Last week the Central Statistics Office (CSO) announced major changes to the way key macroeconomic data - Index of Industrial Production (IIP) and Wholesale Price Index (WPI) are calculated by changing the base year for data computation. Not only that, the baskets for both the IIP and WPI have also been expanded. CNBC-TV18's Latha Venkatesh caught up with Chief Statistician, TCA Anant and discussed about the impact of new WPI and IIP series on FY17 GDP.
Below is the verbatim transcript of the interview.
Q: One wants to know the impact of the new WPI series and the IIP series on the FY17 GDP. The first three quarters' GDP will go up, will it not?
A: I am not going to comment on what the impact is going to be except to refer to the statement which we made when we had released the new series of gross domestic product (GDP) that we will incorporate the wholesale price index (WPI) and Index of Industrial Production (IIP) in GDP compilation. The methodology of incorporation is already available in public domain, but what will be the actual implications on the incorporation, we will have to wait until our GDP releases for me to tell you.
Q: So the next GDP release which is on May 31 will incorporate the impact of the new WPI and IIP, so we should expect a revised Q1, Q2 and Q3 GDP as well on May 31?
A: In any case, at the time of the Q4 release, we revise Q1, Q2 and Q3. So, that in any case would have happened, so yes either way, but they will also incorporate the WPI and IIP numbers.
Q: But surely we are right in guessing that the impact would be to push it higher is it not? After all, the WPI deflator is going to be lower.
A: Generally, I do not like to speculate because these things are not used in aggregate terms in many places, it is used in different ways in different parts of the compilation. So, wait for the release and then you can work out what the changes are.
Q: This is not speculative, this is arithmetic. Arithmetically, the deflator is lower, so the GDP should be higher, the real GDP.
A: You can do you compilation. I will talk about the implication when the numbers are ready.
Q: The new WPI series actually, all the five years that you have given us, the new WPI series is lower than the previous 2012-2013, 2013-2014, 2014-2015. So, when will you give us the entire back series of GDP?
A: Those details, we will incorporate it into the estimate, but will the back series, etc. from 2011-2012 onwards be released, I am not exactly sure. We will let you know of the full details, but we will release enough material at the time of the May 31 release to certainly make use of the numbers. If the full set can be released, we will let you know at that time also.
Q: Let me come to the changes in the IIP. Can we hope that because you have incorporated more number of items and more people from whom you will draw the data, we should have less volatility?
A: To the extent that volatility was caused on account of inadequate number of reporting entities or inadequate weight structure, yes. But I should point out that whatever is natural volatility in production will not be excluded in this process. If the reporting entities for any reason see a drop in their production, that will showing up in the index and that is what an index should do.
Q: That is true, but for capital goods, for instance, you have included work in progress. So, to that extent, that also irons out volatility in capital goods, right?
A: That is correct and that is why it has been done. So, methodologically, whatever we could do to improve the sources of volatility which are on extraneous factors, we are trying to minimise that, but please understand, production is inherently volatile. Sometimes, an entity can increase production, it can reduce production and the index should reflect that.
Q: That is fair point, but we had that rogue element, rubber insulated cables, which would be either 120 percent up or 120 percent down and have a very volatile impact on the IIP. Is that out of the way?
A: There were two elements to it. Rubber insulated cable, as a segment, was volatile because it was highly dependent on the demand from a single purchaser. But the impact it had because the weighting diagram was outdated has been minimised. Please understand, an entity or a production entity can still suffer changes in production because of changes in demand characteristics.
What I am trying to point out is that we have brought the weighting diagram in line with the current production structure and we have also now introduced a review process. We will monitor changes in the production structure if any adjustments need to be made. But there is an inherent naturalness to output volatility which will not go away. This is a point which I used to, earlier, make also when people used to complain about volatility.
Q: Let me come to only the 2016-2017 industrial output. While the new series industrial output is higher than the old series in every year, the difference in 2016-2017 is massive. 0.7 percent as per the old one and 5 percent as per the new one. First of all, why such a big difference?
A: The why lies in the compositional elements and in 2016-2017, the industry which you referred to, rubber insulated cables, if you look at it for a large number of months was contributing very significantly negative. In fact, one of the reasons, if we had started giving you the top-five negative contributions as part of the IIP release was to enable people to make adjustments if they wished to.
A good part of the answer lies there. But other elements would also be there and it is very hard for me to pin down all the reasons. We have given something in the press release of the top major contributors to positive growth and so on and so forth.
Q: Primarily because this IIP difference is so big and the WPI difference also is about two percentage points compared to old series and the new series for FY16-FY17, once again my question, surely the impact is going to be positive on FY17 GDP, would it not?
A: It is not as straightforward because IIP is incorporated into GDP calculations not as an aggregate index, but only for capturing the value added in quasi-corporate manufacturing. Now, the quasi-corporate manufacturing is not the full index, but certain two digit categories, so you have to first take that into account. Then, the quarterly share is used for quaterising annual production for all the segments.
How that impacts will have to be seen based on different two digits which is why I say that converting from IIP to GDP is a complex exercise and I do not want to make comments on what the things are going to be. As analysts, you are perfectly willing to make whatever speculation you like.
Q: One of the complaints with the GDP was that a lot of the deflation is done by WPI, almost 70 percent of the items are being deflated by WPI and only maybe 30 percent by consumer price index (CPI). Are you changing that to deflate more of the elements by CPI?
A: No, because if you remember, when we had discussed deflators, I had said, we do not use WPI or CPI as indices. We try to choose for each compilation category of national accounts a deflator which is most appropriate from the viewpoint of that compilation category. Now, depending on whether price data for that is available in WPI or CPI, we use the appropriate data. It is not done by picking an index and using it. That is not done and in fact, I had explained the process of deflator compilation in detail. It is there in our brochure which has explained the new series methodology. Those issues on how deflators are chosen and so on will continue in GDP compilation in this series and in the future.
Q: I have a last couple of questions on labour data. The Prime Minister has raised that issue and told the Niti Aayog as well to look into labour statistics. Will the Central Statistics Office (CSO) come with a more refined and usable labour statistics? At the moment, we only have surveys.
A: Let me put it this way. First of all, surveys as a source of labour data are important not just in India, but all over the world. We are going to be introducing, we have introduced in fact a National Sample Survey (NSS) has launched a regular labour for survey which will be giving both annual and quarterly estimates of employment as measured through the labour force survey, the details are there. That has been launched, that will continue.
But there is yes, a source of employment data which comes from administrative statistics from establishments which employ people and better capturing the data for those. Now, the task force which you are referring to is trying to see if the administrative statistics pertaining to employment particularly from a variety of administrative records can be better made use of and can be improved, that exercise, you should direct to the task force as and when it completes it works, but that is the intent of that exercise.
Q: I would assume that the Niti Aayog task force will work with you and through you.
A: Yes, I am part of that task force, but I am saying, both are important sources of employment data and the employment data coming from administrative records is a very important source of data and it certainly needs to be looked at.
Q: So, should we get another probably some kind of a national labour figure sometime in FY18? A national unemployment figure?
A: You will certainly get the labour force survey data from the NSS labour force surveys which are what you have been using earlier also and on an annual basis from next year. Exact timelines we will release as and when we are ready to release them.
Discover the latest Business News, Sensex, and Nifty updates. Obtain Personal Finance insights, tax queries, and expert opinions on Moneycontrol or download the Moneycontrol App to stay updated!
