Growing regional disparities in India are a cause for concern. But little is known about the relative importance of possible reasons for the varied growth experiences across the country. This column explores growth imbalances among Indian districts. Proximity to cities, infrastructure, degree of urbanisation and state government policies are found to be key determinants.
The unbalanced nature of India´s growth is becoming a cause for considerable concern. The growing regional disparities appear to have dampened political resolve for further economic reforms, and hence may pose a barrier to India’s future economic growth. Indeed, the causes of this imbalance are key points of contention in the recent debate between Bhagwati and Sen.
Dimensions of India’s unbalanced growth
The evidence on the imbalance in India´s growth has several dimensions. First there is evidence of disparities across state income levels and growth rates, with richer states growing faster (For example, Cashin and Sahay 1996, Trivedi 2003, Ghate and Wright 2012, Bandyopadhyay 2012).
This divergence is puzzling given that there are no political barriers to migration, almost free trade, and a common set of federal institutions. It suggests an important role for geographical barriers, such as transport and migration costs, but these issues have not been widely explored.
Complementing these state-level studies is an extensive literature, typically based on household survey data, on rising returns to education and differences in human capital, opportunities for employment, and the provision of public infrastructure (Cain et al. 2010, Azam 2012, Basu and Maertens 2009, Sachs 2009, Lall, Wang, and Deichmann 2010, Ghani, Goswami, and Kerr 2012, Crost and Kambhampati 2010).
Together these studies suggest a wide array of possible reasons for differing growth experiences across India. However, we know little about the relative importance of each.
Analysing district-level disparities
We use a datasets1 on district-level income and socio-economic indicators to explore the determinants of India’s growth at the district level (Das, Ghate and Robertson 2013). In particular the district-level data allow both socioeconomic and geographic factors to be considered. For example, some Indian districts are cities, while others are rural or semi-urbanised. Moreover some districts are very remote, while others, though rural, may be close to cities. If we take the economic geography literature seriously we would expect to observe different long-run income levels in these districts reflecting differences in transport, migration and communications costs.
We thus consider the determinants of growth across India’s 575 districts. In addition to considering the usual factors based on socio-economic characteristics of each district, we also construct a measure of district “remoteness” - defined as the minimum distance by road from a district headquarter to one of the 10 major urban agglomerations in India.
The first feature of the district-level data is the wide disparity in income levels across states. For example, there is a 9.8 fold difference between the richest state Goa, and the poorest state Bihar. But at the district level, the range of per capita incomes is from 3,858 million rupees (Sheohar district in Bihar) to 139,868 million rupees in Jamnagar district, Gujarat. This is a ratio of, roughly 36, which implies that the real income gap between districts in India can be as large as the gap in real incomes between the world’s richest and poorest countries - such as the US and Rwanda2.
A preliminary analysis of the data shows that, despite this enormous gap in income levels, there is no evidence of convergence3. Rather the trend, if anything is one of absolute divergence with richer districts getting richer faster than poorer districts. This shows the enormous difficulty that the government faces in promoting reforms which may amplify these differences further in the short term.
In more detailed analysis, we then explore the possible causes of these growth rates. We find that infrastructure indicators, such as urbanisation and electrification are significant factors. In addition we find that, even after accounting for many socio-economic characteristics, significant differences in growth rates remain across states. This suggests that state governments play an important role in determining growth rates through providing appropriate institutional or regulatory environments, but also that within states the allocation of public infrastructure is important.
The significance of remoteness
In our analysis, we also find that remoteness is an important factor in explaining differences in growth rates across districts (Das, Ghate and Robertson 2013). As an example, the most remote district from any large urban agglomeration in India is in Manipur, near the Burmese border. It is 2,531 km from Kolkata, the closest city. Conversely, some districts areas are as little as 8 km from a major city. We predict that, in this case, the closer district would have a per capita Gross Domestic Product (GDP) level approximately four times higher than those in the remote parts of Manipur. Likewise, the more remote district would have a growth rate that is in the range of 1.5 to 3 percentage points lower, other factors being equal.
