Why are some places richer than others? Economists offer many answers to this question ranging from better institutions to favourable geography to more education. Much of this research looks at differences across countries in the hope of identifying specific conditions that are present in rich countries but absent in poor countries.

But countries differ along many dimensions such as culture, geography, institutions and income. How do we separate the effect of, say, institutions from culture and geography, which may also affect development?

To get around this problem, we look at the effect of education on economic prosperity across districts of India. Unlike cross-country comparisons, the focus on districts offers a tight comparison, as these sub-national units share common governance and national policies (that is, institutions).

Controlling for differences in geography, we find that districts with higher education, not primary education, are more developed. Since there are few measures of income per capita at the district level, the study measures development using the amount of light density at night in a district (luminosity). Figure 1 shows the relationship between the share of people with higher education and light density in India.

Figure 1. Higher education and light density

Areas with more higher education have higher light density. But does this mean that higher education causes more development? More developed districts may have more money to invest in higher education. In that case, more economic development may lead to higher education, and not the other way around.

We address this chicken and egg problem using the location of Catholic missionaries in India in the early twentieth century. We show that more people completed higher education in districts where Catholic missionaries settled in 1911 – and that the resulting higher education in these districts led to more economic prosperity.

Why would the historic presence of Catholic missionaries lead to more higher education today? When demand for higher education began to increase in India in the 1950s, districts with a historic presence of Catholic missionaries were better placed to respond. Indeed, the number of Catholic colleges expanded after Indian independence, and as they expanded, they did so more in districts with a Catholic missionary presence.

One potential criticism of our argument is that Catholic missionaries may have settled in districts that would have developed anyway. For example, if Catholic missionaries settled in richer and more educated districts, then the link from Catholic missionaries to higher education to luminosity would be invalid. But we show that this is not the case. While Catholic missionaries were more likely to locate in coastal districts and in proximity to railways, they also settled in districts with more tribal groups (less developed areas).

Furthermore, there is no significant relationship between the location of Catholic missionaries in 1911 and measures of development such as the historical provision of education or income tax revenues. Rather, the mechanism from Catholic missionaries to current higher education is via Catholic colleges that were established after Indian independence in 1947.

Higher education is 2.9 percentage points higher in districts with a historic Catholic missionary presence (higher education averages 18 per cent across Indian districts). Using historic Catholic missionary presence as an instrument, we find that increasing higher education by one percentage point increases log light density by 0.07, an economic effect of 1.6 per cent given mean log light density. The main driver of this relationship is the strong impact of higher education on the tertiary sector (largely services).

Our results are robust to alternative measures of development. We also rule out other non-education channels by which Catholic missionaries could affect current income. In particular, we rule out factors such as health, infrastructure and openness to foreign ideas. Such factors may be correlated with Catholic missionary presence and subsequent development. But, we find the effect of higher education on light density is robust to including these non-education channels.

While higher education leads to development, a population being stuck in lower levels of education such as primary-middle schooling do not contribute to Indian growth. Thus focusing merely on primary education (through policies and aid) is a necessary but insufficient condition to foster prosperity. In particular, it may be important also to focus on the higher levels of education that generate economic prosperity.



Amparo Castelló-Climent is a senior researcher at the Institute for International Economics at the University of Valencia. She holds a PhD in Economics from University of Valencia and a MSc in Economics from University College London. Her research interests include the quality and distribution of human capital, the distribution of income and their influence on economic growth, and the interplay between human capital and political institutions. She was previously an associate professor at Universidad Carlos III in Madrid and a researcher at LSE’s Centre for Economic Performance (CEP).

Latika Chaudhary is an associate professor at the Naval Postgraduate School’s Graduate School of Business and Public Policy. Latika does research in economic history, development economics and public finance.



Abhiroop Mukhopadhyay is a professor at the economics and planning unit of the Indian Statistical Institute. He has a PhD in economics from Pennsylvania State University. His areas of interest are microeconometrics, international trade and development economics.