There are large inconsistencies between GDP and other economic indicators: says Garg

Why in the News

A former Finance Secretary has questioned the credibility of India’s latest Gross Domestic Product (GDP) estimates. The objection is not to the level of growth reported but to the absence of a transparent bridge between the old 2011-12 base series and the new 2022-23 base series. The new series has cut the size of the economy for 2024-25 by ₹12.70 lakh crore. The Ministry of Statistics and Programme Implementation (MoSPI) has explained the reduction as the result of a new methodology, wider coverage and improved data. Wider coverage normally raises the nominal size of an economy rather than reducing it. That is the inconsistency now in dispute.

What is the 2022-23 base year GDP series?

1. The base year: The base year is the reference year whose price structure is used to strip inflation out of nominal output. Real growth is measured against that fixed set of prices.
2. What the revision changes: The new series moves the base from 2011-12 to 2022-23. It also changes the data sources and the indices used to estimate output.
3. The back-series: A back-series recomputes earlier years on the new base. Without one, estimates on the old and new bases cannot be compared year on year.

Why does the new series need a back-series?

1. There is no bridge between the two series: No published concordance links the 2011-12 base estimates to the 2022-23 base estimates. A user cannot see which part of the change comes from the new base and which from the new data.
2. A published timetable is the test of intent: MoSPI has been asked to release a back-series covering 2011-12 to 2021-22 and to fix a date for doing so. The absence of any such programme indicates the issue is not being treated as pressing.

Why has a wider dataset produced a smaller economy?

1. The size of the cut: GDP for 2024-25 was reduced by ₹12.70 lakh crore. The revision to the first quarter of 2025-26 is part of that same larger change.
2. Coverage cuts the other way: Better coverage adds activity to the estimate and raises nominal GDP. A revision that widens coverage and lowers the level is unexplained by that argument.
3. An earlier overstatement is one reading: The old system may have overstated output through errors such as double counting. On this reading the new series is a correction.
4. A deliberate write-down is the other: Output may have been overstated to produce stronger growth numbers and then written down under cover of a new series. No evidence of deliberate manipulation was offered for this reading.
5. The official account is contested: The Centre’s explanation for the reduction has been described as “officialese, obfuscatory” and as shedding no light on the change.

What does the deflator gap indicate?

1. The arithmetic does not close: Consumer inflation runs above 4 per cent and producer price inflation at about 9 per cent. The GDP deflator (the economy-wide price index used to convert nominal output into real output) implied by the latest estimates is about 2.5 per cent.
2. The price data behind it is not public: The underlying price series used to build the deflator has not been disclosed. The real growth number cannot be checked without it.
3. Double deflation was applied without the data to support it: Double deflation values a sector’s inputs and its outputs at separate price indices. Indian manufacturing data is not granular enough to sustain that treatment.
4. Parallel running is the suggested safeguard: The older system should be run alongside the new one until the new methodology stabilises.

Why is the statistical system’s independence part of this dispute?

1. The divergence is not noise: Weakness in household incomes, employment, consumption and sentiment has persisted while the headline growth number has not weakened. That divergence cannot be dismissed as statistical noise, particularly where an outcome is politically sensitive.
2. The data infrastructure needs rebuilding: India’s statistical infrastructure requires massive modernisation before its outputs can be defended on technical grounds alone.
3. Freedom from political direction is the precondition: The system can produce reliable numbers only where there is no political interest in results running in a particular direction. Statisticians need greater freedom from political control for that to hold.

What does the GDP number leave out?

1. GDP is not a measure of welfare: Aggregate output says nothing about how the gains from that output are distributed.
2. The income leg is missing: India does not adequately publish the income side of the national accounts. That side shows how value added is divided between labour, corporations and government.
3. Growth alone will not lift per capita income: Per capita GDP remains low. The requirement is 9 to 10 per cent growth together with more effective redistribution and lower unproductive government expenditure.

Challenges to India’s new GDP series

1. No comparable time series exists: A rebased series without recomputed earlier years cannot support any statement about long-run growth. Eg. The 2015 shift to the 2011-12 base was followed by an official back-series only in 2018, and it revised the earlier decade’s growth rates downward.
   The Fix: Publish the 2011-12 to 2021-22 back-series alongside a documented concordance showing which data source replaced which.
2. Single deflation distorts manufacturing value added: Indian national accounts have long applied one price index to both a sector’s output and its inputs. Eg. When input prices fall faster than output prices, single deflation records a rise in real value added that did not occur.
   The Fix: Publish the separate input and output price indices used for each manufacturing sub-sector, so the deflation method can be audited.
3. The informal sector is estimated rather than measured: Output of unincorporated enterprises is extrapolated from formal-sector indicators. Eg. The MCA-21 corporate database used to estimate private corporate output was found to contain dormant and untraceable companies.
   The Fix: Anchor the informal sector estimate to the Annual Survey of Unincorporated Sector Enterprises rather than to a corporate filings database.
4. Benchmark surveys are dated or withheld: Consumption and employment weights depend on large sample surveys that are not released on a fixed cycle. Eg. The 2017-18 Consumer Expenditure Survey was withheld from publication, leaving the consumption basket anchored to 2011-12 for over a decade.
   The Fix: Fix a statutory release calendar for benchmark surveys, with the release date set independently of the government of the day.

Conclusion

The dispute is about verifiability, not about the level of growth. A national accounts estimate that cannot be compared with its own past is not a series, and no methodological note substitutes for that comparison. The statistical system settles this by publishing the recomputed earlier years and the price data behind them, not by explaining itself. Until it does, each quarterly release will be argued over rather than used.

Matching Previous Year Question

“[2021, GS3, 10 marks] Explain the difference between computing methodology of India’s Gross Domestic Product(GDP) before the year 2015 and after the year 2015.”