Many readers of my previous column that showed Kerala as the richest state in the country responded with their views on the state. Two common observations I made about these responses were that though Kerala was rich, it recorded the highest consumption of alcohol (a fact that I confirm from the data) and had the highest suicide rate. God’s Own Country was populated, according to some, with the devil’s own people — those who have a take-it-or-leave-it attitude or are outright lazy in their home state even as their kin work hard enough elsewhere and have an incentive to keep some people back home. Many other readers wanted to know what the situation was like in other states.
Honestly, I usually do not do interstate comparisons. The finding regarding Kerala, however, was just too good to let go. It was an aberration. States have varying sizes, different degrees of natural endowments and different history. Economic progress is often captive to these conditions and governance plays only a marginal role. It is instructive to note the continued dominance of Gujarat and Maharashtra in terms of attracting investments in spite of the riots and civil strife in both states over the past two decades when the governments clearly failed to contain a serious breakdown of law and order.
The country as a whole is likely to have suffered because of these riots, but the erring states do not seem to suffer much after an investor has decided to invest into the nation. Investments, therefore, continue to be driven by the degree of natural endowments, network externalities and economies of scale. Sometimes, bad governance (such as procuring land by fiat and then giving it away cheaply) also attracts investments. Maharashtra took pride in itself for having attracted Enron. So, bad governance attracts bad investments, and good governance will attract good investments. But, to expect good governance to deliver results quickly would be unfair. We must be patient with Nitish Kumar’s efforts in Bihar and must refrain from seeking evidence of his performance in interstate comparison exercises.
Interstate comparisons do not make much sense because often the intra-state disparity is high enough to render the state-level averages more or less meaningless. Small states face less diversity. But, as size increases, diversities also increase and then the state-level numbers start making lesser and lesser sense.
Ideally, estimations should be made over homogeneous regions — regions that have similar agro-climatic conditions and similar levels of economic development. Estimates of socio-economic parameters of such regions are more meaningful and can be used in comparisons.
The Centre for Monitoring Indian Economy (CMIE) has developed such a concept of homogeneous regions. These are clusters of neighbouring districts based on just three parameters — agro-climatic conditions, level of urbanisation and female literacy. Consumer Pyramids provide estimates of household income, expenses, savings, etc. at the level of the homogeneous regions. With the help of this data, we can now understand the sub-state level economic development of households. Note, that the income that is estimated here is not the same as the state domestic product which includes income generated by companies and governments besides that generated by households. We estimate the income and other parameters of households. This is the same data that had shown Kerala households as the richest.
This data shows that household incomes are remarkably uniform across the three homogeneous regions of Kerala. However, households are of a larger size in the northern region. As a result, the per capita income in Kozhikode-Mallapuram is significantly lower than what it is in the rest of the state. Statistically, the coefficient of variation of household incomes is 8 per cent, but the coefficient of variation of per capita income is higher at nearly 18 per cent across the three homogeneous regions of Kerala.
Kerala does quite well in terms of relative uniformity of economic well-being across the state. In this respect, Haryana and Punjab also score very well. Haryana is the best in this respect with a coefficient of variation of 4-5 per cent for household and per capita incomes. Punjab has a coefficient of variation of 9-12 per cent. Notably, all these are small states. And, they are the three richest states of the country.
The next richest state is Maharashtra. (Chandigarh, Delhi and Goa are also rich, but these are more like cities than states and I have, therefore, excluded them from this exercise.) However, Maharashtra’s claim that it is among the richer states of the country is contentious. This is because of the very high diversity of incomes of the several regions within the state.
There are 10 homogeneous regions within Maharashtra. And, the coefficient of variation of household incomes across these is as high as 41 per cent. The coefficient of variation for per capita income is even higher at 43 per cent. Compare this to the coefficient of variation of 4-5 per cent for Haryana or 9-12 per cent for Punjab. It is clear that the average number for Maharashtra is not representative of the state. This is true for many other states. Uttar Pradesh has a coefficient of variation of per household and per capita income of 28 per cent and 29 per cent, respectively. For Jharkhand, it is 29 per cent for both. For Tamil Nadu, the coefficient of variation of household and per capita income are 27 per cent and 24 per cent, respectively; while for Karnataka, they are 21 per cent and 26 per cent, respectively. Even a relatively small state like Orissa has a coefficient of variation of per household and per capita income of 27 per cent and 25 per cent, respectively. States with low disparities across regions (other than the top three states) are Andhra Pradesh and Madhya Pradesh.
These substantive intra-state variations that render the average numbers less representative of the state as a whole make most interstate comparisons meaningless. Inter-HR (homogeneous region) comparisons make a lot more sense. If we do this, we find that the richest HR is Mumbai followed by Chandigarh, Delhi and then Palakkad-Idukki of Kerala and Firozpur-Sangrur of Punjab. These regions are internally uniform and also of similar size (except Chandigarh which is much smaller than the rest).
The poorest is the Malkangiri-Nayagarh region in southern Orissa. This is followed by Cuttack-Mayurbhanj also in Orissa and Madhepur-Kishanganj and Gaya-Banka in Bihar. But, there are several parts of Orissa and Bihar that are quite well off. Baleshwar-Gajapati in Orissa is nearly twice as rich as Malkangiri-Nayagarh. Similarly, in Bihar, the Siwan-Muzaffarpur region is as rich as the Coimbatore-Dharamapuri region of Tamil Nadu.
There is much to learn from this diversity, to strategise marketing plans and chart out developmental policies. Interstate comparisons make a travesty of India’s rich diversity.
The author is managing director and CEO, Centre for Monitoring Indian Economy
