Chief Statistician TCA Anant tells CNBC-TV18 industrial output data seems to have lost momentum ever since December. “If you look at the overall picture, from October there was a small sign of an increase in industrial production but somehow from December onwards the momentum has not been maintained,” he said in an exclusive interview.
February IIP number came in below expectations at 4.1% today, but the shocker was CSO’s revision to the January figure from 6.8% to 1.1%.
Anant attributes this drastic change in data due to errors in calculating the sugar output. “This mistake is serious and it should not happen, but principally this happened because the structure of reporting does not often leave us enough time to do extensive validation possible,” he explained.
Below is an edited transcript of his interview with Latha Venkatesh and Ekta Batra. Also watch the accompanying video.
Q: This is a bit shocking for those who have to take policy decisions. We were told it was 6.8% in January, but now it is a drastic revision to 1.1%. How did that happened sir?
A: I have issued a statement along with the press release explaining that. If you look through the statement, you will see the reason for this revision given in that statement.
Q: You are speaking about the sugar output calculation?
A: That’s correct.
Q: Considering that this is such an important number, isn’t this mistake a little drastic? People could take very serious economic decisions, business decisions based on the fact that the economy is back on track. But then you are told 30 days later that it was never on track at all, it was only 1.1%.
A: I will respond to that in two ways. I agree with you that this mistake is serious and it should not happen. Principally, this happened because the structure of reporting does not often leave us enough time to do extensive validation possible.
But if you notice, every time I have been approached for comments, I have always advised caution in treating monthly data; you should look at patterns which emerge over a systematic period of time. So both these facts must be kept in mind.
Q: We are now being told that SME data was not included in the numbers that came in for October last year. Can you give us an idea of whether that data was not included only for that month and how should we treat the data for October, November and December?
A: I don’t know the context is in which you are asking the question. I will have to check that detail. But as far as I know, this IIP does not include any reports from the small scale sector because when the old index used to, there were problems of getting a regular flow. So when this new index was released, I had pointed out that Ministry of SME is doing a separate exercise to have a system of gathering data from them.
Q: We were told by the Planning Commission that some previous IIP numbers, especially October which came in at minus 4.7%, did not include some small industries data.
A: I don’t know the statement which you are referring to, so I can’t react to that.
Q: Total output in February is higher than the total output in January for capital goods, usually unlikely because there are fewer working days in the first place. Any caveats at all for this February numbers?
A: For capital goods my advice is don’t go on a year-on-year inflation growth rate. Take a look at index itself and what you will notice is that the growth rate is largely due to a base effect because the number last year in February actually came in much lower than expected.
If you look at the series which is there as part of our press release, you will see that the capital goods growth is principally a statistical artifact arising out of a base effect.
Q: The other data doubts that one gets when one gets your press release is that some numbers are often very huge. For instance, industry group printing, publishing and reproduction of recorded media has shown a growth of 60.1%. Why would printing and publishing be 60% higher than a year ago?
A: That has been noted and pointed out many times in the past.
Q: Yes, it has been happening for the past three-four months.
A: IIP indices are not meant to look at item level details. These things at an individual industry level can happen. It is very hard to explain what the reason behind it is, but these do happen. So we do not advise people to use the IIP as guides at the product level.
Q: Would you want to give us any other comment on the February data itself, since you must have supervised the data collection? In your estimate, are there any caveats that we should take and did you notice any trends which were positive or negatives?
A: The caveats are the same which I give all the time. One, do not go only on year-on-year growth rates, look at the index number as it is. In fact look at both month-on-month movements along with year-on-year growth rates and look at the trend which is emerging over a somewhat longer period of time.
If you look at the trend in the month-on-month figures over 8 to 10 months for a year, you will get a much better picture of what is actually happening.
Q: Are you getting the sense at all that we troughed out in the third quarter or you wouldn’t say that because the January revision is now making us question that assumption?
A: If you look at the overall picture, from October there was a small sign of an increase in industrial production but somehow from December onwards the momentum has not been maintained. If you go year-on-year you get very different picture because October came in on a negative growth rate.
It’s not that the momentum was really large in first place, it was not as large as two years ago. The growth is not collapsing, but it is not picking up very much either.
Q: I am not sure if you have noticed the data that we get from the HSBC PMI, but that shows a completely different trend as opposed to what we are seeing in the IIP. Any sort of comments on that?
A: Not really, because I have not analysed the difference. I do know that there is a time difference between the two, so you should match the two carefully. HSBC showed an improvement in sentiment from December-January, and that’s not completely inconsistent because what you are seeing is a growth rate which is positive unlike in October which has become negative.