost half. Yet there has also been an increase in the number of people who are overeducated, particularly in the 1970s and 1980s.</span><br><br><span style="font-family:Verdana; font-size:x-small; ">Overeducation means exactly what it says--people with more educational qualifications than they need to do their job--such as estate agents with Ph.D.s or secretaries with degrees. The estate agent with a doctoral thesis will be no better at being an estate agent than someone with a degree; the graduate secretary will not need any of the skills acquired on a degree course to do the job properly.</span><br><br><span style="font-family:Verdana; font-size:x-small; ">We have looked at evidence from a number of different sources on the extent of overeducation in the UK. Both the 1986 Social Change and Economic Life Initiative (SCELI) and the 1997 Skills Survey ask identical questions of reasonably representative samples of the UK population. In particular, the surveys ask respondents to report which qualifications are necessary for their jobs. Comparing their answers with the actual qualifications that they hold gives a good guide to the extent of overeducation (and undereducation).</span><br><br><span style="font-family:Verdana; font-size:x-small; ">The results of this exercise are reported in <A HREF="1948_table2.htm" TARGET="_blank"> Table 2 </A>. They show that a fairly consistent proportion of the population, around 30 percent, appear to have been overeducated in both 1986 and 1997. This figure is similar to estimates for several other countries, suggesting that this is not a uniquely British phenomenon. Earlier work in the UK suggested a rise in overeducation in the 1970s and 1980s: the figures in Table 2 suggest that this trend may have levelled off in the 1990s. (It's worth noting that Table 2 also shows quite widespread undereducation, with 20 percent of the population having a qualification lower than the one recommended as necessary for their jobs.)</span><br><br><span style="font-size:x-small;"><strong>The evidence mounts</strong></span><br><span style="font-family:Verdana; font-size:x-small; ">We also examined the results of two other surveys, both targeted at particular groups of people and both of which included questions about the educational requirements of respondents' jobs. A 1998 survey of Newcastle University graduates found that, amongst first degree-holders, the incidence of overeducation was 26 percent. The National Child Development Study (NCDS), which has followed a cohort of children born in a particular week in 1958 throughout their lives, was also helpful. The NCDS questionnaire sent out in 1995, when the participants were 37, contained a question about educational requirements of jobs: the answers showed that 47 percent of the sample were apparently overeducated. This comparatively high figure can be explained by the large proportion of workers with very low qualifications reporting that their jobs do not require any particular qualifications at all. For example, workers with just one Certificate of Secondary Education (CSE) pass but who are in jobs requiring no qualifications would be counted as overeducated.</span><br><br><span style="font-size:x-small;"><strong>Overeducation or qualifications inflation?</strong></span><br><span style="font-family:Verdana; font-size:x-small; ">Overeducation isn't the same as qualifications inflation. The latter reflects the tendency of employers to set higher and higher qualification requirements for their employees. They do this because they need some way of identifying high ability candidates. As more people acquire A-levels and degrees as a result of the expansion of higher education, would-be employers narrow the field of suitable candidates by increasing the entry requirements. Even if the nature of the job is totally unchanged, where previously employers might have asked for an A-level as a minimum requirement, they might now ask for a degree.</span><br><br><span style="font-family:Verdana; font-size:x-small; ">The Newcastle University survey was particularly helpful in making the distinction between these two phenomena. The Newcastle alumni were asked for the qualification required to get their job, as well as what qualifications were necessary actually to do it. Qualifications inflation would show systematic differences in the answers given to the two questions. Yet the majority of individuals (76 percent) gave the same response to both. And while 10 percent of the remainder said entry requirements were higher than needed to do the job, 14 percent said they were lower. The 1997 Skills Survey showed similar results: 78 percent of respondents who said they needed a degree to get their job also said that the degree was either "fairly necessary" or "essential" for actually doing the job.</span><br><br><span style="font-family:Verdana; font-size:x-small; ">Not much evidence of qualifications inflation, then, despite the rapid increase in the supply of qualifications. This is understandable: there is, after all, considerable evidence to suggest that jobs are becoming more demanding and more skill-intensive. Employers may be increasing the educational requirements of jobs because of this. The results of the 1997 Skills Survey, when compared with those from the 1986 SCELI, suggest that workers are spending more time training for their job and taking longer to master it. The Skills Survey also suggests that the use of computers increased between 1992 and 1997, and that they were used in a more complex way; it also shows a greater importance being attached to communication, social and problem-solving skills. All of this evidence is thus consistent with the view that job skill demands are genuinely rising.