NSO to use GST data for GDP estimate in new series

The National Statistical Office (NSO) will likely use GST Network (GSTN) data to compute the gross domestic product (GDP) starting with the new national income series to be unveiled in the next financial year. The data of the revenue collections from the principal indirect tax will be a key gauge of private final consumption expenditure (PFCE) in the GDP series, set to be released in February 2026, official sources said.

The NSO has tested Goods and Services Tax (GST) data to compute GDP numbers for the July-September and October-December quarters of the current fiscal year, the sources added. However, the GDP figures released so far don’t include the GST data.

Official sources say that the inclusion of data sourced from GST Network (GSTN) for computing the private consumption activity, a component in the GDP, will likely lead to more accurate estimation of the latter. “This has been done (testing) to check the robustness of data quality,” said a source.

In the coming quarters, the flow of data from GSTN will improve, which will have a direct impact of cutting down the magnitude of GDP revisions, as the data will be real-time and more accurate, the sources say.

In the past few weeks, economists and statisticians have raised concerns on the quantum of revisions in the GDP figures. Last month, the NSO revised the FY24 GDP print upwards by 100 basis points (bps) to 9.2%, which, as per experts, has been the highest in recent times.

The private final consumption expenditure (PFCE), the component carrying a share 56-57% in the GDP, was revised upwards from 4% earlier (provisional estimates) to 5.6% (first revised estimates).

The government final consumption expenditure (GFCE) too was revised higher from 2.5% to 8.1%. However, the GFCE’s share in GDP is much lower, in the range of 9-9.5%.

“The GST data comes with no time lag. It also has a wide coverage, and will certainly help reduce some of the sampling errors we make,” noted NR Bhanumurthy, director, Madras School of Economics.

“We need to ensure stability of the estimates. The GSTN data should be used for estimating PFCE for at least eight quarters, before using it to calculate GDP estimate,” he added.

The statistics ministry has constituted a committee, known as Advisory Committee on National Accounts Statistics (ACNAS), to deliberate upon the various aspects of base revision of the GDP. The committee is meeting on regular intervals to also discuss how the revisions can be reduced.

As per sources, the first GDP estimates, on the new base year will be released in February 2026. The ACNAS is likely to choose 2023-24 as the new base year of GDP.

Sources say the new base will use data from Annual Sector of Unincorporate Sector Enterprises (ASUSE) 2022-23, Annual Survey of Industries (ASI) 2022-23, and other sources. For using GST data, to measure consumption activity, the ministry is coordinating with the GSTN to receive “anonymous information”.

Further, to improve the computation of real GDP numbers, the ministry is working on improving the method of deflation. Sources say the method of deflation will be improved to double deflation, as the single deflation method is a “less sound” method of arriving at real estimates.

Double deflation is a method through which nominal outputs are deflated through an output deflator, and inputs are deflated using a separate input deflator. Currently, India uses a single deflator, using an input price deflator – the Wholesale Price Index (WPI), which inflates the real growth figure in situations when input prices diverge from output prices.

Currently, the estimates of PFCE are obtained by following the “commodity flow” approach, says the ministry on its website. This implies working of commodity balances relating to various items of consumption, taking into account production; intermediate consumption in agriculture, manufacturing and other industries; net imports; stock variations; consumption on government account and gross fixed capital formation.