India September 2026 CPI Inflation Rises to 5.54%; Food Hits 7.74%

India’s consumer-price inflation rose to 5.54% in September 2026, up from the final August reading of 4.83%. Food inflation rose more sharply, from 5.97% to 7.74%, according to the National Statistics Office (NSO) release published on 12 October. The rural inflation rate was 6.08%, while the urban rate was 4.88%.

For households, the figures show another loss of purchasing power. For finance teams, the more useful question is where that pressure is concentrated. The September numbers contain several different stories: fast-rising food prices, very different changes across individual products, and a significant rural–urban gap. They also arrived after the Reserve Bank of India’s October monetary-policy meeting, a point that matters when interpreting the policy response.

The authoritative starting point is the NSO/MoSPI September 2026 CPI release, including its national indices and annexures. September readings are provisional; the August comparisons below are final.

How much did inflation increase between August and September?

The annual inflation rate increased by 0.71 percentage points at the all-India level. Food inflation rose by 1.77 percentage points. Rural inflation rose faster than urban inflation on both headline and food measures.

These are year-on-year inflation rates: they compare prices with the corresponding month a year earlier. The change from 4.83% to 5.54% is a change in an inflation rate, not proof that prices increased by 0.71% during September. A reader checking earlier headlines may encounter a slightly different initial August number. For this comparison, the September release’s explicitly labelled August final series is the consistent baseline.

Onions were up 108%, but the CPI basket did not rise by 108%

NSO’s most striking item reading is onion inflation of 108.09% year on year. That means the price index for that item was more than twice its level a year earlier. But the broader inflation rate depends on how different items are weighted and how their prices change. One extreme item cannot be read as a percentage-point contribution to headline inflation.

The release provides the following selected item rates and published CPI weights. The weights are included so readers can distinguish how fast an item’s price changed from how much that item represents in the consumption basket.

For perspective, onion inflation was 48.78% in the August final column of the September release before rising to 108.09% in September. Garlic inflation accelerated from 43.60% to 54.62%. But ginger slowed from 73.82% to 62.85%, and silver jewellery slowed from 106.81% to 87.80% while remaining exceptionally high. Even the frequently repeated description “vegetable prices rose” hides very different movements: potato and tomato still had negative year-on-year readings.

The published weight is not itself an inflation contribution. Precisely decomposing the CPI change requires the relevant underlying index, weights and calculation method; multiplying a rounded year-on-year percentage by a displayed weight may not reproduce the official overall figure. For a procurement or margin review, the useful next step is to map the company’s actual purchase basket to these movements rather than assuming its costs changed at the national food-inflation rate.

Some service prices were rising differently from the food basket

The food index moved faster than the headline CPI, but the picture was not confined to supermarket purchases. The release’s group-wise figures put food and beverage serving services—a category relevant to eating out—at 9.39% in September. Clothing was at 4.25%, while actual rental payments made for housing were at 2.37%. Those are distinct published CPI categories, not an estimate of “core inflation”.

This matters when discussing whether inflation is broadening. Different expenditure categories can face different pressures, and one headline number cannot resolve whether businesses have raised prices because of wages, imported materials, energy, distribution costs or demand. A review of category and time-series data is needed to make that distinction. The MoSPI eSankhyiki CPI catalogue is the government’s deeper data source for researchers comparing series and expenditure groups.

Rural inflation ran 1.20 percentage points above urban inflation

Rural headline inflation of 6.08% exceeded the urban reading of 4.88% by 1.20 percentage points. Rural food inflation, at 8.06%, was 0.88 percentage points above the urban food reading of 7.18%. Both populations experienced faster annual inflation than in August; the acceleration was larger in the rural series.

For a company with separate rural and urban customers, these are reasons to examine affordability and product mix separately. Consider two sales regions with the same nominal revenue growth. If essential expenses have risen faster in the rural market, spending patterns may shift differently—but that is a hypothesis to test using actual volume, selling-price and customer data, not a result established by the CPI release alone.