</span><br><br><span style="font-size:x-small;"><strong>What is going on?</strong></span><br><span style="font-family:Verdana; font-size:x-small; ">So there's a paradox: jobs are getting more complex but overeducation is still occurring. One obvious question therefore is: does overeducation matter? Perhaps individuals accept jobs that aren't commensurate with their education and skills in the knowledge--or hope--that once they have some experience they will progress to higher level jobs within the organisation they work for. If so, the evidence suggests these hopes are often dashed. Data from a sample of 1980s male graduates reveal that the majority of those who were overeducated in their first job after graduation had still not moved into a graduate-level job six years later, implying that for many overeducation is not a stepping stone to better things.</span><br><br><span style="font-family:Verdana; font-size:x-small; ">Does any of this matter? From the evidence we have, it certainly does for the individuals involved. The overeducated seem to earn significantly less than their similarly educated peers who have found an appropriate job for their skills. The 1997 Skills Survey clearly illustrates this. We classified each individual's actual qualifications to one of five levels, and then the level of education required for their job to one of the same five levels. We then defined the extent of overeducation (on a scale from 0 to 5) as the actual minus the required education level if the former is greater, and zero otherwise. (Similarly, the extent of undereducation can be defined as the required minus the actual education level; again if the former is greater, and zero otherwise.) We then used this information to investigate the effect of being overeducated on earnings.</span><br><br><span style="font-family:Verdana; font-size:x-small; "><A HREF="1948_table3.htm" TARGET="_blank"> Table 3</A> shows the results, given as the percentage change in an individual's hourly wages for each point change in the level of over- and undereducation. The figures are striking: the greater the extent of a person's overeducation, the lower their relative earnings, by a significant amount. Thus there is a 12 percent reduction in women's wages for each point increase in the level of overeducation. For example, if a woman holds a degree, but is performing a job requiring only a sub-degree level qualification, she will earn 12 percent less than a similarly qualified woman working in an appropriate job. For men, the reduction in wages is smaller, but it is still 5 percent for each level of overeducation. These results are consistent with previous research that has also found that there are substantial real wage penalties associated with being overeducated.</span><br><br><span style="font-size:x-small;"><strong>Why does overeducation occur?</strong></span><br><span style="font-family:Verdana; font-size:x-small; ">It's hard to see why anyone would do a job for which they are overeducated. One explanation is, of course, simple bad luck. In an imperfect world people may have to take a job for which they are overqualified; and the financial (and emotional) costs of moving jobs might then make it difficult to change. Another, perhaps more convincing, possibility is that not all individuals with the same level of education are equally productive in the workplace. Perhaps the overeducated are less able in some way than those with the same education level working in a job which matches their qualifications.</span><br><br><span style="font-family:Verdana; font-size:x-small; ">This is a hypothesis we can test, using data from the NCDS. Besides giving us information on respondents' education levels, this survey also provides the results of various ability tests (mathematics and reading skills at age 16, and literacy and numeracy at age 37) which were given to some participants. We examined these test scores to see whether they could explain why some individuals are overeducated and others not. The results in <A HREF="1948_table4.htm" TARGET="_blank"> Table 4</A> underline the importance of good numeracy skills in particular in reducing the likelihood of an individual ending up in a job for which they are overeducated. Each point gained on the numeracy test, for example, cuts the chance of being overeducated by almost two percentage points.</span><br><br><span style="font-size:x-small;"><strong>More educated, less able?</strong></span><br><span style="font-family:Verdana; font-size:x-small; ">So it does seem as if the overeducated are less able, on average, than those appropriately educated for their job, with a lack of numeracy skills being of particular importance. This might give us a clue: perhaps the overeducated have actually got jobs which are appropriate to their lower skills levels. If we could measure job requirements in terms of skills rather than qualifications would we then see that most individuals have appropriate jobs--implying that the labour market is successfully matching the demand and supply for skills?