India’s Consumer Price Index is pro-rich; here’s why

Published: December 10, 2018 4:21 AM

Considering that necessities experience relatively higher inflation as compared to the other groups of commodities, and the plutocratic bias is particularly high in the case of necessities, it emerges that the poorest groups are far more vulnerable to inflation than the more affluent groups.

The official statistics themselves reveal substantial differences between urban, rural, inter-state and national CPI levels. (Illustration: Rohnit Phore)

By Dilip Nachane & Aditi Chaubal

It is indeed, as the common man had always suspected, the official measure of inflation does not reveal the actual inflation experienced by the poor. The official measure of inflation in India is currently the CPI, which uses a Laspeyres-type index number to estimate the changes in the price levels of a fixed quantity basket of goods over time. The basket in use now is based on the 68th Round of the National Sample Survey Organization (NSSO) Consumer Expenditure Survey conducted in 2011-12.

However, such an index suffers from a plutocratic (pro-rich) bias, as it attaches greater weightage to the expenditure pattern of the more affluent groups in comparison to that of the poorer groups. We advocate (for India) in a recent study (Nachane and Chaubal, in the forthcoming International Labour Review) the use of a democratic CPI which uses equal weightage for expenditures of all households as well as a super-democratic index (which additionally adjusts for the size of the household) .

Our findings indicate a strong plutocratic bias, with the actual inflation (as measured by a democratic index) higher by about 0.5-0.8% of the official inflation. Ideally, the inflation rate should reflect inflation faced by the median group of consumers. By contrast, the existing Indian CPI is representative of the inflation facing the higher percentiles of the sample, viz., the 72nd, 78th and the 75th upper percentiles of the rural, urban and combined samples.

Another striking fact that emerges from the calculated plutocratic and democratic weights is that the official weights at 37.5% for necessities seriously underestimate the corresponding democratic weights (45.3%), implying that the true cost of inflation in necessities is commensurately underestimated in the official figure. Further, with the exception of the low-inflation year of 2014, the inflation in necessities (smoothed monthly averages) significantly exceeds, while that of the other commodity groups falls short of, the overall official inflation rate over the recent five-year period. As necessities are overwhelmingly consumed by the poorer sections, while luxury goods mainly figure in the consumption baskets of the rich, inflation seems to hit the poor much harder relative to the rich. This means that not only are the prices of necessities rising faster than the price of commodities in general, but the “true” inflation in necessities is understated by any official index (computed from the official commodity weights).

Finally, following the Hong Kong SRA precept, separate indexes for three consumption brackets in the Indian case, viz., bottom 30%, top 10% and middle 60% were computed by us. These indicate that the weightage of necessities for the bottom 30% expenditure group is more than twice that of the top 10% group, while for luxuries, this weightage is reversed. Reflecting this weightage, necessities constitute the dominant component of the inflation experienced by the bottom 30% group, while in the case of the top 10%, luxuries are the major component of inflation. Considering the two facts, viz., (i) necessities experience relatively higher inflation as compared to the other groups of commodities and (ii) the plutocratic bias is particularly high in the case of necessities, it emerges that the poorest groups are far more vulnerable to inflation than the more affluent groups—a particularly disturbing feature for a country in which, according to the latest World Bank estimates, 456 million people or about 42% of the population live below the international poverty line of $1.25 per day.

Our analysis has serious implications for Indian monetary policy, which largely follows a flexible inflation-targeting approach. The official statistics themselves reveal substantial differences between urban, rural, inter-state and national CPI levels. Our analysis brings out different degrees of plutocratic biases in the various indices. Targeting a single index such as the national CPI effectively means making monetary policy responsive to developments in the leading industrialised states at the expense of the underdeveloped states. The concept of plutocratic bias lends an additional edge to the argument against inflation-targeting. It indicates that the official inflation figure does not really reflect the social costs of inflation, since it predominantly reflects the expenditure pattern of the more affluent sections of the population. As a matter of fact, when we develop inflation indices separately for the bottom 30%, middle 60% and top 10% of the expenditure groups, we find that in years of moderate to high inflation, the actual inflation experienced by the bottom 30% is significantly in excess of the official inflation figure, whereas the actual inflation incidence for the top 10% is lower than indicated by the official figure.

Thus, the existing paradigm of monetary policy is strongly discriminatory on two grounds—against the less developed states as well as against the poorer sections of the population. The official adoption of the democratic (and super-democratic) concepts of inflation introduced above could go a long way in correcting both these biases, as this would remove the wedge between the inflation perceived by the policymakers and the inflation actually experienced by the bulk of the country’s population.

Our analysis also has significant implications for the inflation-indexation of government transfer payments. A prime example of such a transfer is the dearness allowance (DA) of government employees, which are currently indexed by the All-India CPI for Industrial Workers (AICPI_IW). This is most likely to be replaced by the new combined CPI . In view of the plutocratic bias discussed above, a more equitable system would be to index salaries (and pensions) of Class IV government employees with a (democratic) CPI index for the bottom 30%. Similarly, the salaries of Class II and III employees may be indexed by the CPI indices for the middle 60% and top 10% respectively.

Thus, the switch from the current plutocratic indices to a democratic system of indices would eliminate a long-standing anomaly in the official estimation and put in place a more equitable policy framework.

– Nachane is chancellor, Manipur University, and honorary professor, Indira Gandhi Institute of Development Research, Mumbai, and Chaubal is assistant professor (economics), Indian Institute of Technology, Mumbai

Get live Stock Prices from BSE and NSE and latest NAV, portfolio of Mutual Funds, calculate your tax by Income Tax Calculator, know market’s Top Gainers, Top Losers & Best Equity Funds. Like us on Facebook and follow us on Twitter.

Switch to Hindi Edition