Thus remoteness has an important economic effect on growth rates for very remote districts, though for many districts the effect is small. Subject to this, and other infrastructure factors, we find that there is strong evidence of conditional convergence across districts4.
Nevertheless we also find that conditional convergence across Indian districts is a very weak force. It is only about half of the usual rate, known as Barro’s iron law of convergence5 of 2% (Barro 2012). Our estimated convergence rate implies that after a decade of growth, the income gap between two districts would still be 90% of its level at the start of the decade.
Lessons for policy
Our research helps clarify the roles of several alternative theories about India’s unbalanced growth. The differences that exist across Indian districts in terms of proximity to cities, infrastructure provision, and degrees of urbanisation are all important factors. In addition, there are also significant differences across states - suggesting that state-level policy differences are also contributing to India’s growth imbalance. The results are thus suggestive of an important role for the government in addressing India’s growth imbalances, through appropriate state-level reforms and by adopting best practice in the supply of public infrastructure.
Nevertheless there remains much to explain. Remoteness, though important, has large effects only for very remote districts. Likewise the slow rate of convergence suggests that there are other significant, but currently unidentified, barriers to growth across India.
- Indicus Development Landscape and Indicus GDP data sets.
- Real incomes across countries can be compared using Purchasing Power Parity estimates of price levels, such as those produced by the World Bank and the Penn World Tables.
- The idea of convergence is that poorer economies´ per capita incomes will tend to grow at faster rates than richer economies. As a result, all economies should eventually converge in terms of per capita incomes.
- There is "conditional convergence" when economies experience "convergence" but conditional on other factors that influence growth being held constant.
- Barro’s "iron law of convergence" suggests that about two percent of the gap between a rich and poor country disappears each year.
- Azam, Mehtabul (2012), ‘Changes in wage structure in urban India, 1983–2004: A quantile regression decomposition’, World Development 40(6), 1135–1150.
- Bandyopadhyay, S. (2012), ‘Convergence Club Empirics: Evidence from Indian States’, Research on Economic Inequality, 20, 175-203.
- Barro, Robert J. (2012), ‘Convergence and modernization revisited’, Paper Presented at the Nobel Symposium on Growth and Development, Stockholm, September 3-5, 2012.
- Basu, Kaushik, and Annemie Maertens (2009), ‘The growth of industry and services in South Asia and its impact on employment’ In Accelerating Growth and Job Creation in South Asia, New Delhi: Oxford University Press, pp. 81–140.
- Cain, J Salcedo, Rana Hasan, Rhoda Magsombol, and Ajay Tandon (2010), ‘Accounting for inequality in India: Evidence from household expenditures’, World Development 38(3), 282–297.
- Cashin, Paul, and Ratna Sahay (1996), ‘Internal migration, center-state grants, and economic growth in the states of India’, International Monetary Fund Staff Papers, pp. 123–171.
- Das, Samarjit, Chetan Ghate and Peter E. Robertson (2013), ‘Remoteness and Unbalanced Growth: Understanding Divergence Across Indian Districts’, ICRIER Working Paper 268, September 2013.
- Ghani, Ejaz, Arti Grover Goswami, and William R Kerr (2012), ‘Is India’s manufacturing sector moving away from cities?’, Technical Report, National Bureau of Economic Research.
- Ghate, C., and S. Wright (2012), “The V-factor: Distribution, timing and correlates of the great Indian growth turnaround”, Journal of Development Economics, 99(1), 58-67.
- Crost, Benjamin, and Uma S Kambhampati (2010), “Political market characteristics and the provision of educational infrastructure in north India”, World Development 38(2), 195–204.
- Lall, Somik V, Hyoung Gun Wang, and Uwe Deichmann (2010), ‘Infrastructure and city competitiveness in India’, Technical Report, Working paper, World Institute for Development Economics Research.
- Sachs, Jeffrey D (2009), ‘South Asia story of development opportunities and risks’, in Ejaz Ghani and Ahmed Sadiq (ed.), Accelerating Growth and Job Creation in South Asia, New Delhi: Oxford University
- Trivedi, K. (2003), ‘Regional Convergence and Catch-up in India between 1960 and 1992’, Mimeo, Nuffield College.