Reality check on Eco Surveys: A case of hits and misses on GDP numbers
Of the 7 Surveys presented under this govt, 3 were close to the growth rate
If someone were to rate the finance ministry’s advisors on the accuracy of their economic growth forecasts during the tenure of the current government, the person would, in all probability, give them five out of 10.
Of the seven Surveys presented during the regime, three predicted either a somewhat correct number or underestimated growth. One of the Surveys can’t be given a score because the base year changed. The remaining three, all presented during the past three years, were way off the mark in predicting growth numbers.
Economic Survey 2013-14
This Survey was authored by a team led by Arvind Mayaram, then economic affairs secretary. One should note that the Surveys project economic growth for the next year and generally takes the GDP expansion/contraction for the current year from the advance estimates. In this respect, this Survey seems to underestimate economic growth (see chart). However, by the time the GDP numbers came, the methodology for computing it had changed. GDP is now calculated at market prices (which includes indirect taxes), unlike GDP at factor cost (which excludes indirect taxes), which had been the practice till then. Besides, value addition is taken into account in various sectors such as agriculture, industry, and services from the supply side of the computation.
2014-15
This was the first survey under Arvind Subramanian as chief economic advisor. Taking inspiration from the International Monetary Fund’s World Economic Outlook, this Survey departed structurally from its predecessors and was presented in two volumes. Volume one discussed the outlook and prospects as well as a number of analytical chapters addressing topical policy concerns. Volume two described recent developments in the major sectors of the economy and contained statistical tables and the data. In a sense, volume one was forwardlooking but gained from the perspective provided by the recent past, which was the subject of volume two.
The Survey was close to predicting the correct GDP growth number, though closer to the lower end of its range. Against its prediction of 8.1-8.5 per cent range, GDP growth turned out to be
8 per cent.
2015-16
Economic growth in 2016-17 turned out to be much higher than predicted by this Survey - 8.3 per cent, versus 7-7.5 per cent as projected, despite demonetisation.
This may also be due to revisions in the growth rate in later years. Though one may say this Survey was also off the mark, underestimating the growth rate would not have had as disastrous an impact as overestimating it.
2016-17
Growth in 2017-18 fell within the range predicted the Survey predicted (see chart). Volume one and volume two were presented with a time lag. Volume one was presented in February 2017 after the government announced demonetisation in November 2016. Volume two was presented in August 2017, a month after goods and services tax was introduced.
2017-18
From here onwards, the Surveys slipped up in projecting growth and overestimated it by a huge margin. Against its projection of 7-7.5 per cent, the rate turned out to be 6.1 per cent for 2018-19.
2018-19
The first Survey authored by current Chief Economic Advisor Krishnamurthy Subramanian was way off the mark in predicting growth. The economy grew at 4.2 per cent against the 7 per cent the
Survey projected. This was probably due to the slowdown, which had cast its shadow on the economy, and the initial impact of Covid-19.
2019-20
At the beginning of calendar 2020, the year 2020-21 was a difficult one to predict. That’s why the Survey, tabled in Parliament at the end of January, projected the economy to grow by 7 per cent. Now the economy is officially projected to contract by 7.7 per cent.
Economist Ila Patnaik, who was one of the main authors of the Economic Survey for 2013-14, said the finance ministry needed to devote resources to the modelling and forecasting capacities and do it as a structured and regular exercise.
Another economist said in many other countries such models were done with other expert organisations in a far more organised and transparent manner.