29-06-2025
GVA data haze: Has India been overcounting the output of its informal sector?
Gift this article
In his Statistically Speaking column published on 19 June in Mint ('India's informal sector is being tracked better than ever before'), former chief statistician of India T.C.A. Anant, while taking on those who argue India's GDP in the current series may be overestimated, has contended—among other things—that such criticism requires a more careful analysis.
In his Statistically Speaking column published on 19 June in Mint ('India's informal sector is being tracked better than ever before'), former chief statistician of India T.C.A. Anant, while taking on those who argue India's GDP in the current series may be overestimated, has contended—among other things—that such criticism requires a more careful analysis.
His article ends on the optimistic note that better-than-ever availability of timely and regular informal-sector data presents an opportunity for India's ongoing base-revision exercise to measure value added by the informal sector in a manner that offers more clarity. Also Read: India's informal sector isn't off the map: It's being tracked better than ever
Overall, the informal (or unorganized) segment of the economy contributes 45% of the total gross value added (GVA) and accounts for 33% of the non-agriculture sector. For many sub-sectors within the non-agriculture sphere, value added for the base year is estimated as a product of GVA per worker (GVAPW) and the workforce—through what is called the 'labour-input method.'
These benchmark estimates are then moved for later years as per the relevant selected indicators, like corporate growth, the volume/quantity index, sales tax and others. For 2011-12, the base year for the current series, the source of GVAPW data was the National Sample Survey 67th Round (2010-11).
It is important to appreciate the definitional nuances involved, as these have a vital bearing on both the final estimates of GVA and their interpretation. In enterprise surveys, an 'own-account enterprise' is defined as a business that is run without any worker hired on a fairly regular basis, while an enterprise employing at least one paid worker is termed an 'establishment.' Further, an establishment with six or more workers is termed a 'directory establishment' (and the rest are 'non-directory establishments'). Also Read: Data deficiency: India needs to map its informal economy better
In the 2011-12 series, base-year GVA estimates of the unorganized segments of many sectors—such as trade, hotels and restaurants, telecom and educational, health, real estate and professional services—in which the labour-input (direct or modified) method was adopted were made on the basis of a palpably higher GVAPW for establishments in rural areas and directory establishments in urban areas.
This was done on the professed reasoning that '...most of the establishments in urban areas are Directory Establishments. This is also in consonance with the practice adopted in previous series."
The use of GVAPW makes calculations complex.
As a case study, consider the informal segment of the trade sector. To estimate its size, the GVAPW of directory establishments was used in the 1981 series. In 1999-2000 as well as 2004-05, the GVAPW used was that of 'all enterprises,' not just directory establishments. For 2011-12, that was changed, even though the contention cited above that most enterprises in urban areas are directory establishments was not correct, as the share of such establishments in the urban trade sector was merely 2.1% (see graphic above).
This should be viewed alongside the fact that the GVAPW of directory establishments was placed at 2.40 times that of own-account enterprises (the vast majority with a 77.3% share). Likewise, in rural India, the estimated GVAPW of establishments (with a share of 11.4%) was 2.31 times that of own-account enterprises (the overwhelming majority with a share of 88.6%). So the method adopted for the base year 2011-12—which took a thin slice of relatively productive businesses to represent the whole—overcounted the value added per worker and would thus have resulted in a higher GVA figure.
The patterns in other sectors are also more or less along the same lines.
To illustrate, it may be seen from the above graphic that in trade, there is a significant decline in GVAPW growth among all categories of workers during 2015-22 in comparison with 2010-15, while in the National Accounts, informal trade is seen to have grown annually at a rate of 10% over the period 2015-22.
This gap is too wide for changes in the workforce during that period to account for it. A similar trend is observed in other sectors too. It is highly probable that apart from base-year estimates, subsequent growth rates in these informal segments have been overestimated in India's National Accounts, particularly after 2015-16. In our opinion, the only debatable issue is the extent of this overestimation. Also Read: TCA Anant: How India's statistical system could win the ongoing war of narratives
Data from agriculture is generally not commented upon much. However, Drèze and Oldiges ('India's Puzzling Cereal Gap,' published in Down To Earth on 3 January 2025) have highlighted significant and increasing data gaps between the production of cereals (like wheat and rice) and all their known uses, reaching nearly 70 million tonnes in 2022-23, enough to feed half a billion people for one year.
India's gap between net cereal availability and household consumption is stated to have started widening around 2008. While it was only around 30 million tonnes in 2011, the gap had grown enormously by 2022-23. Is India's cereal production overestimated? This issue needs to be settled, lest it also leads to GVA overestimation. Also Read: GDP's dirty little secret: Why we should be tracking GVA instead
Since the Union ministry of statistics and programme implementation proposes to start conducting its survey on the informal sector annually, it is acceptable to argue that it is being tracked more closely. The jury, however, is still out on whether the informal sector is actually being tracked better. Before adoption, it is advisable to analyse these results for their utility in sectoral GVA estimation. Else, it may end up raising more questions about the economy instead of delivering the clarity we hope for.
These are the authors' personal views.
The authors are, respectively, former additional director general and director general, ministry of statistics and programme implementation, Government of India. Topics You May Be Interested In
