India measures its national output three different ways. It measures each State's output one way.
On 9 October 2026 the National Statistics Office (NSO), under the Ministry of Statistics and Programme Implementation (MoSPI), released a draft guideline for compiling Gross State Domestic Product (GSDP) from the expenditure side, on the revised base year 2022-23. Comments are open until 28 October 2026. It is a methodology document, not an estimate, and it will not produce a headline number this year or next.
It is also the most consequential statistical release of the week, because of one sentence buried in the middle of it: at present, some States and Union Territories compile estimates of fixed capital formation and government consumption, but Private Final Consumption Expenditure is not compiled by any State or Union Territory at all.
We do not know what the households of any Indian State spend.
Three ways to count the same thing
Output can be measured at three points in its life, and the three must in principle give the same answer.
The production approach adds up value added — what each enterprise sells minus what it bought in to make it. Sum it across agriculture, industry and services and you get Gross Value Added; add product taxes and subtract product subsidies and you get GDP.
The income approach adds up what that production paid out: wages and salaries, operating surplus, mixed income of the self-employed, consumption of fixed capital.
The expenditure approach adds up who finally bought the output:
PFCE + GFCE + GFCF + change in inventories + valuables + net exports
- PFCE — Private Final Consumption Expenditure — households and non-profits buying for their own use.
- GFCE — Government Final Consumption Expenditure — the government's own purchases of goods and services, essentially the running cost of the state.
- GFCF — Gross Fixed Capital Formation — investment in assets that last: machines, buildings, roads, software.
- Change in inventories — output produced but not yet sold.
- Valuables — gold, jewellery, precious stones held as a store of value rather than consumed. India is one of the few economies where this line is large enough to matter.
- Net exports — exports minus imports.
One rupee of output is one rupee of somebody's income and one rupee of somebody's spending. The identity is not a theory; it is bookkeeping. What makes it useful is that the three are compiled from independent data sources, so the extent to which they fail to agree is itself a measure of how good the statistics are. That is why the Indian national accounts publish a line called discrepancies: the gap between the expenditure-side total and the production-side total, left visible rather than forced to zero.
At the national level India does all three. At the State level it does the first, sector by sector — the older documented practice was production approach for agriculture, forestry, fishery, mining and registered manufacturing, and income approach for the rest. That is what a GSDP figure is. It tells you where output happens. It is silent on where it is absorbed.
Why production and consumption part company
A State's production and its residents' consumption are two different things, and in a large federal economy they can be very far apart.
A coastal State that hosts a refinery books enormous output. Its households do not drink the diesel. A State with a large pharmaceutical cluster produces medicine sold across the country. A hill State with little industry may have household consumption well above its own production, financed by remittances, transfers and tourism receipts.
Production-side GSDP measures the first thing. Almost every policy question that interests a finance commission, a welfare ministry or a State budget officer depends on the second. Household consumption is what determines poverty, nutrition and the incidence of an indirect tax. It is also, after the shift to GST, the base on which a State's own consumption-tax revenue actually rests — GST is a destination-based tax, and destination is consumption, not production.
So the gap matters. We have been running a destination-based tax system for nine years without a State-level measure of destination consumption.
The problem that makes this genuinely hard
There is a reason no State compiles PFCE, and it is not laziness.
Look again at the identity. The last term is net exports — and for a State, "exports" means goods and services sold outside the State, which includes everything sold to other Indian States. There are no customs posts between Karnataka and Maharashtra. Nobody records a truck of cement leaving one and arriving in the other as a trade transaction. Inter-State trade is the largest unmeasured flow in the Indian economy.
This is not unique to India. Australia publishes State-level expenditure accounts, and its statistical agency is explicit that State expenditure must equal final demand plus exports minus imports — whether international or interstate — plus the change in inventories. It is also explicit about the compromise: it measures that goods trade on a merchandise-trade basis rather than the conceptually correct balance-of-payments basis, because that is the data that exists. A well-resourced statistical system with decades of practice still has to settle for the achievable version.
India's draft guideline reaches for the same solution, and names it plainly. Where direct State-level data exist, use them. Where they do not, allocate a national total across States using an allocation indicator.
That is the sentence to read carefully, because allocation is where a measurement can quietly become an assumption. If national PFCE is distributed across States using an indicator that is itself a function of State income, the resulting "estimate" of State consumption will track State production by construction — and the whole point of compiling the expenditure side was to learn something the production side did not already tell us. The guideline's own emphasis on State-specific data sources is the right instinct. Whether the final document holds that line against the convenience of national allocation is the thing to watch when the comments close on 28 October.
