We have used more reliable data to decompose economy: Pravin Srivastava

It is difficult to draw a correlation between some economic indicators and GDP growth, says Srivastava

After what turned out to be an eventful week for Chief Statistician of India Pravin Srivastava — his ministry released the back series of national accounts for the UPA tenure on Wednesday and GDP growth for Q2FY19 on Friday — he speaks to Abhishek Waghmare on the rationale behind the GDP back series and the slowing down of growth in Q2. Edited excerpts:

The data on GDP growth in Q2 shows a dip in growth. Is it a matter of concern?

I don’t think so. A growth of above 7 per cent when the fundamentals of the economy are becoming stronger still makes India the fastest growing large economy. Further, we must look at quarter-over-quarter, where growth is visible. Growth of 7.4 per cent in manufacturing in Q2FY19, over 7.1 per cent in Q2FY18, is not a cause for worry. The high growth in Q1 was on a much lower base.

What would you attribute the slowdown in Q2 to?

Mining was affected, probably due to high crude oil prices and increased imports in the period. But other sectors need to be looked into, at the sub-sectoral level. However, with inflation under control and crude prices reducing, cause of concern is now less.

The GDP back series has drawn some criticism since it revised GDP growth in the “boom period” of 2004-08 at levels lower than recent years. Corporate earnings, income tax, bank credit and consumer durables sales were higher then. How to reconcile the opposing sets of data?

It is difficult to draw a correlation between some economic indicators and GDP growth. Estimation of the latter is a complex exercise that uses data from public sources and a globally accepted methodology, which may not directly reflect select indicators. There could be a correlation between some indicators and economic growth, but it may not be statistically significant. Perhaps the same indicators overestimated the economy at that time.

Then which factors contributed to the downward revision?

The benchmark surveys of the National Sample Survey Office (NSSO) were not available while revising the base year to 2004-05. We have used more reliable data to decompose the economy, look at it using this data, and then re-compile and add it up at the sectoral level, to get back-casted estimates. As a result, the over-estimation that had happened in communications, unorganised trade and financial services sectors got revised.

What would you say about the sharp political criticism that ensued?

It is unfair to make a political statement on an apolitical data exercise, or to associate it with a particular government. We have robust data that has stood the test of time, and which has been accepted by economists. 

The presentation of the data happened in NITI Aayog, and it raised questions on whether the think tank was involved in the statistical exercise. 

The NITI Aayog had a purely supportive role and was limited only to dissemination of the data. We, at the ministry of statistics, do not hold press conferences. We asked them to be a part to help explain the data to the public and enhance credibility. 

How will you convince the public the methodology used to do it is the best?

Given the availability of data, we have used only tested methods to arrive at an optimal solution. To have a consistent series was very important. 

Is the exercise to revise the base year to 2017-18 currently on?

Not yet, since results of the base surveys are yet to come.

Looking forward, which data sets should we expect in the coming months?

The Periodic Labour Force Survey will arrive in December. Probably in January, we will come up with the survey on unorganized sector for the 2016-17 period.