On the pitfalls of estimating GDP

For consultant functions.
| Picture Credit score: Getty Photos

Gross Home Product, or GDP, is essentially the most vital measure of a rustic’s financial dimension. It is usually a common denominator for evaluating indicators throughout nations and areas or for sizing up tax burdens or welfare expenditures. GDP is normally extra significant at “fixed” costs or in “actual” phrases — netting out the impact of worth adjustments. The actual GDP is estimated for the “base 12 months”, requiring quite a lot of datasets on output, costs, and employment. Each 5-10 years, the GDP base 12 months is revised to account for adjustments in relative costs and output composition. The Nationwide Statistical Workplace (NSO) is tasked with “ revising” the GDP sequence, normally drawing upon experience from many fields.

The continuing GDP sequence with the bottom 12 months 2011-12 is due for revision. 2020-21 is the proposed new base 12 months. All required main datasets are mentioned to be accessible apart from Census information. The NSO is contemplating utilizing the products and companies tax (GST) information to estimate worth addition, changing the at present used Ministry of Company Affairs’ MCA-21 database for the Personal Company Sector (PCS), which accounts for about 38% of GDP.

Why the change?

In any case, the MCA-21 database was introduced in solely within the final revision, with 2011-12 as the bottom 12 months. Earlier to that, the Annual Survey of Industries (ASI) was the long-standing workhorse for estimating manufacturing unit manufacturing value-added. The Reserve Financial institution of India’s (RBI) small pattern of huge firms, with the bulk paid-up capital of PCS, was used to estimate the non-financial company sector output. The statistical company modified it to the MCA-21 database because the ASI claimed to overlook out on worth addition outdoors of manufacturing unit premises in a company entity. Likewise, reportedly, the RBI pattern was insufficient to account for the quickly rising PCS. Furthermore, the provision of the in depth and up-to-date MCA-21 information, obtained from the necessary submitting of company annual returns and quarterly company outcomes — it was contended — would allow fuller capturing of the company output.

The 2011-12 base 12 months GDP (changing the 2004-05 base 12 months sequence) confirmed a touch smaller absolute GDP dimension and a quicker development fee. However for the manufacturing sector in 2013-14 at fixed costs, the annual development fee was (+) 5.4% within the new sequence, in comparison with (-) 1.90% within the earlier sequence. Such a pointy divergence within the fee and path of business development by the 2 GDP sequence was a shock. Furthermore, the upward revision of the economic development fee didn’t sq. with associated macro aggregates, similar to financial institution credit score development or industrial capability utilisation, resulting in widespread scepticism of the brand new GDP estimates. Statistical investigations zeroed in on an untested or inadequately vetted MCA database because the supply of the overestimation downside.

The official company, nevertheless, defended its new estimates, claiming they seize worth addition extra fully, utilizing a way more in depth database, improved estimation strategies, and following the newest template of worldwide finest practices. Critics, nevertheless, questioned if an even bigger dataset is essentially a greater information set. And if the brand new estimates had been higher or overestimates. The statistical dispute remained unresolved as the federal government refused to make the MCA information accessible for impartial scrutiny or reveal its estimation methodology for verification.

Systematic overestimation

With time, nevertheless, it has been attainable to match estimates of Gross Worth Added (GVA) within the manufacturing sector as per GDP sequence (within the Nationwide Accounts Statistics) and by the ASI — based mostly on manufacturing accounts of registered factories — for a reasonably long period. We in contrast (i) GVA and (ii) Gross Mounted Capital Formation (GFCF) (mounted funding) at fixed costs for 2012-13 to 2019-20 as reported by the NAS and ASI. The outcomes had been startling. The common annual development fee of GVA in NAS was 6.2%, whereas it was solely 3.2% in ASI. The distinction was a lot sharper in GFCF: 4.5% by NAS and 0.3% by ASI, respectively. These comparisons present a scientific overestimation in NAS estimates (based mostly on the MCA-21 database) in comparison with the ASI-based estimates, vindicating the doubts raised concerning the integrity of the GDP estimates.

The proof introduced here’s a cautionary story for the proposed use of GST information for GDP estimation. It’s a stark reminder of the necessity for the official company to protect towards the hasty utility of unverified datasets and shaky methodologies with out ample testing and validations for GDP estimation. NSO should provoke pilot research to confirm the GST dataset’s suitability for worth addition estimation of particular industries, sectors, and States. Such validation is essential to make sure the estimation’s truthfulness and instil confidence within the integrity of the GST information. Alternatively, NSO might discover reverting to ASI to estimate GDP manufacturing, because the database is now accessible with a shorter time lag.

GST information is usually a game-changer for GDP estimation within the proposed revision. It’s a giant and up-to-date database, nevertheless, its particulars are in a black field, because it has not been open for coverage analysis. With out systematic analyses and cross-validation disaggregated by manufacturing and institutional sectors and areas by impartial businesses, the validity of GDP estimates on GST information shall be onerous to ascertain.

R. Nagaraj is with the Centre for Liberal Schooling, IIT Bombay.