India’s GDP growth revisions shouldn’t raise eyebrows: They’re routine

This revision, which brought annual growth to one of the highest levels in the post-covid period, sparked considerable debate in the media as well as on social media over India’s data quality and statistical governance. Surprisingly, even well-informed commentators demonstrated a lack of understanding about the reasons behind variations in GDP estimates.

This article explains the factors contributing to these revisions in National Account estimates and explores possible improvements in future GDP series.

Also Read: The state of India’s economy is not as bright as GDP data may suggest

GDP revision patterns: To put the discussion in perspective, it is useful to examine the pattern of GDP revisions under the current (2011-12) series, as shown in Table 1.

The data indicates that revisions have historically occurred in both upward and downward directions. The most significant changes typically take place between the provisional estimates and first revised estimates and between the first and second revised estimates.

Why do GDP revisions occur?: GDP revisions result from changes in data availability. Table 2 outlines the key data sources and their availability timelines for GDP computation.

Key insights from data availability timelines: From the second table, it is clear that most of the key data inputs are finalized only after the release of provisional estimates. Since many elements follow fixed statutory timelines, most of this calendar is essentially irreducible, with limited potential for reducing variations between the provisional estimates and first revised estimates. Major changes that occur between the first and second revised estimates are on account of late filings of corporate accounts, availability of data from the Annual Survey of Industries (ASI) and the completion of the reconciliation process with gross state domestic product (GSDP) estimates.

The availability of ASI estimates is a crucial determinant of the variability between the first and second revised estimates. It plays a key role in allocating manufacturing value-added to different states. This is important in helping states finalize their GSDP accounts (and thus GSDP reconciliation).

Potential improvements in GDP estimation: The volatility of estimates in the current National Accounts series is explained by patterns of current data availability. Newer and richer data sources have since become available, which can be considered in the next base revision. Some of these are discussed here.

Also Read: GDP growth: India’s latest economic data stirs up a range of emotions

First, leverage GST data for faster estimates: The current ASI methodology relies on National Sample Survey investigators visiting selected establishments, which is both resource-intensive and time-consuming. An alternative approach could involve using goods and services tax (GST) filings to generate faster, albeit coarser, estimates of establishment-level value-added. This could help reduce fluctuations between the first and second revised estimates.

Second, begin direct data collection from corporate headquarters: Instead of depending solely on field-level data collection, a survey conducted among corporate head offices for establishment- level data could accelerate the availability of data on formal-sector enterprises. GST registration information can assist in identifying a sampling frame for such an exercise. Such a survey would improve the existing ASI by going beyond manufacturing to include services.

Third, use Annual Surveys of the Unincorporated Sector: The ministry of statistics and programme implementation (MoSPI)n has recently introduced India’s Annual Survey of Unincorporated Sector Enterprises (ASUSE) to track commercial activity in the informal sector. A fact-sheet for ASUSE 2023-24 was released in December 2024—nine months before ASI data for the same period. Incorporating ASUSE data into National Accounts could improve informal-sector estimates and reduce dependence on ASI availability.

Also Read: Gross domestic product: Grossly inadequate as a measure of well-being

Conclusion: Understanding GDP revisions requires a clear grasp of the various factors driving variability. If we analyse previous years in the same manner as done in Table 2, it becomes evident that systematic improvements in data quality and availability have been made over time.

While the revision process is complex and data-driven, attributing changes in GDP growth to political factors is neither accurate nor productive. Instead, the focus should be on further refining data collection, incorporating alternative data sources and improving the timeliness of Indian statistical releases.

The author is a former chief statistician of India and visiting professor at the Institute for Studies of Industrial Development.