Decoding India’s GDP base revision

Why in the News

India’s nominal Gross Domestic Product (GDP) has been revised down by roughly 3 percent across the three years in which the old and new series overlap, under the New GDP Series with base year 2022-23. The Ministry of Statistics and Programme Implementation (MoSPI) set out the methodological improvements and updated data sources behind the revision when it released the series, along with a comparative table giving activity wise revisions and their reasons. The principal driver is a better measurement of India’s unincorporated services sector, which the earlier series estimated by carrying benchmark figures forward on proxy indicators. The contested point is whether a lower headline number means a smaller economy or only a better measured one.

What is a GDP base year revision?

1. Base year: It is the reference year whose price structure and economic composition the national accounts are built on, so every later estimate is expressed against that year’s conditions.
2. What a rebasing changes: It updates the data sources, the coverage and the methods together, so it changes the estimated rupee size of the economy and not merely the growth rate.
3. Direction is not fixed: International statistical practice recognises that the estimated size of an economy can move up or down after a rebasing, depending on what the new data and methods reveal.
4. India’s current shift: The base has moved from 2011-12 to 2022-23, with three overlap years across which the two series can be compared directly.

How large was the revision, and over which years?

1. Year wise cuts: Nominal GDP was revised down by about 2.7 percent in 2022-23, 3.5 percent in 2023-24 and 3.8 percent in 2024-25.
2. An independent estimate: The World Bank’s India Development Update of April 2026 put the cut at 3 to 4 percent in each of the four years from FY23, attributing it mainly to a reassessment of the informal economy.
3. Volatility fell in the new series: The same update found quarterly growth between FY 2023-24 and FY 2025-26 to be less volatile and more broad based than previously estimated.
4. Size is not activity: A lower estimate does not mean the economy became smaller or slowed in those years, since part of the change is simply a different and better measured starting number.

Which sectors were revised up, and which down?

1. Agriculture and allied activities: Revised up by about 3.8 to 5.9 percent.
2. Financial services, real estate, professional services and ownership of dwellings: Revised up by roughly 7.8 to 9.0 percent over comparable years.
3. Trade, transport and storage: Revised down by around 23 to 26 percent, the sharpest movement in the exercise.
4. Trade and road transport in detail: Trade Gross Value Added (GVA), the value an activity adds before product taxes and subsidies, was cut by 36 percent and road transport by 16.9 percent.
5. Hotels and restaurants: Revised up by 5.7 percent, mainly on the revised estimates for the unincorporated sector.

Why did the unincorporated sector drive the change?

1. The old method: In the 2011-12 series the unincorporated sector was estimated by moving benchmark estimates forward with proxy indicators, so the sector’s actual size was never measured afresh between benchmarks.
2. The new inputs: The new series uses the Annual Survey of Unincorporated Sector Enterprises (ASUSE), which enumerates unregistered non farm enterprises, and the Periodic Labour Force Survey (PLFS), which measures employment and how it is distributed across enterprise types.
3. Direct measurement: Together these give a direct basis for measuring the sector instead of an extrapolation anchored to an ageing benchmark.
4. The correction is not uniform: Revisions within the unincorporated sector vary from activity to activity rather than moving in one direction.

Why did a single year’s revision carry into later years?

1. How the estimates are built: India’s quarterly and provisional GDP estimates are constructed from the previous year’s quarterly figures.
2. The updating indicators: Those figures are then updated using information such as Goods and Services Tax collections and industrial production.
3. The carry forward: Once the 2022-23 estimate was revised under the new methodology, every subsequent annual and quarterly estimate moved down with it as a matter of arithmetic.

How common is a rebasing revision across other economies?

1. Nigeria and Indonesia, 2014: Both rebased their national accounts and both saw their previously estimated nominal GDP levels revised.
2. Brazil, 2015, and South Africa, 2018: Each rebasing likewise produced a revision to the previously estimated level of nominal GDP.
3. Mexico, 2019, China, 2021, and Spain, 2024: All three changed their previously estimated nominal GDP on rebasing.
4. India’s own precedent: The earlier shift from base year 2004-05 to 2011-12 also changed the estimated size of the Indian economy.
5. What the set can bear: These are cited as country and year only, without the methodological detail that would allow a like for like comparison, so they establish that revision on rebasing is routine and nothing further.

Challenges to the new GDP series

1. Transparency of sources and methods: Independent verification of the estimates depends on a detailed Sources and Methods publication, which lags the release of the series itself. Eg. The comparative table issued with the new series gives activity wise reasons but not the underlying computation.
   The Fix: Publish the full Sources and Methods volume alongside the series release rather than months after it.
2. Deflator weakness: Real GDP is deflated largely with the Wholesale Price Index, which does not cover services, so measured real growth in services can be distorted. Eg. India has no full Producer Price Index of the kind most large economies use for deflating output.
   The Fix: Complete the Wholesale Price Index base revision and introduce a Producer Price Index for deflating services output.
3. Residual extrapolation in the informal economy: ASUSE and PLFS improve coverage, but a portion of informal activity is still estimated rather than enumerated. Eg. Enterprises that operate seasonally or from a dwelling are the hardest to capture in an establishment survey.
   The Fix: Run ASUSE on a fixed annual calendar and publish its enterprise coverage rate, so the extrapolated share is visible to users.
4. Irregular rebasing intervals: Uneven gaps between base years let the series drift away from the actual structure of the economy between revisions. Eg. The 2011-12 base remained in use for well over a decade, through a period of rapid digitisation and sectoral change.
   The Fix: Institutionalise a base year revision every five years, which is the international practice.
5. Institutional independence: Confidence in the numbers rests on the statistical system being visibly insulated from the government of the day. Eg. Past resignations from the National Statistical Commission and the withholding of completed survey results drew attention to exactly this.
   The Fix: Give the National Statistical Commission a statutory basis, so decisions on methodology and release are not administrative ones.

Conclusion

A statistical system is judged by whether it changes its numbers when better evidence arrives, not by whether the numbers hold still. The unresolved half of this exercise sits on the price side: coverage of output has improved while the indices used to convert output into real terms have not been rebuilt to match. The next marker is whether the promised documentation of sources and methods arrives in a form that lets independent researchers reproduce the estimates rather than only read the reasons for them.

Back2Basics: National Statistical Commission

1. Nature: It is the apex advisory body on India’s official statistical system.
2. Origin: It was set up in 2005 by a government resolution, following the recommendation of the Rangarajan Commission on statistics, and has no statutory backing.
3. Composition: It has a part time Chairperson, four part time members, the NITI Aayog Chief Executive Officer as an ex officio member, and the Chief Statistician of India as Secretary.
4. Mandate: It advises on statistical priorities, standards and survey design, and its recommendations are given effect through the Ministry of Statistics and Programme Implementation.

Matching Previous Year Question

“Explain the difference between computing methodology of India’s Gross Domestic Product(GDP) before the year 2015 and after the year 2015.”