S Chandrasekhar (Indira Gandhi Institute of Development Research)

In this paper, we provide a rich description of distribution of Monthly Per Capita Household Earnings (MPCHE) following which we present estimates of inequality and undertake a decomposition analysis by rural and urban. By focusing on household earnings, our work complements the literature on consumption poverty in India. We highlight differences across states in rural and urban median MPCHE in form of a Pen’s Parade (following the vivid description of Jan Pen, 1971). Following Milanovic (2012) who assesses the income distribution of different countries vis-à-vis the world income distribution, we benchmark the distribution of household in rural and urban areas of each state and National Sample Survey region against the all India distribution of MPCHE. The graphical exposition clearly highlights the persistence of inequalities and lagging regions within India. The framework outlined by Kuznets (1955) and later formalized by many other including Anand and Kanbur (1993) is useful for analyzing evolution of inequality with development. We focus on specific aspects of the Kuznets process. Consistent with the framework we find divergence in rural and urban well-being. For the year 2018-19, we estimate the inequality in rural and urban monthly per capita household earnings (MPCHE), as measured by Gini coefficient to be 0.366 and 0.456 respectively. For India as a whole inequality as measured by mean log deviation (MLD) is 0.341, and when we decompose this by rural and urban we find that the between-group component accounts for 20.6 percent of inequality. While we indeed find large differences in inequality within India, it is not true that states that have higher average MPCHE are necessarily more unequal. India is heterogeneous and the relative importance of urbanization varies across states.
In order to identify what explains difference in inequality within a state vis-à-vis India, we adapt the decomposition method outlined by Mookherjee and Shorrocks (1982) This allows us to account for the difference in inequality between that of a state and India as the sum of difference in inequality attributable to differences in within state and within all India group inequality, difference in inequality due to difference in the national and state urbanization rate on within and between group inequality and the contribution of relative differences in rural-urban mean MPCHE. Thus we can answer the following question in the context of Kuznets process. How much of the difference in state and all India inequality is because of difference in urbanization rate of a state and the national average? Is this the dominant contributor to the rural-urban between group component of inequality or is it the differences in the relative rural and urban earnings which explains the lion’s share of the observed differences?