Discrepancies are inherent part of expenditure side of GDP as of now: T C A Anant

Interview with Chief statistician of India

As economic growth came in at 7.9 per cent in the fourth quarter of 2015-16, many argue that much of it could be attributed to discrepancies. Chief Statistician of India T C A Anant dispels these notions. He tells Dilasha Seth and Indivjal Dhasmana that the principal method of calculating the gross domestic product (GDP) is by taking into account the production-side estimates and not an expenditure one. Edited excerpts:

Many experts have criticised the latest GDP figures, saying much of the nearly eight per cent growth in the fourth quarter could be attributed to discrepancies.

The principal estimate, which has the most data and is the most robust, are from the production-side estimates, not expenditure-side ones. Discrepancies arose on the expenditure side. It is useful for those who comment on growth to have knowledge about time series. It won't be difficult for people to look at past quarterly estimates, both in the new and old series, to see there were many instances - both at the time of provisional and revised estimates - when such high or even higher discrepancies occurred. In an ideal world, production-side and expenditure side, GDP would be a completely independent calculation, with independent detailed data sets.

In India, we do not have that level of detailed information, especially for the expenditure side. We do not get regular expenditure data covering the entire economy with the same degree of periodicity with which we get production data. So, for most parts, expenditure data is estimated, based on the production side and using a set of rules evolved in the base year calculation. Base year is a much more complete data from which we build rules, which are extrapolated to do expenditure-side calculation. As a consequence, the expenditure-side numbers do not add up as we were using the rule of thumb, which was valid in the base year to generate estimates of consumption, gross fixed capital formation (GFCF). Exports and imports are the only timely independent available data. Others are projections for the most part.

When the revised estimates come later, will this be addressed to some extent?

Yes, to some extent. We will get more data on expenditure. But, the general principle is the fact that there are more projections and rate calculations on the expenditure side than production side of calculating GDP.

GFCF, a proxy of investment, fell in the fourth quarter of FY16. Even then, the GDP growth was close to eight per cent. What explains this?

Again, GFCF is derived from projections based on capital goods data in the Index of Industrial Production (IIP). Since those were negative in Q4, that explains the negative GFCF.

But, even then manufacturing rose over nine per cent in the quarter. Isn't it surprising?

We mostly use corporate data. Manufacturing has come in lower in the provisional estimates from what we got in advance estimates. We had estimated 9.5 per cent in advance estimates, while now in provisional, it is 9.3 per cent. The fact that it is much higher than IIP is because manufacturing estimates are largely dependent on corporate value added. IIP principally plays a role in explaining role in non- corporate, quasi-corporate and households at this stage.