By PC Mohanan, Former Member, National Statistical Commission
Rebasing macroeconomic indicators are routine exercises in all official statistical systems. The rebasing is needed to account for the changes in the underlying structure of an economy or consumption, production, etc. over time. The dearth of such exercises in recent years in India was a major criticism of national statistics. The continued use of outdated base years for key indicators is also an issue that the International Monetary Fund has repeatedly flagged during their annual consultations.
We did not have a retail price index covering the entire population till 2010. There were indices representing specific population segments like industrial workers, agricultural labour, rural labour, and urban non-manual employees. Except for the last index, the others are still in use. Inflation was computed from the Wholesale Price Index (WPI) released by the ministry of commerce and industry on a weekly basis. Since the monthly Consumer Price Index (CPI) was released with 2010 as the base year, headline retail inflation is derived on a monthly basis. The CPI is released separately for rural, urban, and combined segments at the state and national levels. Group and sub-group level indices are also made available. The CPI also finds use as a deflator in the GDP.
Clearly, the release of the CPI with an updated base year is a much-awaited, welcome step. Though the index comes out as a single number, much efforts go into the selection of items and their weights besides the continuous collection of price data from selected markets and other sources for its regular computation.
The CPI measure changes over time in general with the prices of goods and services that households acquire and consume. The government and the Reserve Bank of India use it widely as an indicator of retail inflation. It plays a critical role in inflation targeting and monitoring price stability, besides indexing dearness allowance of employees to neutralise price changes in their wages and salaries. The behaviour of the new index will be closely watched by all.
The just-released CPI with 2024 as the base year has many improvements over the current series which had a base year of 2012 and was used since 2013. For the previous index, the distribution of weights for different items of consumption was based on the household consumer expenditure survey of 2011-12. In case of the new index, the 2023-24 survey provided the weights and item basket. The government had junked the household expenditure survey of 2016-17 due to quality concerns.
The expert group that carried out the base revision has done a good job of reviewing the methodology followed till now, and it has addressed many shortcomings of the old index such as those relating to house rent, online purchases, and improved commodity groupings. As recent household expenditure surveys reveal, food-heavy consumption has been changing over the years. The share of food in average monthly per capita expenditure is down from 53% in 2011-12 to around 47% in 2023-24 in rural areas. In urban areas, the share is down to 40% from 43%. The drop in cereal consumption is even steeper. Using the 2023-24 survey, in the new CPI the weights for food and beverages is now down from 54% to 42% in rural areas and from 36% to 30% in urban areas. In the new index, a reduced weightage for food would also reduce the volatility caused by changes in food prices.
The new CPI has a much expanded list of items and markets including 12 online platforms for price data collection. It has excluded social transfers and made house rent data collection more objective. One of the basic problems faced in price data collection on a continuous basis is the change in product specifications over time, which is proposed to be addressed via structured product descriptions. Most importantly, the formula used for item-level index is a chain-based short index formula that does not require base-period prices for calculating the index every month. The index is compiled using the current month’s prices, the previous month’s prices, and the previous month’s index. This change could impact comparability with the previous index in the long term.
Broadly, the new index shows a slightly higher level of inflation. The CPI-based inflation in December 2025 was 1.33% while it is 2.75% for January 2026 in the new index. Interestingly, unlike in the previous index, the rural-urban differential in inflation appears to have come down for most item groups. While the all-India weighting diagram from the survey would be robust, it may not be the case for states due to smaller survey sample sizes. For example, the inflation rates for Kerala used to be an outlier in the last series due to very high weights for groups such as gold and personal effects. In the new series, the January inflation for Kerala is only 3.67% compared to the 9.49% in December 2025. For Telangana, the current inflation rate is 4.92% while it was only 1.77% in December. At the state level, one needs to be more careful in interpreting inflation rates.
The efforts to make the CPI a more relevant and effective indicator of price changes is welcome. But one also notices that for many of the underlying computations for the new CPI series, information from the 2011 census forms the backdrop in the absence of any census since then. This will certainly diminish the efforts to make the CPI more relevant to the present. In fact, this issue would be much more serious while revising the GDP, as it needs quantum estimates where population is used to inflate individual numbers. In contrast, the inherent simplicity of the price index has less space for controversies.
Disclaimer: The views expressed are the author’s own and do not reflect the official policy or position of Financial Express.