“Sources and Methods for Compilation of National Accounts Statistics”

1. At a Glance

• National Accounts Statistics (NAS) are India's macroeconomic accounts — GDP, GNI, savings, capital formation — compiled by MoSPI's National Accounts Division (formerly CSO) using standardized sources and methods documents released each time the base year is revised. [1]
• MoSPI periodically publishes a "Sources and Methods" volume documenting data inputs, estimation techniques, and classification systems used for each NAS base-year series. [1]
• Currently topical because the base year is being shifted from 2011-12 to 2022-23, with new methodology discussion papers released for public consultation. [2][3]
• Relevant for Prelims (agency names, base years) and Mains GS-III (economy, statistics, governance of data systems).

2. Why in the News

• Government announced a new GDP series with base year 2022-23, replacing the 2011-12 series, incorporating new data sources and methodological changes. [3]
• MoSPI released a discussion paper on "Methodological Improvements in Compilation of National Accounts Aggregates Using Expenditure Approach" on 21 November 2025 for stakeholder feedback. [4]
• A companion Press Note on the New Series of GDP Estimates with Base Year 2022-23 was released in February 2026. [5]
• A related discussion document, "Counting What Counts: Strengthening India's National Accounts," was released in January 2026. [6]

3. Background & Evolution

• It has been CSO/MoSPI practice to publish a "Sources and Methods" document whenever the base year of the NAS series is revised, detailing data inputs and compilation methodology. [1]
• Base year revisions (chronology): 1948-49→1960-61 (published 1967); 1960-61→1970-71 (1978); 1970-71→1980-81 (1988); 1980-81→1993-94 (1999); 1993-94→1999-2000 (2006); 1999-2000→2004-05 (2010); later series moved to 2011-12; now moving to 2022-23. [1]
• Each revision reflects updated economic structure, new survey instruments (e.g., NSS/ASI/Annual Survey rounds), and improved coverage of the informal/services sector. [1]
• The 2011-12 series was itself controversial for growth-rate divergence from a "back series" recompiled for earlier years — clarified by government via press release. [7]

4. Core Static Facts

Item	Detail
Nodal body	National Accounts Division, MoSPI (successor to Central Statistics Office/CSO) [1]
Current base year (moving from)	2011-12 [3]
New base year	2022-23 [3]
Approaches used	Production/income approach and expenditure approach — framework retained, methodology refined [3][4]
Key document	"National Accounts Statistics: Sources and Methods" — published per base-year revision [1]
Recent discussion papers	Methodological Improvements — Expenditure Approach (21 Nov 2025); Production/Income Approach changes (21 Nov 2025) [4]
Press Note	New Series of GDP Estimates with Base Year 2022-23 (Feb 2026) [5]

5. Multi-Dimensional Analysis

Economic

• Base-year revision recalibrates GDP composition/weights, affecting reported growth rates, sectoral shares, and fiscal ratios (debt-to-GDP, deficit-to-GDP). [3]
• New data sources (e.g., updated enterprise surveys, GST-based turnover data) aim to better capture informal-sector and services contribution. [3]

Administrative

• Compilation depends on inter-ministerial data — Registrar General (population), Ministry of Agriculture, RBI (financial sector), corporate MCA-21 database, GST Network — making coordination a persistent challenge. [1]
• Periodic "back series" recomputation is needed to maintain time-series comparability after a base-year shift, which is technically complex and politically sensitive. [7]

Governance / Ethical

• Transparency exercise: MoSPI released discussion papers for public consultation before finalizing methodology — signals participatory statistical governance. [4]
• Past base-year revisions (2011-12 series) faced credibility questions over growth divergence with back-series data — government issued formal clarification. [7]

Scientific / Technological

• Methodological improvements leverage richer administrative data (GSTN, MCA-21) replacing older survey-only estimates, improving granularity and timeliness. [3][4]

Historical

• Six-plus base-year revisions since 1948-49 reflect evolving statistical capacity and economic structural change (agrarian to services-dominated economy). [1]

6. Recent Developments (last 12-18 months)

• 21 November 2025: MoSPI released discussion paper on methodological changes to the expenditure approach for GDP compilation, seeking public feedback. [4]
• 21 November 2025: A parallel discussion paper on production/income approach changes was also released. [4]
• January 2026: Document "Counting What Counts: Strengthening India's National Accounts" published. [6]
• February 2026: Press Note on New Series of GDP Estimates with Base Year 2022-23 released, formally announcing the shift from 2011-12 to 2022-23 base. [5]
• PIB explainer "Redefining Growth: India's Revised GDP Estimates and the New Measurement Framework" and an FAQ "Understanding the New Series of GDP" published to aid public understanding. [8][9]

