India’s growth in the 2000s: Four facts

Arvind Subramanian 05 December 2011



Posting a growth of income per capita of 6.1% per annum during the first decade of this millennium (2001–09, the ‘oughties’), India now has three decades of respectable growth performance behind it, a point that is often obscured by the near-universal tendency to equate India’s growth turnaround with the policy turnaround that occurred in 1991.

India’s growth performance, especially across the states within the country, since the takeoff in the late 1970s/early 1980s has been the subject of considerable research interest (for example Ahluwahlia 2000), and controversy about when growth turned around and why (Economist 2005). But all the previous research covers the period until 2000. In a new paper (Kumar and Subramanian 2011), we compare growth performance across states during the 2000s and establish four key stylised facts about interstate growth performance.

Growth increased in most states

Figure 1 plots the income per capita growth rate for the 21 largest states for two time periods – between 1993 and 2001 (the ‘nineties’) and between 2001 and 2009. The figure shows that all states, with the exception of three, lie above the 45 degree line, ie, their growth in the 2000s was substantially greater than in the 1990s. Indeed, average growth across the 21 states doubled from 2.8% in the 1990s to 5.8% in the 2000s. The largest improvements were posted by Uttarakhand (7.0 percentage points), Maharashtra (5.8 percentage points), and Chhattisgarh (5 percentage points) with Gujarat, Orissa, and Bihar not far behind.

Figure 1. Average growth of net state domestic product per capita

Sources: Central Statistics Office and authors’ calculations.

Divergence in the growth performance across states continues

The strong performance of the hitherto laggards – Bihar, Orissa, and Chhattisgarh – has been one of the remarkable stories of the oughties. So did the “gale winds of divergence” noted by Rodrik and Subramanian (2005) and Kochhar et al (2006) change direction and become forces for convergence during the 2000s?

Figure 2 plots the growth rate of the states for the period of 2001–09 against the income per capita in 2001. It shows that richer states on average grew faster so that the inequality across states is actually increasing. We find that far from changing directions the forces of divergence continue. We show this more formally by estimating a standard growth regression. In this framework, average annual growth rate of income per capita over the period of 2001–09 is regressed on the logarithm of initial income per capita in 2001. We find that the coefficient on the log of initial income per capita is positive and statistically significant, indicating divergence across states over the period of 2001–09. We repeat the exercise for different time periods and for different sample sizes (main states and all states) and, broadly, find that the result on divergence continues to hold.

Thus, the strong growth performance of the laggard states should not obscure the more general pattern that across the Indian states, namely, that we still do not see the phenomenon of convergence, whereby the poorer states, by virtue of growing faster than the richer states, start catching up with the latter’s level of income.

Figure 2. Growth during 2001–09 and income in 2001

Sources: Central Statistics Office and authors’ calculations.

Faster growing and more globalised states took a bigger hit during the crisis

The crisis of 2008–10 highlighted the vulnerability that is the flip side of the dynamism that globalisation has engendered. Growth declined in India, and capital fled, as in most other countries, albeit to a lesser extent. But the question remains if there was any differential in the growth performance across states during the crisis years and which states took a bigger hit. We find that the states with the highest growth during the pre-crisis years were the ones which registered greater decline in growth during the crisis years.

Could it be the case that states that were the most open or globalised before the crisis were affected the most during the crisis? We cannot easily measure the degree to which each state trades internationally but we can estimate crudely how tradable is the economic profile of each state. Since manufacturing and business services tend to be highly tradable, we use these – specifically, the share of manufacturing and business services in total state output – as proxies for the openness of each state. We find a negative correlation between this share and change in growth during the crisis (Figure 3). Karnataka, Maharashtra, Tamil Nadu, and Gujarat are among the most open states and they also experienced the greatest growth declines. In contrast, Bihar, Jammu and Kashmir, and Assam, which produce relatively few tradable goods, were the most resilient during the crisis. Of course, there are likely to be a multiplicity of factors at work which precludes drawing any clear causal conclusions, but the simple correlations seem to be consistent with globalisation conferring dynamism and stoking growth but at the same time inducing vulnerability.