Where this sits in the base-year revision
This is one piece of a larger rebasing exercise, and it helps to see the sequence.
MoSPI has moved the national accounts from base year 2011-12 to 2022-23, with the new GDP series released on 27 February 2026. The Consumer Price Index moved to a 2024 base on 12 February 2026, and the Index of Industrial Production was scheduled for May 2026. The Wholesale Price Index had already been rebased to 2022-23, which is why the national accounts and the wholesale price series now share a base and can be used together without splicing.
The new base brought methodology with it: integration of administrative data — GST returns, PFMS, e-Vahan vehicle registrations — alongside survey data, and a shift from single to double deflation, which deflates inputs and outputs by their own separate price indices instead of applying one deflator to value added. The 2022-23 base was recommended by the Advisory Committee on National Accounts Statistics, a 26-member expert body constituted in June 2024 under the economist Biswanath Goldar. The full methodological account is to appear in the Sources and Methods publication.
A base-year revision at the national level has to cascade to the States, because a GSDP series on an old base cannot be added up or compared against a national series on a new one. The expenditure-side guideline is part of that cascade — and MoSPI has chosen to use the rebasing to add a capability rather than merely renumber an existing one.
The institution that has to do the work
GSDP is not compiled in Delhi. It is compiled by the State Directorates of Economics and Statistics (DES) — one in each State and Union Territory — working to a common methodology issued by NSO.
This is the part of the Indian statistical system that gets the least attention and carries a great deal of the load. The entire architecture described in the guideline depends on State DES staff being able to execute it: identify State-specific data sources, apply allocation indicators consistently, and produce something comparable with 35 other jurisdictions doing the same thing. The release says so in as many words, describing the exercise as strengthening the statistical capacity of State DESs.
That is an honest admission of where the binding constraint lies. The methodology can be written in Delhi in a year. The capacity to apply it uniformly across every State cannot. This is the same institutional story as the statistical system Mahalanobis built — a national design that works only as well as its weakest State office.
What has and has not happened
A draft guideline is a draft guideline. No State has compiled expenditure-side GSDP. No PFCE estimate for any State exists. No date has been announced for one.
What has happened is that the framework now exists, it is public, it is open to comment until 28 October 2026, and it has been issued on the same base year as the national series so that the two can eventually be reconciled. On the strength of the environmental-economic accounts the same ministry has committed to building, the pattern is recognisable: MoSPI publishes the method, invites comment, and then builds the series over several years.
The useful question for an examination answer is not whether this is welcome. It is what the expenditure side can tell you that the production side cannot — and why the answer involves trucks crossing State borders with nobody counting them.
🔑 Revision block
- What: NSO/MoSPI released a draft guideline for compiling GSDP from the expenditure side, base year 2022-23, on 9 October 2026; comments open till 28 October 2026.
- The gap it addresses: some States/UTs compile GFCE and GFCF; no State or UT compiles PFCE. State accounts are production- and income-side only.
- Expenditure identity: PFCE + GFCE + GFCF + change in inventories + valuables + net exports.
- Three approaches: production (value added), income (wages, operating surplus, mixed income), expenditure (final purchases). Independently compiled; the gap appears as discrepancies in the accounts.
- Why it is hard for a State: "net exports" includes inter-State trade, which nobody records — there are no customs posts between States.
- The method where data are missing: allocation indicators distributing a national total across States — with the risk that the indicator imports the assumption being tested.
- Base-year context: GDP rebased 2011-12 → 2022-23, new series 27 February 2026; CPI to 2024 base 12 February 2026; WPI already on 2022-23; double deflation adopted; base recommended by ACNAS under Biswanath Goldar.
- Who compiles it: State Directorates of Economics and Statistics, to a common NSO methodology — the stated aim is strengthening their statistical capacity.
- Exam hook: GST is a destination-based tax; destination is consumption. A production-side-only State account cannot measure the base of the tax the State collects.
🎯 Practice MCQs
Q1. The draft guideline released by the National Statistics Office on 9 October 2026 relates to the compilation of: (a) GSDP estimates from the expenditure side (b) GSDP estimates from the income side (c) District Domestic Product estimates (d) Gross Value Added by registered manufacturing
→ (a) The guideline covers expenditure-side GSDP on base year 2022-23. State accounts are already compiled on production and income approaches.
Q2. Which expenditure component is currently not compiled by any State or Union Territory? (a) Government Final Consumption Expenditure (b) Gross Fixed Capital Formation (c) Private Final Consumption Expenditure (d) Change in inventories
→ (c) MoSPI records that some States/UTs compile GFCE and GFCF, but PFCE is compiled by none. This is the specific gap the guideline is meant to close.
