)
While doubts over the new gross domestic product (GDP) data series refuse to die down, the ministry of statistics and programme implementation (MoSPI) is pretty sure of its methodology and has rebutted these charges.
The latest to raise doubts over the methodology is R Nagaraj, a professor at Indira Gandhi Institute of Development Research (IGIDR), in the Economic and Political Weekly (March 28 issue). To address his criticism, the Central Statistics Office (CSO), under the ministry, is issuing a point-to-point rebuttal of these in the magazine's coming issue (April 18), sources said.
Meanwhile, to assuage doubts, a committee headed by National Statistical Commission Chairman Pronab Sen has also been appointed to review the new GDP exercise sources added.
Nagaraj, who was a non-official member of CSO's sub-committee on 'Private corporate sector, including PPPs', had accused the organisation of adopting a different method in its final report than what was agreed upon in the sub-committee, thereby over-estimating performance of the private corporate sector.
The CSO's article will argue that a major point of departure between the two estimates is that the sub-committee was asked to restrict output in its draft report to the total revenue posted in the ministry of corporate affairs (MCA) database.
Explaining, Chief Statistician T C A Anant told Business Standard, "At the time of preparing the final report, it was pointed out that the sum of all line items under the revenue head did not tally with total revenue in the MCA database."
Further, it was felt that individual items reported under the revenue head would give a better real picture. Thus, output was estimated using individual components and not the total revenue. This increased the output of the non-financial corporate sector for 2012-13 by 4.9 per cent. This change, coupled with some other changes in intermediate consumption, resulted in the gross value added (GVA) for 2012-13 rising 31.8 per cent.
It is essentially this difference, which translates to an over 200 per cent increase in savings of the non-financial private corporate sector (PCS) that Nagaraj had pointed to.
The other point of departure between the sub-committee's methodology in the draft report and the final one is that estimates from the MCA database were scaled up in the latter to get complete coverage of all active companies. This is so because MCA21, the ministry's e-governance initiative, covers only 85 per cent of the total paid-up capital (PUC) of all active companies. So, it was scaled up by 15 per cent to provide a true picture.
Also, the advisory committee was concerned that as the number of companies filing returns in the database might fluctuate from year to year, it would be difficult to compare across years. Simply using the MCA database would increase the volatility of the estimates.
The number of active companies in the country is around 900,000, while the MCA database contains around 500,000 companies, accounting for 85 per cent of the PUC. According to the note prepared by CSO, "In the year 2012-13, the PUC of these companies was 85 per cent of the PUC of all active companies, so a factor of 1.15 was applied to account for all the active companies."
For 2011-12 and 2012-13, around 500,000 companies' database was in MCA21, so the scaling up was of the same magnitude. This would have raised the levels but the growth rate would not have changed due to the scaling up, Anant explained.
However, for the 2013-14 estimate, only 300,000 companies had filed their returns by October 2014. Thus, the scaling up was of a much higher magnitude. This was because CSO issues GDP estimates by January. Hence, it was decided to specify a cut-off date to see the MCA21 data base, fixed as November 30. For 2013-14, not all companies had filed their data base by then.
But extrapolation could inflate estimates. It is quite likely that big companies, which presumably account for more value addition, are more regular with their filings. Thus, using the same growth rates to adjust for smaller companies could distort the estimates, leading to higher than actual value addition. Anant acknowledged this but stressed that as these would be revised, it would be prudent to wait for the revisions before arriving at a firm conclusion.
On the issue of the difference in the GVA growth rates, the CSO note says the change for 2012-13 for the non-financial corporate sector is not significant. The rate estimated in the draft report was 14.7 per cent, only marginally lower than the 15 per cent in the final report. The industry-wise changes in growth rates are due to revisions in industrial classification, which has increased the number of companies classified in the 'trade' category. But despite detailed clarifications, doubts about the new series remain, as the new numbers seem substantially out of sync with other economic indicators. A report from the Emerging Advisors Group, a global consultant, examines the growing divergence between key economic indicators and GDP growth under the new series.
It contrasts the real growth of 12 economic indicators during the high-growth phase of 2005 and 2006, with their growth in the second half of 2014 (July to December), when GDP growth according to the revised methodology was just shy of eight per cent.
The 12 indicators it identifies are agricultural and industrial production, credit growth, electricity and cement generation, vehicle sales and freight traffic growth, export volume, corporate sales, the revenue of listed companies and government revenue.
The variance is staggering. Only one indicator - power generation - grew faster in 2014 than in earlier years. For most other indicators, growth in 2014 lagged those in 2005 and 2006.
Part of the explanation for the growing disconnect between these indicators could be that these are volume indicators, while the GDP is measuring value addition. This explanation suggests that during these years, there has been a significant increase in productivity levels.
But, Anant said it would be unfair to use one data point to talk about trends, as figures for previous years are not known. "You are comparing the new series with the old one and coming to the conclusion," he said.