Figure 3. Change in growth during crisis years and openness

Sources: Central Statistics Office and authors’ calculations.

Demographic dividend seems to be disappearing

Hope in India's future growth is founded on the demographic dividend. Demographics affect growth because different age groups exhibit different economic behaviour. A higher share of the working-age population has a positive effect on growth through various channels – a higher labour supply, higher savings as working-age groups tend to save more than the young and the old, and greater investment in education and health as the number of children being raised declines and the lifetime over which the investment can be recouped becomes longer. Thus, a favourable change in the age structure ie, an increase in the share of the working-age population, as captured by the growth in the share of the working-age population, has the potential to positively influence growth.

The demographic dividend is routinely touted by analysts and forecasters as one basis for optimism for India’s economic future. And corroborative evidence was provided in two recent papers by Kumar (2010) and Aiyar and Mody (2011). But the pattern of growth in the 2000s appears to muddy the waters. A simple scatter plot of the growth of income per capita for 2001-09 and change in the share of the working-age population in total population (working-age ratio) between 2001-09 (based on Census 2001 projections) shows that the two are negatively correlated (Figure 4), a fact that we confirm econometrically.

Figure 4. Growth in the share of the working-age population and growth of income per capita

Sources: Central Statistics Office, Census of India, and authors’ calculations

This could be due to the fact that there are significant differences before and after 2001 in the states which see a favourable demographic structure. Post-2001, based on the population projections from the 2001 census, Kumar (2010) shows that 49% of the growth in the working-age population is likely to have been concentrated in four states, the BIMARU states (Bihar ­­– undivided, Madhya Pradesh – undivided, Rajasthan, and Uttar Pradesh – undivided)..While the BIMARU states, especially Bihar, did perform better in the 2000s than in the 1990s, they still lagged behind the other states. That might explain why we find here that the growth in the share of the working-age population is not positively correlated with economic growth in the 2000s. At least so far, these states have not been able to fully utilise the young population to their advantage. But this might change in the future.


India’s growth has been distinctive in many ways. It has relied on services rather than manufacturing as an engine of growth; growth has been skill-intensive rather than intensive in the use of India’s abundant factor, labour; India, despite being poor, is exporting skills and technology in the form of FDI to countries much richer than itself.

The analysis of growth in the 2000s throws up one more quirk, relating to Kerala. The conventional wisdom is that this state is Scandinavian in its social achievements but sclerotic in its growth performance because of investment-chilling labour laws and strong trade unions. However, the data suggest that the conventional wisdom and the caricature are dead wrong. Kerala posted among the highest rates of growth in the 1990s (4% per capita), continued its stellar performance in the go-go 2000s (7.5%), and exhibited great resilience during the crisis, experiencing virtually no decline in growth.

The Indian growth miracle, including the experience of the 2000s, continues to confound.


Ahluwalia, M (2000), “Economic performance of states in the post reforms period”. Economic and Political Weekly May 6: 1637–48.

Aiyar, S and A Mody (2011), “The Demographic Dividend: Evidence from the Indian States”. IMF Working Paper 11/38..

Economist (2004), “Who put the shine into India? Was it Sonia Gandhi's economic lieutenant, her husband, or her mother-in-law?” 27 May.

Kochhar, K, U Kumar, R Rajan, A Subramanian, and I Tokatlidis (2006), “India’s Pattern of Development: What Happened, What Follows?” Journal of Monetary Economics 53: 981–1019.

Kumar, U (2010), “India’s Demographic Transition: Boon or Bane? A State-Level Perspective”. Economics Program Working Paper #EPWP 10–03. New York: The Conference Board.

Kumar, U and A Subramanian (2011), “India’s Growth in the 2000s: Four Facts”. Working Paper Series WP 11-17. Peterson Institute for International Economics.

Rodrik, D, and A Subramanian (2005), “From ‘Hindu growth’ to productivity surge: the mystery of the Indian growth transition”. IMF Staff Papers52 no. 2: 193–228.



Topics:  Development

Tags:  growth, India

Chief Economic Adviser to the Government of India