Q3. In the expenditure approach to measuring domestic product, "valuables" refers to: (a) Government holdings of foreign exchange (b) Gold, jewellery and precious stones acquired as a store of value (c) The market value of listed equity (d) Land and buildings held by households
→ (b) Valuables are acquisitions held as a store of value rather than consumed or used in production. The line is unusually large in India.
Q4. The base year of the new national accounts series released on 27 February 2026 is: (a) 2011-12 (b) 2017-18 (c) 2024-25 (d) 2022-23
→ (d) GDP moved from 2011-12 to 2022-23. The Wholesale Price Index had already been rebased to 2022-23; CPI moved to a 2024 base.
Q5. The principal measurement difficulty in compiling expenditure-side GSDP for an Indian State is: (a) Inter-State trade flows are not directly recorded (b) States do not maintain budget documents (c) Household surveys are not conducted in India (d) Product taxes cannot be separated from subsidies
→ (a) Net exports for a State include sales to other States, and there is no border recording of inter-State movement of goods and services.
Q6. "Double deflation", adopted in the revised national accounts, means: (a) Deflating nominal GDP twice in successive years (b) Applying both a wholesale and a retail price index to the same aggregate (c) Deflating inputs and outputs by separate price indices (d) Adjusting for inflation and for population growth together
→ (c) Double deflation deflates output and intermediate input separately, instead of applying a single deflator to value added.
Q7. Gross State Domestic Product estimates in India are compiled by: (a) The Reserve Bank of India (b) State Directorates of Economics and Statistics, to a common NSO methodology (c) The Finance Commission (d) The Comptroller and Auditor General
→ (b) State DESs compile the estimates; NSO issues the methodology to keep them comparable across States.
Q8. The "discrepancies" line in the Indian national accounts represents: (a) Errors detected in State budget documents (b) The difference between nominal and real GDP (c) Unreported income in the informal sector (d) The gap between independently compiled expenditure-side and production-side totals
→ (d) Because the approaches use independent data sources, their totals differ; the gap is published rather than suppressed.
Q9. The body that recommended 2022-23 as the new base year for the national accounts is: (a) The Advisory Committee on National Accounts Statistics (b) The National Statistical Commission's Standing Committee on Prices (c) The Fifteenth Finance Commission (d) The Central Statistics Advisory Board
→ (a) ACNAS, a 26-member expert body constituted in June 2024 under Biswanath Goldar, recommended the base year and the new data sources.
Q10. Why does a production-side-only State account sit awkwardly with GST? (a) GST is levied on production, which State accounts do not measure (b) GST is destination-based, so its base is consumption, which production-side accounts do not measure (c) GST revenue is not shared with States (d) GST applies only to inter-State supplies
→ (b) GST accrues where goods and services are consumed. A State account that measures only output cannot measure the base from which the State's own consumption-tax revenue arises.
📋 How this gets asked (PYQ pattern)
Statistical-system questions have a settled shape in CDS and OTA papers, and this release sits squarely inside it.
The first pattern is the identity. Candidates are asked to pick the component that does not belong in the expenditure approach, or to complete the list. Learn the six terms in order and learn which two are distinctively Indian preoccupations — valuables, because household gold is large, and net exports, because at the State level it is the term that cannot be measured.
The second is institutional attribution. Who compiles GSDP? Who issues the methodology? Who recommended the base year? Papers reliably test the difference between the body that designs a statistical product and the body that produces it. NSO writes the guideline; State DESs compile the estimates; ACNAS advised on the base.
The third is the base-year chain. Three series have moved in a short span — GDP to 2022-23, CPI to a 2024 base, WPI to 2022-23 — and a question that mismatches a series with its base is easy to set and easy to get wrong. Keep them as a short table.
The fourth is the one that separates a written answer from a recalled list: explaining what the expenditure side adds. The production side tells you where output occurs; the expenditure side tells you where it is absorbed. In a federal economy with a destination-based consumption tax and large inter-State flows, those are different maps, and the policy questions that matter — poverty, nutrition, tax incidence, transfers — sit on the second one. A candidate who can say that in three sentences has understood the release. A candidate who recites PFCE, GFCE, GFCF has only read the headline.
Preparing for CDS/OTA? Statistical releases look dry and are the cheapest marks in the paper, because the facts are finite and stable: who compiles, on what base, by which method. Build the base with our CDS/OTA study material and the economy section, follow the daily CDS current affairs, and prepare with our faculty in the upcoming Cavalier courses in Delhi.
✍️ Written by Aditya Tiwari — Faculty, Economy & Polity, at The Cavalier. Reviewed by the Cavalier Faculty Desk.