</span><br><br><span style="font-family:Verdana; font-size:x-small; ">Wrong again. That isn't what we found when we tested this hypothesis using the International Adult Literacy Survey (IALS). This survey measures each person's skills, using three tests focusing on literacy and numeracy. We first took each individual's average score across the three tests and allocated them to one of four skill bands. We then used responses to the survey's questions about how often participants needed to use certain reading, writing and arithmetic skills at work. We used the responses to these questions to measure the skills required on the job, and again allocated the results to one of four required skill bands. We then arbitrarily defined an individual as "overskilled" if their skill level was two or more levels higher than their job requirement, and similarly "underskilled" if their skill level was two or more levels below that required by their job. A significant number of participants fell into one of these two categories, particularly the former.</span><br><br><span style="font-family:Verdana; font-size:x-small; ">Even in spite of our rather arbitrary classification of actual and required skills to the four bands, the extent of over- and under-skilling can be shown to have real effects on individuals' earnings, similar to that found amongst the over- and undereducated. We found that, even taking into account the gender, age, education level and full- or part-time status of workers, individuals with good skills who work in jobs which only require low skills earn much less than their peers who find jobs that do match their skill level. For example, a worker with very high skills (level 4) working in a very low level job (level 1) earns &#1637,000 a year less than a similarly skilled individual working in a level 4 job. Not using one's skills can therefore have very real effects on one's earnings.</span><br><br><span style="font-family:Verdana; font-size:x-small; ">The trail hasn't gone entirely cold, however: the IALS does offer some clues about the causes of overskilling (and, in consequence, overeducation). As we noted, the actual skills variable is a composite measure, based on numeracy and literacy tests. We've already established that the overeducated are more likely to have poor numeracy skills. It turns out that the overskilled are also more likely to be deficient in this respect. This finding is reinforced by information elicited from those participants who have had some education or training in the 12 months prior to the survey. <A HREF="1948_table5.htm" TARGET="_blank"> Table 5 </A> shows the percentage of participants who studied in a particular subject, divided into those who are classed as overskilled and those who aren't. The figures clearly show that the overskilled are more likely to have studied subjects like arts and humanities; and correspondingly less likely to have studied subjects with some quantitative element, such as engineering, science, business management and so on. This suggests that at least in some non-quantitative courses, the skills being acquired aren't those demanded by the labour market.</span><br><br><span style="font-size:x-small;"><strong>Too much education?</strong></span><br><span style="font-family:Verdana; font-size:x-small; ">Overeducation and overskilling are real phenomena in the British labour market. They have real--and financially uncomfortable--consequences for the individuals affected. That doesn't mean there are now too many graduates in the UK: indeed, the figures suggest that graduate pay has remained stable or even increased relative to the wages of unqualified workers over the last decade or so; and it's not clear that overeducation has increased in recent years. Nevertheless, the proportion of employees who are overeducated remains large and should be addressed.</span><br><br><span style="font-family:Verdana; font-size:x-small; ">Rather than reducing the number of graduates, one solution might be to pay more attention to the mix of subjects being studied. There has been an increase in the demand for skilled labour, particularly related to technology and computerisation. It should in theory be possible to match this increase in demand to the growing supply of skilled labour: yet the evidence suggests that this is not happening, at least in part because those who find themselves overeducated for the jobs they do lack the specific skills--especially quantitative skills--to enable them to move to other, better paid jobs.</span><br><br><span style="font-family:Verdana; font-size:x-small; ">A word of warning though. There is a limit to which economics can address subjects of this nature. We cannot offer conclusive evidence which enables us to settle arguments about the nature and extent of education. But we can offer a cautionary note, to ensure that the content of education is given due weight at a time of rapid expansion of its provision.</span><br><br> <span style="font-family:Verdana; font-size:x-small; font-style:italic; ">This article was taken from CentrePiece magazine, published by the Centre for Economic Performance at the London School of Economics and Political Science. </span></td> </tr> </tbody></table> </body></html>