About a third of non-government non-financial companies in the services sector are not traceable is the finding of a National Sample Survey Office (NSSO) survey for 2016-17 that has just been released. Since such entities could be shell/fake/bogus companies included in the MCA-21 database of “active” companies used for estimating the gross domestic product (GDP), the new finding could imply that private corporate sector GDP is being currently overestimated, denting the official growth narrative.

The background

In 2015, the Central Statistics Office (CSO) issued a new GDP series with 2011-12 as the base year, replacing the earlier series with the 2004-05 base-year as a routine matter. Usually, the revision leads to a slight expansion of the absolute GDP in the base year, but its growth rate does not change, implying that the underlying pace of economic expansion in the two series has remained the same. This time was different, however. The absolute GDP size — the sum of the value of all (unduplicated) goods and services produced in a year — got diminished slightly in the base year, and its growth rates went up subsequently.

Faced with public scrutiny and scepticism, the CSO defended the revision by claiming that it had followed the latest global template (the System of National Accounts 2008), applying improved methodologies to a newer and larger data set; hence the new GDP was kosher. In a first, the new series estimated private corporate sector (PCS) GDP directly using the Ministry of Corporate Affairs’ (MCA) statutory filing of financial returns, MCA-21. Accounting for over a third of GDP, as the non-financial PCS now spans widely, the revision has affected the estimates of many industries and services. Hence the GDP debate has mostly centred on the PCS.

Since the MCA-21 database is much larger than those used earlier — like the Annual Survey of Industries (ASI) for manufacturing or the Reserve Bank of India sample of large companies for estimating corporate saving and investment) — the CSO claims the new GDP better captures the economy’s value addition, especially of smaller enterprises and services activities. Critics, however, wonder if it is a case of a better description or an overestimation.

Screening and setback

Though now contributing over a half of GDP, services sector output estimates are poorer, outside of the public sector and large private sector companies. To redress the shortcoming, the CSO is committed to launching an annual survey of services (on the lines of the ASI). As a first step, the NSSO carried out a survey of non-government and non-financial companies/establishments in 2016-17, using three list frames (or the universes of enterprises) to draw samples for the survey, the MCA sample being the largest. A 10% sample was drawn from the CSO’s universe of 3.5 lakh active non-financial companies. After due screening, at the survey stage, the NSSO, to its shock, found that 45% of the selected companies did not respond to the survey.

The NSSO report says, “About 45% of MCA units were found to be out-of-survey/causality.... Non-response of a large number of units was a major setback for this survey. This happened due to unit non-response, closure of the unit, unit found to be the one other than headquarter, unit out of coverage or unit non-traceable.”

To its dismay, the NSSO found the results to be so poor that it had to abandon the planned output of two-volume survey results, and instead settle for a brief technical report which was released recently. It candidly admitted the difficulties the NSSO faced in the survey: “Many units, particularly of the MCA list, were not identifiable due to lack of proper/adequate postal addresses. Therefore, many notices could not be delivered. A large number of out-of-coverage units was also found in the list. Affixing signatures on Schedule 2.35 for out-of-coverage units was time-consuming and difficult as owners were reluctant to sign. In many cases, it was found that the selected enterprises had not prepared the Annual Audit Report for 2015-16 or the balance sheets any time before.”

The inference could be that such companies are likely to be shell/fake/spurious entities that remain legally registered (but merely on paper), without actually producing goods and services.

Impact of estimation

What does this imply for GDP estimation? Apparently, quite a lot since they are part of the universe of active companies for which GDP is estimated. Though it may not be possible to show how much difference it would make to the GDP level and growth rate (in the absence of more information), the survey findings could bring down the growth estimates.

However, those knowledgeable have dismissed such an apprehension on two counts: one, shell companies add value to the economy, hence their deletion would underestimate GDP; two, as all active companies are said to submit their audited accounts at least once in three years, the contribution of shell companies is well captured in the MCA database.

Both arguments seem questionable. Shell companies, by definition, do not produce goods and services; they help the promoter/owner to hide profits or evade taxes/regulation. A dictionary meaning of a shell company is “a company existing as a legal entity but having no significant assets, independent business operations, etc., often owned or controlled by another company and used for various, often illegal purposes”.

The argument that all active companies under the MCA have filed statutory returns at least once during the last three years is a bureaucratic fiction. If it were true, there would be at least one year for which there would exist data for seven-eight lakh companies, which has never been the case. In reality for most years data are available for around three lakh active companies, estimates for which are inflated (or multiplied) for the (fictitious) universe of about 10 lakh active companies. As the database is not made public and the methodological details are not adequately revealed, there is no way of verifying the veracity of the official estimates — an issue critics have flagged since 2015.

If the share of shell/fake/bogus companies in the universe of active companies in the MCA-21 database is as high as what is found in the NSSO’s services sector survey, then GDP estimates based on a more realistic list of working companies are likely to be smaller. Hence this could affect the corporate sector’s GDP level and its growth rate.

Case for scrutiny

In sum, the NSSO’s survey of active companies in the services sector discovered that 45% of them could not be traced or misclassified; hence they could represent or be shell/fake/bogus companies. The finding throws into sharp relief the poor quality of the MCA-21 data set, which has formed the backbone of the new GDP series. The NSSO survey results have added more questions about the beleaguered GDP series, strengthening doubts that have arisen from various aspects of the revision process. As a first step towards dispelling the growing distrust in the new GDP series, the government should put up the MCA-21 data for public scrutiny and lift the opacity of the methodology used in estimating corporate sector output.

R. Nagaraj is with the Indira Gandhi Institute of Development Research, Mumbai