New GDP methodology: Most people do not need to know how GDP is calculated. But one distinction matters when manufacturing growth is being measured: an increase in the value of factory output can reflect higher production, higher prices, or both. If the objective is to measure real growth, the price effect has to be removed.
That is the issue behind one of the more important changes in India’s new GDP series. The Ministry of Statistics and Programme Implementation (MoSPI), which shifted the national accounts to a 2022-23 base year in February, has adopted double deflation for manufacturing. The methodology was set out in greater detail on September 21 when MoSPI released its Sources and Methods for Compilation of National Accounts Statistics. The new series was also updated in August to incorporate the new 2022-23 Output Producer Price Index (PPI) and Index of Industrial Production.
READ | India’s GDP data debate goes beyond the headline
The change addresses a weakness in the earlier method. Under single deflation, the nominal gross value added (GVA) of manufacturing was adjusted using a single broad price measure. But manufacturing output and the inputs used to produce it do not necessarily experience the same price changes. If the price of steel, energy or chemicals rises much faster than the selling price of the manufactured product, using one index for both can give a misleading picture of real value added.
Why manufacturing needs two price measures
Consider a factory that sells goods worth ₹1,000 crore and uses ₹800 crore of raw materials, electricity, packaging and other intermediate inputs. Its nominal GVA is ₹200 crore.
Suppose the price of its output rises by 5%, while the prices of its inputs rise by 10%. Applying the same price index to both sides would fail to capture the change in the relationship between what the factory produces and what it has to buy to produce it.
Double deflation treats the two sides separately. Output is first converted into real terms using an appropriate output price index. Intermediate consumption is separately adjusted for price changes. Real GVA is then derived as real output minus real intermediate consumption. MoSPI describes the approach as:
Real GVA = Deflated gross value of output − Deflated intermediate consumption.
The distinction is important because one industry’s output can be another industry’s input. Steel is an output of the steel industry and an input for automobile and machinery manufacturers. Cotton yarn is an output of spinning mills and an input for textile manufacturers. A price measure that captures only final products would therefore be inadequate for calculating the real value added of manufacturing.
The new methodology addresses this by using a detailed Output PPI compiled by the Office of the Economic Adviser in the Department for Promotion of Industry and Internal Trade. The index includes intermediate products as well as final goods. Its basket covers items such as wheat, milk, bauxite, coking coal, electricity, refined palm oil, cotton yarn, cement and semi-finished steel. This allows the price of an item produced by one industry to be used in measuring the price of that item when it enters another industry’s production process.
READ | GDP growth numbers need scrutiny, not shortcuts
The distinction is worth making because an experimental Input PPI does exist. The Office of the Economic Adviser began publishing a trial Input PPI for manufacturing in March 2026. But MoSPI has said that this trial index was not used to compile the new national accounts. Instead, it maps the input structure obtained from the ASI to relevant items in the Output PPI.
This is a more granular exercise than applying one aggregate deflator to manufacturing. MoSPI ranks items by their contribution to output and intermediate consumption and selects those accounting for 80% of the respective values. Their current-price values are then allocated using weights derived from the ASI and deflated separately using the relevant item-level Output PPI.
New GDP methodology: Double deflation has limits
The change improves the methodology, but it does not eliminate the dependence of GDP estimates on the quality of price and production data.
India’s PPI system itself is still developing. The 2022-23 Output PPI is now being used in the national accounts, while the Input PPI remains a trial series. Producer price indices for services are also not available for every service purchased by manufacturers. Where an appropriate producer price index is unavailable, MoSPI uses other price measures, including consumer prices and implicit deflators.
There are also two manufacturing categories in which double deflation is not being applied. MoSPI has excluded manufacture of food products, including processed food and oils, and pharmaceuticals because the share of imported inputs is high and mapping those inputs to the available item-level Output PPI is difficult. Real GVA for these categories is estimated using volume extrapolation instead. Double deflation therefore covers 28 of the 30 manufacturing compilation categories, rather than the entire sector.
Imported inputs present a broader problem. An Indian manufacturer may buy a raw material from China at a price that moves differently from the price of a comparable product made in India. A domestic producer price index cannot automatically capture that difference. Better measurement of manufacturing GVA will therefore depend partly on improving information about the prices actually paid for imported and domestic inputs.
This is also why double deflation should not be treated as a device that produces a single, unquestionable measure of manufacturing performance. Its advantage is that it recognises that output and input prices can move differently and incorporates that difference into the estimate.
READ | India GDP growth: Private capex is the real test
What a negative manufacturing deflator can mean
The new method can produce results that appear counterintuitive.
Manufacturing GVA in real terms can increase even when the prices of both manufactured goods and their inputs are rising. The reason is that the implicit price deflator for GVA reflects the difference between the price movements of output and intermediate consumption. If input prices rise faster than output prices, the price component of value added can decline even though both sets of prices are increasing. MoSPI itself has pointed out that this can produce negative growth in the implicit GVA deflator.
This measure should not be confused with consumer inflation. A negative manufacturing GVA deflator does not mean that manufactured goods across the economy have become cheaper. Nor does it mean that consumers are experiencing deflation. It reflects the net price movement associated with the value added by manufacturers after output and intermediate consumption have been separately adjusted.
That distinction will become increasingly important as analysts compare manufacturing GVA with the Index of Industrial Production, wholesale prices or consumer prices. IIP is a volume index and is therefore more directly comparable with gross value of output at constant prices than with GVA. GVA also incorporates the changing relationship between output and intermediate inputs, which IIP does not measure.
Better GDP estimates will still depend on better data
The manufacturing methodology is part of a much wider revision of India’s national accounts. The new series uses 2022-23 as the base year and incorporates additional administrative and survey data, including GST records, e-Vahan, the Public Financial Management System, the Annual Survey of Unincorporated Sector Enterprises and the Periodic Labour Force Survey. It also brings the national accounts more closely into a Supply and Use Table framework intended to improve consistency between production, expenditure and income estimates.
The significance of double deflation, therefore, lies less in producing a dramatically different number than in changing how the number is constructed. Manufacturing GVA is being measured with greater attention to the price behaviour of both what factories sell and what they buy.
That does not settle every question about GDP measurement. The quality of the estimate will continue to depend on the coverage, timeliness and representativeness of the underlying data. It will also take the forthcoming historical back series to show how far the new methodology changes the measured path of India’s growth. MoSPI currently expects that back series by December 2026.
The useful test of the new system will therefore be the quality of the data it produces over time. Double deflation gives statisticians a better framework for measuring manufacturing value added. Whether it delivers a materially better picture of the economy will depend on how well India can fill the remaining gaps in price and input data.