Another common error is to interpret the rural–urban inflation gap as proof that the absolute price level is higher in rural India. CPI inflation measures changes over time in the relevant baskets; it is not a direct cross-region cost-of-living price comparison. State-level breakdowns in the NSO annexures can be useful for local analysis, with the official warning that some item-level state estimates are based on relatively thin samples.

The 2024-base index provides a different, month-to-month comparison

India’s current CPI series is published with 2024=100. The combined headline index was 108.74 in August and 109.49 in September, a change of 0.75 index points. Comparing those rounded values gives an indicative month-to-month index increase of about 0.69% (0.75 ÷ 108.74 × 100). That is separate from the published 5.54% year-on-year inflation rate.

Why do the two measures differ? The annual figure compares September with September of the previous year, while the monthly calculation compares September with August. Base effects, seasonal patterns and recent prices can make those stories differ. The example above uses the release’s rounded index values; statistical releases can use unrounded underlying figures, so it should not be presented as an independently reported official monthly inflation rate.

For an illustration of the annual percentage only, a hypothetical unchanged basket costing ₹1,00,000 in September 2025 would cost approximately ₹1,05,540 one year later if its inflation exactly matched the all-India CPI. Actual households do not buy identical baskets, and a company’s input-price inflation will often differ from consumer CPI. The illustration is arithmetic, not an observed household expense.

The RBI had already raised the repo rate before these figures were released

The chronology deserves care. On 7 October 2026, the RBI’s Monetary Policy Committee increased the repo rate by 25 basis points to 5.50% and changed its stance to calibrated tightening. The official October MPC resolution (Press Release 2026-2027/1264) records that decision. The NSO’s September CPI release followed on 12 October. It is therefore wrong to say that the published 5.54% September reading caused a rate decision taken five days earlier.

The fresh CPI print nonetheless adds evidence for later monetary-policy assessments. At 5.54%, September inflation was above the 4% medium-term target and below the 6% upper limit of the RBI’s tolerance band. That is relevant to future decisions, but does not establish that the next meeting will produce a particular rate change. Policymakers will consider the persistence and breadth of inflation as well as growth and financial conditions.

The historical August RBI MPC minutes help show how the central bank’s discussion developed before October’s change. They describe an earlier 5.25% repo rate, not the current October policy setting. Separately, the September household inflation-expectations survey launch concerns how RBI gathers expectations; its announcement is not a new measured CPI figure or a policy-rate decision.

What is actionable for a finance or audit team?

A finance team can use this release without turning a macroeconomic statistic into a blanket pricing rule. Three concrete applications are especially relevant.

• Budget assumptions: Label the source, series, geography, date and provisional/final status whenever CPI is included in a forecast. Use actual payroll, lease, input-cost or customer data rather than applying 5.54% indiscriminately to every expense line.
• Gross-margin analysis: Where the business buys food products, compare invoice-level price movements, supplier arrangements, inventories and sourcing regions against relevant item data. The onion reading is a reason to investigate exposure, not proof that all materials costs doubled.
• Management reporting and audit evidence: If management attributes a sales, pricing, impairment or cost variance to inflation, reconcile that explanation to the entity’s actual activity. A government aggregate supports macroeconomic context, not a substitute for transaction-level corroboration.

One practical control is to keep the raw index alongside any annual inflation percentage in a working paper. It makes the base year and the period being compared visible, reduces confusion between a percentage and a percentage-point change, and helps identify when an earlier provisional figure has been updated.

Next data point: the October CPI release

NSO has scheduled the October 2026 CPI release for 12 November 2026, or the next working day if that date is a holiday. The next release may indicate whether food-price acceleration persisted or eased, and it will provide another observation for comparing inflation across regions and categories.

The September collection covered 1,407 urban markets (including online markets) and 1,465 villages; NSO reported a 100% response rate for the exercise. This describes completion within the selected sample, not a survey of every market or household. It is one reason to use the national aggregates confidently for their intended purpose while treating very granular local item comparisons with more care.

For now, the September release supports a clear, bounded conclusion: annual consumer inflation accelerated to 5.54%, food inflation reached 7.74%, and rural inflation exceeded urban inflation. Its most useful implication for professionals is to separate headline inflation, category-level exposure, and already-announced RBI policy before making a forecast or explaining a financial result.