The envious may see those with degrees getting better seats in the job market 'house' at their expense, but investment in higher education has benefits for the country's economic prospects too, argues Norman Gemmell

There are two views on higher education: "nice 'non-work' if you can get it, paid for by those who can't," or "a productive investment for the country benefiting the educated and uneducated alike". Which view people prefer is often based more on prejudice or personal experience than on a careful weighing up of the evidence.

Of course there is plenty of room on the spectrum between these two "tabloid" extremes but many people seem willing to plump for one view or the other. Not surprisingly, those who have enjoyed higher education are more likely to sign up for the latter view, while early education drop-outs are more likely to go for the former.

So is there any more objective way to persuade the educational have-nots that encouraging others to acquire more education - to the extent of subsidising them - really is in every one's interest?

³ÉÈËVRÊÓƵ

The arguments are, at least, fairly straightforward. The "nice work" view essentially sees education as an allocation system which gives those with a degree a "ticket" for one of the better seats in the labour market "house". It does not make the educated any better at doing the, usually higher paid, job which their degree has bought them, and it certainly does not make the less educated better off.

In effect, this view treats the total income in the economy as a pie of fixed size which can be cut into different sized slices - in broad terms the more educated you are, the bigger the slice you get! The higher incomes earned by the educated may of course reward greater intellectual ability or greater effort but, if so, surely there are cheaper ways of signalling this ability difference to prospective employers. In any case, if the education system simply allocates the intrinsically more able to the higher-paid jobs, why should the taxpayer fund such a system?

³ÉÈËVRÊÓƵ

For the alternative "productive investment" view to be valid not only must it be shown that higher education (or indeed schooling) makes the recipients better at the jobs they subsequently claim, but that some of the gains spill over to others. In other words, gains to the economy as a whole have to exceed the gains accruing to the educated individuals.

This might be observed in a factory production-line system for example, in which workers of different skills contribute sequentially to the final product. Improvements in the productivity of any one group of workers can benefit those further along (or further back) in the production process by improving the quality of the final product.

Additional gains might be realised if the more skilled workers use their education to devise improved production methods for the less skilled workers. The economic pie in this view is therefore not of fixed size but can expand tomorrow by investing a proportion of it today in appropriate education.

Is it possible to discriminate empirically between these alternative arguments in any systematic way?

Ideally one would hope to observe workers with different levels of education undertaking similar tasks and see whether the more educated performed these tasks more efficiently. Unfortunately such controlled experiments are almost never possible. Not only must workers be identical in all relevant respects (including innate ability) except education levels but they must be performing the same tasks and it must be possible to identify the individual contribution of each worker.

Not surprisingly, with modern technologies and production methods this is rarely possible and, in any case, people with different education or skill levels tend to specialise in different tasks within the firm or the wider economy.

Recently economists have begun to follow a different approach. If it could be demonstrated that an economy's overall rate of productivity growth is positively related to the educational attainment of its labour force - its stock of "human capital" - this would provide at least some support for the "education as productive investment" view.

This does not mean comparisons such as: "country X educates a higher proportion of its 18 to 21-year-olds than country Y and country X experiences a higher rate of economic growth than country Y". Such evidence is simplistic for at least three reasons. First, in such an anecdotal approach it is almost always possible to find counter examples where countries with more education also have lower growth rates.

³ÉÈËVRÊÓƵ

³ÉÈËVRÊÓƵ

Second, we would not expect that producing a few more graduates would have an immediate and substantial effect on economic growth. However to the extent that education expansion builds up a more educated labour force, over a (perhaps lengthy) period of time we could expect to see the improvements in labour quality affecting economic performance.

Third, without controlling for the other determinants of productivity growth there is a strong risk either that any observed associations between education and productivity growth are spurious or that we fail to observe a "true" association because of the countervailing effect of some omitted productivity determinant.

For example, suppose an economy's productivity is affected both by the amount it invests in physical capital embodying new technologies and by its investment in labour skills (via education). We would not be surprised to observe that the country which invests more in skilled labour has lower productivity growth than its neighbour if the neighbouring country simultaneously invested more in new technologies. This would be reinforced if the benefits from more educated labour could only be realised where there is simultaneous investment in new technology.

Clearly any potential positive impact of more education on productivity will not be realised in economies with low physical capital investment. The appropriate conclusion however is not that education does not impact positively on economic growth. For a given level of investment, education does impact positively on growth.

Recent empirical evidence suggests that this is precisely what we do observe. If we look at the group of Organisation for Economic Cooperation and Development countries over the post-1960 period, and after controlling for the effects of investment and other growth determinants, there is fairly strong evidence to suggest that countries which began the period with more tertiary educated labour, and which expanded that labour more rapidly, experienced faster rates of per capita income growth.

There also appears to be some evidence that countries which expanded their secondary school educated labour forces more rapidly may have benefited from higher physical capital investment as well. Perhaps then, education can also boost growth indirectly by providing the quality of labour necessary to exploit profitable investment opportunities.

Lastly, the good news does not seem to be confined to OECD countries. Though tertiary education is fairly limited in most third world countries, there is now evidence that for those countries, expansion in the number of workers with primary and secondary education has had a positive impact on their economic growth rates, at least since 1960.

As is often the case in the social sciences, it is dangerous to draw definitive conclusions from a small number of studies. Accumulation of evidence is required before we can be confident that we do not simply have a few "rogue" results. Nevertheless, the availability of new international data in recent years is allowing more systematic investigation of the education- economic performance link.

Initial results seem to be encouraging for those who would argue that being a student is not just about enjoying beer and wild parties at public (or, increasingly, private) expense!

Norman Gemmell is professor of development economics at the University of Nottingham.

³ÉÈËVRÊÓƵ