Report №8: Externalities to higher education: a review of the new growth literature

• Recent advances in growth theory have brought into focus the role of education in the creation of human capital, in the production of new knowledge, and the possibilities of education-related externalities. In principle these externalities could be either consumption or production related though it is only the latter which have been explored.
• New growth theories have proposed a number of mechanisms whereby education affects productivity levels: education is important for successful research activities (eg. by producing scientists and engineers) which are, in turn, important for productivity growth; and education creates human capital which directly affects knowledge accumulation and therefore productivity growth. Though higher education is not often modelled explicitly in these theories there is a prima face case for higher education because of its twin outputs of graduates typically embodying 3-6 years of post-16 education and production-relevant research.
• New growth theories in which education generates human capital typically incorporate at least one of two crucial assumptions: (i) that there are constant returns to all (physical plus human) capital used in production (implying increasing returns to all factors including 'raw' labour); and (ii) that there are positive externalities to human capital in production. The empirical basis for these assumptions is essentially unknown. If there are human capital externalities in production we should expect to observe increasing returns to 'broad' (physical plus human) capital at the 'aggregate' level. The aggregate here will depend on the pervasiveness of spillovers - whether at the firm, industry, sector or economy-wide level.

Empirical evidence

• Countries with higher average years of education (including higher education where appropriate) of their labour forces tend to grow faster, other things being equal. There is some, tentative evidence that the magnitude of this effect observed at the economy-wide level may exceed that observed at more micro levels indicating possible externalities.
• Other things being equal, OECD countries which expanded their higher education more rapidly from 1960 experienced faster growth. The direction of causation however is unclear.
• Education seems to be important more via its effects on productivity rather than directly as a factor input.
• The only specific group of graduates which have been examined for productivity growth effects are 'scientists and engineers' and results generally confirm positive effects.
• There is some evidence that education affects physical capital investment in the economy as a whole, which in turn raises income growth rates, though the specific role of higher education in less clear in this process.
• There is increasing evidence that research and development activities may be important for productivity growth and some of the gains from such activities spillover to other firms and countries. The additional link from higher education to research and development (R&D) is yet to be confirmed but some evidence is beginning to suggest that HE may be important for the development of innovative research and the ability to acquire and adopt it.
• Productivity advantages in 'leading' countries may take many years to spill over to others so that there can be relatively long-run gains to countries which use their education/research systems productively to generate new ideas or technologies. Within countries, regionalised production is common, though not specific to 'new' industries. If HE has a role to play in this process it is probably in sustaining rather than initiating new industries in specific localities, via the provision of appropriate labour skills. HE research does not appear to be primarily aimed at local firms, though key local entrepreneurs sometimes originate in the local HE sector.