The Reserve Bank of India today released a DRG Study entitled, ‘Regional Inequalities in India in the 1990s: Trends and Policy Implications’. The Study is co-authored by Dr. Nirvikar Singh, Professor, University of California, Santa Cruz, USA along with Mr. Jake Kendall, World Bank, Shri R. K. Jain, Director, DEAP and Dr. Jai Chander, Assistant Adviser, DEAP.

There are concerns that the regional inequalities have accentuated in the post reform period. While several statistical analyses conducted at the state level support this hypothesis, there is limited systematic analysis/examination of sub- state regional inequality. Disaggregated studies have typically relied on qualitative or descriptive statistics.

The DRG study on ‘Regional Inequalities in India in the 1990s: Trends and Policy Implications’ attempts to fill the gap in the existing empirical analysis by providing a more fine-grained and quantifiable understanding of the trends in regional inequality in India.

The study is a first of its kind in representing detailed analysis of Indian data below the state level. The study finds that given the size and heterogeneity of Indian states, cross state comparisons and growth regressions tend to be limited in providing useful insights into policy formulation. By recognising the district as an administrative unit and mapping its growth performance and the determinants, the study seeks to provide additional guidance to policy makers. The study also highlights the need for better data collection and empirical work to be undertaken as inclusive growth requires basic understanding of the parameters of broad based growth.

The main findings of the study are:

• At the regional level, partial measures of economic activity do not indicate any strong evidence of conditional convergence or divergence. There is, however, clear evidence of conditional convergence in per capita consumption level. Three points are noteworthy in these results:
  
  
  • First, the convergence result is the strongest for urban households.
    
    
  • Second, the main conditioning variable is petrol consumption, which could be an indicator of the quality and quantity of road infrastructure (and which could also be related to access to urban areas).
    
    
  • Third, dummy variables for the poorer states do not indicate that they were doing worse than the benchmark average state (Andhra Pradesh), though some of the regions with the largest negative residuals were in the poorer states.
    
    
• The district level results do indicate conditional convergence, but not absolute. The conditioning variables used are measures of roads, literacy and credit, so the results are supportive of the importance of infrastructure and human development, as well as access to finance. These results are quite robust across a wide variety of specifications, and are consistent with a well-understood model of development, that emphasises human capabilities and appropriate access to markets as determinants of growth. This study can, therefore, be used to identify districts which require additional policy intervention along the three dimensions used, as well as districts where the performance is worse than the average, even after conditioning on development measures. In the districts where the performance is worse than the average, social backwardness or policy implementation shortcomings may be the problem.
  
  
• The results for conditional convergence hold across states, as well as within most states in the sample. This indicates that attention to improving these variables in districts where they are at relatively low levels can have a growth payoff. It could also improve the inclusiveness of growth, as measured by convergence of income levels across geographic regions.

Alpana Killawala
Chief General Manager

Press Release : 2009-2010/1547