Former Chief Economic Adviser (CEA) Arvind Subramanian has said India’s GDP growth in the period 2011-12 to 2016-17 is likely to have been over-estimated, and the tag of fastest-growing major economy may not hold.
In a research paper, ‘India’s GDP Mis-estimation: Likelihood, Magnitudes, Mechanisms, and Implications’, published by Harvard University, Mr. Subramanian has argued that GDP growth during that period was actually 4.5% rather than the 7% presented by the official data.
“India changed its data sources and methodology for estimating real gross domestic product (GDP) for the period since 2011-12,” he writes in his paper, ‘India’s GDP Mis-estimation: Likelihood, Magnitudes, Mechanisms, and Implications’, published by the Center for International Development at Harvard University.
“This paper shows that this change has led to a significant overestimation of growth,” the paper adds. “Official estimates place annual average GDP growth between 2011-12 and 2016-17 at 7%. We estimate that actual growth may have been about 4.5%, with a 95% confidence interval of 3.5-5.5%.”
Responding to Mr Subramanian's paper, the Ministry of Statistics and Programme Implementation late on Tuesday evening reiterated its stance that the methodology adopted was in line with international standards as set by the United Nations and was as such robust.
17 indicators
Mr Subramanian, whose term as CEA from October 2014 to June 2018 coincided in part with this period of "overestimation", stressed that his paper deals with the technical origins of the GDP overestimation and not the political aspects of it.
He argued that one of the problems with the new methodology for calculating GDP growth since 2011 was that the growth numbers no longer correlated with other indicators of economic growth such as electricity consumption, two-wheeler sales, airline passenger traffic, index of industrial production, and export figures, to name a few.
One of the problems highlighted by the former CEA was that growth numbers no longer correlated with other indicators of economic growth such as electricity consumption, two-wheeler sales, airline passenger traffic, index of industrial production, and export figures, to name a few.
In total, Mr Subramanian looked at 17 such indicators and found that “the correlations between most indicators and GDP growth broke down in the post-2011 period”.
Methodological issues
Former Chief Statistician of India and expert on India’s GDP calculations Pronab Sen countered Mr. Subramanian’s thesis, arguing that it is the result largely of the methodology.
“If you think about GDP growth, it can come from three distinct factors,” Dr. Sen explained. “One is growth in volumes, the amount that is produced. The second is growth in productivity, and the third is improvement in product quality. What Arvind has done is that the indicators he has used are all volume indicators, and having done that, he has said they were very strongly correlated prior to 2011 but not after that period.”
The reason for this breakdown in correlation, Dr. Sen explained, is precisely because the shift in methodology in 2011 meant that the value of goods and services were now considered to estimate growth and not their volume.
“In estimating the growth of the high-frequency indicators pre-2011, he has in a sense replicated the method in which the GDP growth was calculated during that period, and then said that there is a correlation between these indicators and GDP growth,” Dr. Sen said. “Post 2011, when we moved to value indicators from volume indicators, the relationship is weaker because the other two drivers would start getting picked up by the values.”
“If he had made the statement that in the post-2011 growth, only 4.5% came from volumes and the remaining 2.5% came from other factors which we don’t know, then that would have been correct,” Dr Sen added.
Impact of prices
Mr Subramanian also argued that the shift in 2011 to using values rather than volumes meant that price changes, especially in important inputs such as oil, would have started to have a big impact on the final growth number.
“Under the old, establishment-based GDP estimates, price changes mattered less because real growth numbers were largely based on volumes not values,” Mr. Subramanian says. “Under the new system, however, values had to be deflated by prices to get real magnitudes. And this mattered crucially for the manufacturing sector where the often-dramatic changes in oil prices can heavily influence input costs”
Dr Sen also acknowledged that this problem with the new methodology should be addressed. However, he pointed out that while price changes did have an effect now, the direction of that impact was not evident. So, to say that there was only an over-estimation of GDP growth would not be correct. Gauging the effect of price changes on GDP growth using the new methodology would be a different exercise, separate from the one Mr Subramanian has conducted, Dr Sen said.
“Earlier, prices didn’t matter, as he said,” Dr. Sen said. “But it’s not obvious that the direction of error is constant. When oil prices are falling, then you would be over-stating GDP, but when they are rising, you would be under-stating GDP.”
Mr Subramanian’s paper also points towards the fact that the way the informal sector in India was measured was using formal sector proxies, which was an increasingly inaccurate approach. This, too, is a correct assessment, according to Dr Sen, who said this problem has become even more acute post-demonetisation, when the informal sector’s growth fell away.
“But remember that the period Arvind is talking about is pre-demonetisation, so this argument I don’t think applies with as much force then,” Dr. Sen said.
Mr. Subramanian derives several implications from the findings of his paper. The first is that growth needed to be restored to high levels. The second that the quality and integrity of data in India needs to be improved, something called for by several other economists. And the third is that “India must restore the reputational damage suffered to data generation in India across the board”.
He also called for the creation of a taskforce to revisit the entire methodology and implementation of GDP estimation.
