The gap between reported and felt inflation is about to widen: RBI should take note

On 12 March, when February inflation printed at 3.2%, the war in West Asia had been raging for 13 days. Oil company retail agencies, which sell cylinders of liquefied petroleum gas (LPG) at an administered price to registered users, shuttered their doors, anticipating supply cuts. Consumers were forced to turn to the informal market, where the price had risen five-fold by 12 March, relative to the regulated price even without the Ujjwala price subsidy.

Retail agencies later re-opened to long lines and police protection.There was a massive gap between the official inflation rate and the fuel price spike. This kind of dissonance between reported and experienced inflation is intrinsic to the nature of a single inflation rate, which is a weighted average over a diverse array of consumable goods and services. In the LPG case, it arose also because war in West Asia broke out during the interval between data collection and the issue date of February inflation.

Before going any further, a few comments are in order about the new Consumer Price Index (CPI) series, which made its debut in February 2026 along with the new GDP series. Both are a magnificent improvement over what they replaced. The new CPI covers 358 items, up from 299 in the old series.

For each of these 358 items, the ‘unit-level’ weights (which then get aggregated upto divisions like food and beverages) were obtained from the national Household Consumption Expenditure Survey of 2022-23. I salute the team that generated the unit-level weights separately and additively for rural and urban sectors for each of 28 states and eight Union territories, summing to 100 at the national level across all items. These weights yield a portrait of what we consume based on where we