The relationship between human capital and economic development -status: It’s complicated
Human capital for development
Conventional wisdom on economic growth and development emphasizes the importance of human capital. But is the link between education and income as robust as we think? At the individual level, the link is fairly well established (micro studies find robust labour market returns to education) but empirical evidence linking education to income at the macro level has produced mixed findings. The road to eliciting the macroeconomic return to human capital is full of landmines, some of which our paper, presented at the UNU-WIDER Conference on Human Capital and Growth, tries to tackle in the context of India, a large developing country.
The study of India is both of interest in itself, and also is a key part of our empirical strategy in extricating the impact of human capital on economic development. Motivated by previous work on India, we focus on tertiary education. This focus is motivated by the fact that India is a service sector-driven economy, and it is expected that tertiary education should correlate to economic growth and development. But this correlation may not necessarily imply that tertiary education leads to development.
To establish that it does, we use history—the spatial location of catholic missionaries in early 20th century. We show that the historical presence of such missionaries affects the current tertiary completion rates but does not affect development through any other channel. Using the differences in tertiary education completion rates, induced by history, our study indeed finds that tertiary education has a positive causal impact on economic development. To validate this finding, we address various potential obstacles, with a mixture of empirical methods and historical facts.
Obstacle 1: The measuring of human capital
The first obstacle to such research is the measurement of human capital. Human capital is usually proxied by average years of education. Such indicators typically reflect differences in literacy/schooling rates across countries. Focusing attention on the average tends to mix the impact of various levels of education completions. This can hide important impacts. For example, what level of human capital drives growth most?
In some cases, the decomposition of human capital to the various education completion levels can be startling! For example, our analysis reveals that only the share of adult population with completed tertiary education matters for the growth of the state level gross domestic product in India. So is the mass of individuals who drop out of school useless for economic development? This is clearly something for policy makers to ponder.
Obstacle 2: Accounting for institutions
It is well recognized that the correlation between human capital and economic development does not imply that human capital leads to economic development. So establishing the theorized causation is key.
Cross-country analyses, the workhorse of this field, often struggle to account for various institutional and cultural differences between countries that co-vary with human capital. One strategy is to look at how levels of economic development and growth vary with human capital within areas that share similar institutions and culture. In our case we used the districts within states of India.
Districts within a state share the same institutions, state government, culture and are ethno-linguistically more similar than countries. The need for district-level income data is solved by the use of a proxy—light density at night as measured by satellites (luminosity per km2). Does the link between tertiary education and income go through, even at this level of disaggregation? A resounding yes!
Obstacle 3: Reverse causality
… this may still be a completely spurious correlation. Higher income and faster growth rates generate more resources to invest in education. This means that analyses that use the correlation between the current levels of human capital and income proxies may suffer from reverse causality. To address this problem, one needs to search for circumstances and instances where human capital was accumulated for reasons independent of income.
In our work, we posit and show that the proportion of adult population with tertiary education is higher today in districts where, in 1911, Catholic missionaries were settled. This is largely driven by the spread of Catholic educational institutions after India’s independence in 1947, and is linked partly to a special protection status given by the Indian constitution to educational institutions run by minority communities. As the educational institutions run by the Catholic Church expanded, they did so more around the old Catholic missionary locations. Hence these districts show a much higher human capital accumulation, in particular tertiary education.
This is not surprising given studies from all over the world, that show that Christian missionaries promoted education among their flock with Catholics involved in more elite education. However, where the context of India is different is that while in other parts of the world, Christianity also played a big role in shaping other institutions—technology, infrastructure development, savings behaviour, etc. (most of them being countries where Christianity is now a dominant religion)—in India their role was largely limited to education. Our work gives evidence for this hypothesized non-association. Further, even if we were to control for the impact of all these factors on economic development, the impact of tertiary education on development remains the same, thus confirming that the transmission of Catholic missionaries to economic development is only through education.
Obstacle 4: Did Catholics settle in developed areas that would have developed anyway?
A criticism of our strategy could be that perhaps Catholics settled down in areas which would have anyways been developed today? For example, if they settled in what are today state capital cities, we would incorrectly infer that Catholic missionaries caused the development of the cities through education. We show that this is not the case. In fact, apart from being situated on the coast, a fact driven by history and controlled for in our work, Catholic missionaries were largely located in areas not necessarily beneficial to development; for example, in tribal areas, in hilly areas. Their location does not correlate with measures of income in the early 20th century. Interestingly, their locations do not even correlate with the presence of educational institutions in 1901. They do so only post-1950s, consistent with the mechanism of influence suggested above.
The history of Catholic missionaries therefore yields us the intra-state district-level differences in tertiary education that help us identify the causal impact of human capital on economic development. As hypothesized, our empirical analysis finds a positive effect of tertiary education on night luminosity. Similar effects are obtained for per capita consumption: one percentage point increase in the population share of individuals with completed tertiary education increases consumption per capita by 3.55%.
Implications for policy
Our research points out to the importance of tertiary education completion, especially for countries where GDP is driven by the service sector. However, in spite of the happy finding from our study that there is a strong link between tertiary education completion and economic development, human capital accumulation as a strategy for economic development is undermined by the fact that only 9% of adult population in India has completed tertiary education. Tertiary education completion requires good quality schooling. Hence our study does not imply that other levels of education are unimportant, rather they are key to the human capital–development link. In this context, mass schooling of poor quality, reflected by poor tertiary education completion rates, is not that useful for the macro-economy. However, arguably, this is exactly the kind of progress that has been made under the Millenium Development Goals.
Our thoughts are echoed by a recent report from the United Nations that says that the low primary completion rate (evidenced in India and many developing countries) may hinder the next Sustainable Development Goals of universal secondary enrolment. Perhaps the quest for mass education may need to be tempered by the need to provide quality education. Or else a large amount of resources need to be committed to provide quality school education to all. Only then will we be able to unleash the full potential of human capital.
Abhiroop Mukhopadhyay is Associate Professor in the Economics and Planning Unit at the Indian Statistical Institute, New Delhi.
- Castelló-Climent, Amparo, and Abhiroop Mukhopadhyay (2013). “Mass education or a minority well educated elite in the process of growth: The case of India”, Journal of Development Economics, 105: 303-320.
- Castelló-Climent, Amparo, Latika Chaudhary, and Abhiroop Mukhopadhyay (2016). “Tertiary Education and Prosperity: Catholic Missionaries to Luminosity in India”
Blog | Improving adult health increases children’s schooling - evidence from Zambia
Blog | Ending child marriage and closing the education gap
Blog | Land certification and schooling in rural Ethiopia
Blog | Migration and the transfer of gender and democratic norms