“You can slice and dice the data anyway you want, but India’s GDP growth rates between 2004 and 2011 were bound to come down in the backcasting computation effort,” said TCA Anant, former Chief Statistician of India.
When the new base year of 2011-12 came out, the Central Statistics Office (CSO) had documented and recognised that the 2011-12 GDP and Gross Value Added (GVA) estimates were lower than what the 2004-05 (previous base year) series had projected (estimated) for 2011-12, according to Anant.
“If you are bound to get a lower estimate in 2011-12, you are bound to get a lower growth rates for the period (2004 to 2011),” Anant told
Anant attributed the uniform fall in GDP growth rates from 2004 to 2011 partly to the statistical effect and partly to a fall in the tertiary sector’s gross value added (GVA) due to changes in the methodology.
Reasons behind GVA fall
Three broad factors contributed to the fall in the tertiary sector’s GVA: UN’s System of National Accounts 2008 (SNA); fall in the GVA of retail trade; and lowering of the GVA of the unorganised trade.
While SNA only had a small role — measurement of financial services (treating the RBI as a non-market institution) — the most significant change, according to Anant, came in the GVA of retail trade wherein the CSO moved the indicator from the erstwhile Gross Trading Income (GTI) in 2004-05 base year to use of sales tax data for the latest backcasting. Anant felt that the boom in mining sector in mid-2000 may have boosted the GTI, leading to substantially higher gross value added for organised retail trade in those years.
“The implicit growth in retail trade (in 2011-12) is much lower than what national accounts had been projecting using the erstwhile GTI indicator and NSS benchmark of late nineties for 2004-05 base year. We had to bring down the growth rate of trade. The growth rate of retail trade has now been brought down in the latest backcasting through the changes in the benchmark and the change in the indicator to sales tax from GTI,” Anant said.
The improvement that is happening because of the changes in governance will improve the quality of the country’s measure of national income, he said, adding this can lead to higher or lower growth rate, but the quality of measure would be better.
Statistical effect
On the statistical effect, Anant pointed to what is widely known as the Laspeyres and Paasche (L&P) quantity index — an aspect highlighted by Saumya Kanti Ghosh, Chief Economic Advisor, SBI Group, in a recent research note.
Citing an example of a phenomenon in statistics, Anant said: “When I go from 90 to 100, the growth rate is 11 per cent. When I come from 100 to 90 , the growth rate is 10 per cent.” The growth rate declined because the base changed and this will always happen, he said.
“Rest assured, if you had the same data when you backcast, then also growth rate will be lower again. That is purely the L&P index problem which bedevils all of the constant price estimates. The change is exactly the same and the percentage is lower,” Anant said.
Informal sector growth
The latest backcasting has moved the indicator for measuring the value added in the informal sector. “Till now, the problem was the manner in which national accounts were using labour force estimation to project informal sector. What has happened in the current base revision is a change in the method of projection,” he said.
In a large number of cases, informal growth is being linked to contemporary indicators as much as possible as compared to growth in labour force over the past five years as was done for the 2004-05 base year, Anant added.