7. Prelims Hooks

• NAS "Sources and Methods" is published by MoSPI's National Accounts Division, not the Finance Ministry. [1]
• The practice of releasing a Sources and Methods document is tied specifically to base-year revisions of the NAS series. [1]
• India's GDP base year is shifting from 2011-12 to 2022-23. [3]
• Earlier base-year shift (1999-2000 → 2004-05) was published in 2010. [1]
• GDP is compiled via production/income and expenditure approaches — the overall framework is retained even after base-year revision. [3]
• MoSPI released its expenditure-approach methodology discussion paper on 21 November 2025. [4]
• The "back series" refers to recomputed historical GDP data to align with a new base year/methodology, once clarified via a dedicated PIB release. [7]
• MoSPI's National Accounts Division was earlier called the Central Statistics Office (CSO). [1]
• The Press Note formalizing the 2022-23 base-year series was released in February 2026. [5]

8. Why Growth Can Look Different Even If Nothing Real Changed

• A base year is just the year whose prices and weights the whole series is built on
• The new series uses 2022-23 instead of 2011-12 [3].
• The Advisory Committee on National Accounts Statistics picked 2022-23 because it was a normal economic year and the big surveys needed for income estimation were available for it [10].

• That second reason matters: 2020-21 and 2021-22 were COVID years, so their shares of farming, factories and services would have been odd.

• Two things move at once in a revision, and students mix them up

• One, the weights change — services are a bigger part of the economy in 2022-23 than in 2011-12, so the same growth in services now pulls the total up more.
• Two, the data sources change — new surveys and tax records find output the old method missed.

• So a higher or lower growth number after revision does not mean the economy suddenly behaved differently. It means we are measuring with a different ruler.

• This is also why debt and deficit ratios shift

• Fiscal deficit (the gap between what government spends and what it earns) is always reported as a share of GDP.

• If the new series shows a bigger GDP for the same year, the same rupee deficit becomes a smaller percentage — with no change in government behaviour [3].

• The honest defence of doing this at all

• An economy measured with a 2011-12 ruler in 2026 keeps under-counting things that barely existed then — online platforms, delivery services, digital payments.
• Not revising is also a distortion. The problem is not the revision; the problem is comparing across the break carelessly.

9. The Real Repair: Counting Small Unregistered Businesses Every Year

• What was broken earlier
• Most Indians work in tiny unregistered units — a tailor, a roadside repair shop, a self-employed driver. This is called the household sector or unincorporated sector.
• There is no register of these units, so their output could not be measured yearly. It was estimated by assuming they grew at roughly the pace of the organised units we could measure.

• If small firms are actually doing worse than big firms, that assumption hides the gap completely.

• What the new series does instead

• Two annual surveys now exist: ASUSE (Annual Survey of Unincorporated Sector Enterprises) and PLFS (Periodic Labour Force Survey, which counts who is working and at what) [10].

• Because both run every year, the household sector's contribution is now estimated from actual survey answers each year, not carried along on a formal-sector indicator [10].

• Why this is the most important single change in the note's topic

• "Sources and Methods" is not a paperwork document. This is exactly the kind of thing it records — which number came from which survey.
• For Mains, this is your concrete example of a method change improving coverage, instead of the vague line "better data was used".

10. Tax Records Are Not Survey Records — Where GST and MCA-21 Can Mislead

• The new series leans harder on administrative data
• GST data — specifically outward taxable supply, that is, the sales a firm declares — is now used across manufacturing and non-financial services, including as a high-frequency input for quarterly estimates [10].

• MCA-21, the company filings database of the Ministry of Corporate Affairs, covers the corporate sector, including digital and platform services [10].

• The mechanism that can go wrong: a firm entering the tax net looks like new output

• A shop that already existed but sold without a bill contributes nothing to GST records. The day it registers, its sales appear in the data.
• Nothing extra was produced. Only the visibility changed.

• Unless the compilers separate "newly formalised" from "newly produced", formalisation can show up as growth. This is the single sharpest critical point you can make about the new sources.

• Filings are made for tax reasons, not statistical ones

• Firms file to meet a legal duty and to manage tax liability. A survey form is answered for measurement.

• So coverage follows the tax law's thresholds and exemptions, not the shape of the economy. Anyone below the GST threshold is simply absent from that source.

• This fault line already caused a credibility fight once

• When the 2011-12 series and its recomputed back series (older years redone on the new method) showed different growth for the same past years, the government had to issue a formal clarification [7].
• The lesson: a revision that is technically correct can still lose public trust if the old and new numbers are not explained side by side.

11. The Quiet Change in How Quarterly GDP Is Built

• Old way: Pro-Rata benchmarking
• The yearly total was split into quarters in fixed proportions taken from indicators.

• At each year-end the series had to jump to meet the new annual total — creating artificial steps that were not real economic events.

• New way: the Proportional Denton method [10]

• It also forces the four quarters to add up to the annual figure, but it spreads the adjustment smoothly instead of dumping it at the join.

• Result: no artificial break between the last quarter of one year and the first of the next [10].

• Why an aspirant should care

• Quarterly GDP headlines drive market and policy reaction. A step caused by arithmetic used to be read as a slowdown or a revival.
• This is a good one-line example for the question stem on "credibility of growth estimates" — it shows credibility can be damaged by a joining technique, not only by politics.

