Recent discussions about the GDP series have sparked an avalanche of debate about the measurement and credibility of the national accounts. Much of it is uninformed, though there are genuine concerns that need to be addressed.
In this piece, setting aside the noise and personal insinuations, we focus on the technical aspects of the debate and address the criticisms thematically.
First, the criticism that triggered this debate was former Finance Secretary Subhash Garg’s assertion that India’s reported GDP growth of 7.8 per cent overstated the economy’s actual pace. He arrived at this conclusion by comparing the GDP series for FY 2026-27, using the revised base year 2022-23, with that for FY 2025-26, which used the base year 2011-12. This comparison is inappropriate because it compares apples and oranges.
However, the follow-up question on the 7 per cent downward revision, from Rs 86.05 lakh crore to Rs 80.0 lakh crore, is worth examining.
Also Read: GDP math under fire: What critics and fanboys are both getting wrong
What explains the GDP revision
Revisions always happen when rebasing is done, and India is no exception. For instance, in 2019, Vietnam had a positive revision of almost 25 per cent, according to some estimates.
In India’s case, the earlier exercise in 2011-12 also resulted in a downward revision. Furthermore, given that India is undertaking the rebasing exercise along with the introduction of the Producer Price Index (PPI) and an updated Index of Industrial Production (IIP) — both of which did not happen together in previous revisions — much of the revision has been loaded upfront. Historically, too, it has been the same, with subsequent quarters having only a marginal effect.
Moreover, according to Statement 1.6 in NAS 2025 and NAS 2026 (MoSPI, 2025; MoSPI, 2026), the single biggest cause of this revision is the “trade, repairs, hotels and restaurants” section, which alone contributes 87-112 per cent of the revision. In simple terms, without this segment, the revisions would have been marginal compared to the old series. This trade segment is also part of the private corporate sector, not the informal sector alone, which in part negates the next issue of informal-sector overestimation.
Also Read: Delimitation goes beyond just votes. It is also about India’s economic future
A counter to half-baked perspectives
Estimation of the informal sector and the estimated GVA has been another source of needless contention. Referring to NAS 2025 and NAS 2026 (MoSPI, 2025; MoSPI, 2026), we find that total GVA was revised down at the overlap, but only by 3.34 per cent in 2022-23 and 3.67 per cent in 2023-24.
This is an order of magnitude below the roughly 22 per cent cumulative overestimation claimed by Arvind Subramanian, Josh Felman and Abhishek Anand in their March 2026 Peterson Institute paper, or the 40.9 per cent claimed by Jatinder S Bedi and S Nagaraj in their India Forum piece last month.
An additional dimension refers to the adoption of the double deflator in the new series.
India has followed the single deflator for many years and, based on a slew of expert opinions, including the IMF’s recommendations, has decided to move to double-deflator measurement. For instance, if we are measuring the GVA of chocolate output, there are inputs such as sugar, cocoa and others, each with its own price movements. The double deflator deflates the value of output separately from the inputs, each by their own prices. This ensures a better estimation of GVA for many products.
There has been a stream of criticism of this change on the grounds that we do not have enough input measurements for double-deflator measurement, including by Pronab Sen, former chief statistician of India, in a recent interview with Business Standard. While the concern over data demands is real and merits future work, the new PPI covers inputs and outputs, and the new Index of Services Prices extends coverage to many services hitherto not directly measured. This is essentially the incremental approach recommended in 2019 by economist Bishwanath Goldar.
Subramanian criticises the double deflator on account of its poor match with CPI inflation. This is ill-informed because GDP measures value added — output net of intermediate consumption — and the deflator is a derived measure netting output and input price changes, so it is not expected to track CPI. Ironically, he had recommended the adoption of double deflation as a reform in the 2019 version of his paper.
Bolstering the GDP measurement argument is the fact that there is a slew of indicators suggesting that the economy is doing well. For instance, passenger vehicle sales growth is between 24-27 per cent, two-wheeler sales are between 14-28 per cent, and Nifty 50 profit growth for April-June is 18 per cent. All of these indicators point to a trendline of positive and healthy growth rather than tepid growth.
Lastly, is everything good with India’s statistics? Naturally, for any statistical exercise, there are methodologies and numbers that need to be revised and updated. This is best practice across the world, and India is no exception.
However, questioning the integrity of statistical authorities based on half-baked perspectives is not only counterproductive but mischievous as well. We would also be better served by a more refined discussion on this subject.
TCA Anant is Visiting Professor at ISID, Delhi, Adjunct Professor at TISS Hyderabad, and former Chief Statistician of India. Sriram Balasubramanian is an economist and best-selling author of Dharmanomics. Views are personal.
(Edited by Asavari Singh)