JOIN USLies, damned lies and statistics during ‘Achhe Din’
In terms of reliability of official data, India is getting dangerously close to Chinese territory. The year that witnessed demonetization, when 86% of the cash in circulation was declared as illegal tender overnight and which was perhaps one of the greatest assaults on private property in recent history, is now being touted as the year of highest GDP growth in the last eight years. This was the year when all other macroeconomic indicators took a nose-dive, but the GDP for that year has been inexplicably revised upwards to 8.2%, which has left everyone confounded.
The Narendra Modi government came to power promising higher GDP growth rates, and the very next year, in 2015, it delivered. However, it was not entirely due to a booming economy or increased economic activity. Instead, the Central Statistical Office (CSO) changed the methodology of GDP calculation. There were three broad changes – changing the base year, moving to GDP at market prices instead of factor costs, and expanding the database for calculating GDP.
Changing the base year, a routine practice by CSO and other statistical agencies across the world, allows the data to better reflect the current economic scenario. Moving from factor costs to market prices (removing subsidies and adding indirect taxes) was done to align India’s national accounts with that of international best practices. The third step, however, proved to be problematic and attracted criticism. The previous method used RBI’s Annual Survey of Industries (ASI) to measure industrial output, but that was shifted to the MCA21 corporate database. The ASI was a small sample of 2,500 companies and also had a significant lag of two years. The MCA21 has five lakh registered companies and was, thus, supposed to be a more comprehensive and timely database. So far, so good. The problem, however, was that a lot of firms on the MCA21 database were in-operational and many of them did not file returns. The few firms that did file returns were the relatively bigger companies. The CSO would then extrapolate these numbers to the entire economy, severely overestimating the GDP figures. The multiplier that is used to extrapolate from the sample MCA data is not known to the public and cannot be independently verified.
Confounding Revisions
Since then, there has been consternation among the various stakeholders with each GDP data release. In early 2017, all eyes were on the GDP growth rate figure for the third quarter of the 2016-17 fiscal year, the period when demonetization was inflicted upon the economy. Surprisingly, the data showed that the economy grew at a healthy 7%. Further, a few weeks ago, the annual GDP growth rate for 2016-17 was revised upwards from 7.2% to 8.2%. This was baffling, to say the least.
One other missing piece in the puzzle was the fact that it was impossible to compare the new GDP series with the old one. If the economy was growing at 7% despite demonetization and GST, analysts wanted comparable figures for the previous years. After a long delay, there were two separate sets of GDP growth rates released, using the new methodology, from 2004 onwards. The calculations by National Statistical Commission, the agency which officially signs off on GDP data, show that the growth rates during the UPA era were higher than that during the Modi regime. These calculations were sidelined. The Niti Aayog separately published another set of calculations which show precisely the opposite!
According to this back-series data, the economy grew at an average of 6.67% in the nine years ended March 2014 when UPA was in power, while it has grown at 7.35% in the four years ended March 2018. However, the fact is that it is nearly impossible to come up with reliable calculations for calculating the back-series data, as the MCA database was not stable prior to 2010-11.
There are other macroeconomic indicators that can give a much clearer picture of how well the economy was doing in the two periods. GDP is the most comprehensive indicator of the level of economic activity in a country, but it is not the only one. Other macroeconomic indicators like automobile sales, bank credit, tax collections, energy consumption, net exports, and inflation can all give a snapshot of economic performance. Data from government sources clearly shows that performance on all these metrics was better in the UPA period than in the NDA period.
The larger worrying trend here is the systematic assault on India’s statistical institutions. While any one of the aforementioned discrepancies can be assumed to be an error, taken in total, it seems to be a deliberate and systematic approach towards data manipulation and omission. Taken in conjunction with the recent events surrounding the employment data, the suspicions seem vindicated.
The chairman and a member of the National Statistical Commission resigned over the government’s refusal to release the unemployment data, despite all the necessary approvals being in place. Further, they felt sidelined when Niti Aayog released the GDP back-series data without consulting the NSC. This government has also discontinued the Employment-Unemployment Survey conducted by the Labour Bureau in 2015-16 after the previous two surveys showed employment decline by 16 million.
Good policy-making requires reliable data. Collecting and publishing data has always been a difficult affair in India, given the magnitude of the task, compounded by the presence of a large informal and unorganized sector. We do not need deliberate manipulation and withholding of data to make the task even harder.
(The writer is Research Fellow, Takshashila Institution, Bengaluru)