12. The Weakest Link Is Upstream: The Surveys That Feed the Accounts

• National accounts can only be as good as the surveys behind them
• The National Statistical Commission (NSC) noted with serious concern that on occasions the unit-level data (the raw record for each household or firm) and the summary tables built from them do not match, and asked the government to find the reason and fix institutional responsibility for such failures [12].

• If the base table and its own raw data disagree, every estimate built on top inherits that doubt.

• Delay is a form of error too

• Processing of NSS data was weak until the late 1990s, with long delays in finishing tabulation and a pile-up of pending rounds [13].

• A survey published years late cannot be used for a current-year estimate, so compilers fall back on assumptions — which is how gaps quietly enter the accounts.

• The NSC that is supposed to guard quality has no statute behind it

• The NSC was created by a Government of India resolution dated 1 June 2005 and constituted from 12 July 2006 [12].
• A resolution is an executive decision, not a law passed by Parliament. So the NSC's recommendations do not legally bind the ministry it oversees — a standing GS-II point on institutional design.

13. What Should Be Fixed Next, and Who Must Do It

• MoSPI should publish the back series along with the new series, not after it
• In the 2011-12 revision, the recomputed older years came separately and the mismatch became a controversy needing a clarification note [7].

• Releasing both together removes the space in which "the numbers were changed" claims grow.

• MoSPI should report, separately, how much of the rise comes from firms newly entering GST

• GST sales data is now a main input, including for quarterly estimates [10].

• A published note on the share of the increase caused by new registrations would settle the formalisation-versus-growth argument with data instead of opinion.

• Government should give the National Statistical Commission a statutory basis

• Today it rests on a 2005 executive resolution [12], so its concerns — such as unit-level data not matching published tables [12] — carry no binding force.

• A law would let it demand corrections rather than record them.

• MoSPI should keep the public-consultation route it used this time and make it the standing rule

• For this revision, discussion papers on the expenditure approach and the production/income approach were put out for feedback before the methodology was frozen [4], and a Sub-Committee on Methodological Improvement for the base revision examined the changes [11].
• Making pre-release consultation compulsory for every future base-year change turns a one-time good practice into a permanent safeguard.

14. Anchors for Answers

• Data: Base year shifted from 2011-12 to 2022-23; chosen because 2022-23 was a normal economic year with the needed survey data available [3][10]
• Data: GST outward taxable supply now used as a high-frequency input for quarterly GDP; MCA-21 covers the corporate and digital-platform sector [10]
• Report/Committee: Advisory Committee on National Accounts Statistics — recommended 2022-23 as base year [10]; Sub-Committee on Methodological Improvement for the Base Revision, MoSPI [11]
• Report/Committee: National Statistical Commission — flagged mismatch between unit-level data and published summary tables, sought fixing of institutional responsibility [12]
• Law/Case: NSC created by Government of India Resolution dated 1 June 2005, constituted 12 July 2006 — non-statutory body [12]
• Scheme: ASUSE (Annual Survey of Unincorporated Sector Enterprises) and PLFS — now give annual survey-based estimates for the household/informal sector [10]
• Method: Proportional Denton replaces Pro-Rata benchmarking for quarterly GDP, removing artificial year-end breaks [10]

15. Mains Relevance

• GS-III: Indian Economy — "Statistics," Government Budgeting, GDP/GNP measurement issues, growth vs. employment data debates.
• GS-II (secondary): Governance — transparency and public consultation in policy/statistical processes.
• Possible question stems: 1. Discuss the rationale and challenges involved in periodic revision of the base year for India's National Accounts Statistics. (250 words) 2. Examine how methodological changes in GDP compilation (production/income vs expenditure approach) affect the credibility of growth estimates. (250 words) 3. What is a 'back series' in national accounts, and why is its recompilation necessary after a base-year revision? (150 words)

16. Related Topics to Study Next

• GDP, GNP, NNP concepts and measurement approaches — foundational for understanding NAS methodology.
• MoSPI structure and functions — parent institution; also compiles CPI, IIP, employment surveys.
• NSSO/NSS surveys — key primary data source feeding into NAS.
• GST Network (GSTN) as an administrative data source — increasingly used in new GDP methodology.
• Periodic Labour Force Survey (PLFS) — related MoSPI statistical product, often confused administratively.
• Fiscal Responsibility and Budget Management (FRBM) Act — deficit/debt ratios recalibrated after base-year change.
• UN System of National Accounts (SNA) framework — international methodological standard India aligns with.

17. Common Errors / Trap Areas

• Confusing CSO (old name) with NSO/National Accounts Division — CSO and NSSO merged to form National Statistical Office (NSO) under MoSPI; aspirants often misattribute NAS to CSO alone without noting the restructuring.
• Mixing up base-year revision with annual revision of GDP estimates (Provisional/First Revised/Second Revised estimates) — these are different exercises.
• Assuming GDP is compiled by only one approach; India uses both production/income and expenditure approaches, reconciled together. [3]
• Misdating the base-year shift — current shift is 2011-12 → 2022-23, not to be confused with the earlier 2004-05 → 2011-12 shift. [3]
• Treating "Sources and Methods" as a one-time document — it is republished each time the base year changes, not a static single publication.