In this blog post, Ewout Frankema of Wageningen University & Research and Marlous van Waijenburg of Harvard Business School introduce their latest article for the Economic History Review, entitled “What about the race between education and technology in the Global South? Comparing skill premiums in colonial Africa and Asia”.
—
The historical co-evolution of technology, schooling and labour-market institutions, also known as the ‘race between education and technology’, is a central theme in the literature on modern economic growth and inequality. The idea of a ‘race’, as Jan Tinbergen (1975) called it, refers to three central features of the modern economy: continuous technological change that requires flexible education systems to equip labour forces with the knowledge and skills to work with frontier technologies, and to keep the cycle of innovations going. While in-depth studies on the race between education and technology have provided valuable insights into long-term patterns of economic development, the existing literature has so far largely focused on the ‘West’ (cf. Goldin and Katz, 2008).
Our article shifts the focus to the race in two major parts of the Global South: Africa and Asia. This shift forces us to think carefully about colonialism as a central factor influencing all three aspects of the race. As the conceptual framework in Figure 1 shows, colonialism was a major driver in the transfer of new technologies from the metropole to other parts of empire and created demand for new or different types of skills. Colonialism was also important in shaping the supply response. The introduction of mass education, often aided by missionary organizations, increased training possibilities for complementary skills. These ‘Western’ types of technology and education, in turn interacted with indigenous technologies, knowledge and institutions of human capital formation. Finally, through a range of (coercive) labour-market interventions, colonial rule produced specific frictions in the market for skilled and unskilled labour, mostly to avoid the high costs of market-clearing wage rates. All of these forces had an effect on skill-premiums—the additional wage premium that is paid to skilled workers relative to unskilled labour.
Historical datasets on skill premiums have provided important insights into the long-term dynamics of the interplay between technology and schooling. When education systems are efficient in adapting to the latest technologies, skill premiums are likely to be modest. Any rise in demand for skills is soon met with extra supply. Studies on skill premiums in early modern Western Europe have shown that the wage premium for traditional artisans (carpenters) was relatively low and stable already since the mid-fourteenth century. This observation is in line with a vast literature that emphasizes the well-developed and comparatively efficient nature of European apprenticeship systems (Epstein and Prak, 2008). Little, however, is known about the evolution of skill premiums in the Global South.
Our article presents the first skill-premium database for 34 African and 16 Asian economies that covers both the colonial and post-colonial periods. Our Africa Asia Occupational Wage Database contains wage series for indigenous males in six main occupations: unskilled labour, carpenters, electricians, car mechanics, lower-level clerks and bank tellers. This selection allows us to observe patterns for traditional artisanal skills, for ‘new’ blue-collar occupations that spread to the colonies from the metropole and for white-collar occupations that required proficiency in literacy and numeracy.
Our series reveal three major patterns which are all illustrated in Figure 2. First, skill-premiums in both Africa and Asia were much higher than in the West for most of the twentieth century. Second, despite some significant internal variation, skill premiums in Africa were on average much higher than in Asia around 1900. Third, skill premiums in both Africa and Asia fell dramatically in the course of the twentieth century, ultimately converging with levels that had long been observed in Western countries.
These insights raise two major questions about the race between education and technology in Africa and Asia. First, why were skill-premiums so much higher in Africa at the start of the twentieth century (and possibly also in the centuries before)? Second, what drove the dramatic and universal fall in skill-premiums?
While detailed answers to these questions require a more extensive research programme, we offer some initial hypotheses in our paper. As for the Africa-Asia gap, it is unlikely that higher African skill-premiums can be explained by larger demand shocks caused by imported technologies. While African economies were clearly in flux, there is no evidence that they grew faster than Asian economies in the period 1870-1940. It is therefore more plausible that there were differences between Africa and Asia on the supply side: in the adaptation of schooling systems to changes in technology regimes under colonial rule. To think about this question more systematically, we designed a framework (Figure 3) that presents a spectrum of apprenticeship institutions defined by access and formalized standards of teaching. One of the hypotheses we develop is that greater reliance on lineage-based systems of intergenerational skill transfer and the more intensive use of slavery may have lowered opportunities for the diffusion of formal vocational education in large parts of Africa.
The answer to the second question—the universal free-fall in skill premiums—seems more straightforward, at least at first sight. Clearly the long-term demand for skilled labour was on the rise across broad segments of the economy. Construction workers and engineers were needed for infrastructural projects, clerks were demanded for expanding colonial state bureaucracies, companies, trade houses and so on. The dramatic fall in skill premiums must therefore be largely caused by expanding supplies of skilled workers, a long-term effect of expanding mass education. Nevertheless, other factors, such as changes in the colonial governance of labour markets, played a role as well. The coercion of workers in African as well as Asian colonies declined over time, and may have underpinned rising wages for unskilled workers. In addition, the free or forced immigration of skilled workers to areas of high demand may also have put downward pressure on skill premiums.
In sum, our article is the first to reveal major long-run patterns in wage premiums for a large number of African and Asian countries. Our findings raise a number of questions, for which we provide provisional answers. A more thorough understanding of the dynamics of the race between education and technology, requires a larger research agenda than our paper can cover. This research direction, however, not only holds great promises for our understanding of the human capital–growth nexus in non-Western economies, but can also shed light on some of the deeper roots of the ongoing South–South divergence that is rapidly transforming global economic relations.
References
Epstein, S. R., and Prak, M., Guilds, innovation, and the European economy, 1400-1800 (Cambridge, 2008).
Goldin, C., and Katz, L., The race between education and technology (Cambridge, MA, 2008).
Tinbergen, J., Income Distribution: Analysis and Policies (New York, 1975).
To contact the authors: