Here is a short paper written by Montek Singh Ahluwalia (2000), which has come to be very important in the field of inter-state inequality in India:
The purpose of the paper is to analyse whether growth became more unequal in the post-reform period, and if yes, what the causes of this inequality are. The first part of the paper reviews the growth performance of states in both the pre-reform and post-reform period through a number of summary statistics such as coefficient of variance. I have mentioned a few of these in the last two posts. There is a brief section of what this means for poverty.
The second part of this paper, which deals with the determinants of growth in the states, is the section I find most insightful. The author picks up possible culprits behind the divergence in growth between states and then tests their validity at a very basic level. It would be worthwhile to mention a few of his results here:
- Investment Ratios: After recognising the inadequacy of investment data at the state level, the author runs regressions of the growth rate on (1) public investment (2) private investment (3) sum of both. He finds, not so surprisingly, that the only statistically significant coefficient in the post-reform period was found for the model containing only private investment. This could support the hypothesis that after reforms, public investments had to be reined in due to aggressive fiscal targets, and this increased the importance of private investment. Private investment, however, is not as equally distributed as public investment and hence would tend to go to states that are more efficient at using their resources, and hence are richer. This would be a worthy topic of future research.
- Plan Expenditure: Using only the available data of state-level plans, the author runs regressions of growth rate on the planned expenditure as a percentage of state GDP. Expectedly, he finds that the coefficient was significant in neither the pre-reform or the post-reform period. This would support the assertion made earlier that planned or public expenditure has begun to matter less.
- Human Resources: Taking the literacy rate as a proxy for the quality of human resources, the author runs regressions of the growth rate on the literacy rate. In the pre-reform period, the sign of the literacy rate comes out to be negative, which is counter-intuitive. In the post-reform period, it comes out positive as would be expected; however, in neither period is the coefficient statistically significant. When combined with the investment ratio to create an interactive dummy, however, the coefficients turn out to be statistically significant. Hence, it could be concluded that investment yields high growth only in the presence of good human resources.
- Infrastructure: The independent variable used here is the CIME composite index of the relative infrastructure capacity of different states. When a regression is run, both the coefficients and the R-square are not statistically significant. The author notes that three components of this index - percentage of villages electrified, per capita energy consumption and teledensity - have a statistically significant relation with growth rate; he however cautions that no major implication should be drawn from this.
No comments:
Post a Comment