India’s Q4 GDP Data May Be Overstated By Rs 2 Lakh Crore, Says Former Chief Statistician
Former Chief Statistician Pronab Sen dissects India’s Q4 GDP numbers.
India gross domestic product data for the January-March quarter, which showed the economy growing at its slowest pace in over a decade, may have been overstated, according to former Chief Statistician of India Pronab Sen.
“My own sense is that this particular figure that they have quoted in the current release is high by about Rs 2 lakh crore. One should knock that off,” Sen, who headed the country’s statisics ministry from 2007 to 2010, said in an interview. “My personal estimate of the loss that took place in March is about Rs 4 lakh crore.”
Sen’s assumption is based on the fact that the country was under a total lockdown for a whole week during the end of March. That would’ve meant a loss of about Rs 2 lakh crore based on the previous fiscal’s average GDP, he said. “And prior to that for three weeks we had various states that went into lockdown. The tourism sector had been wrecked.”
So all of that would total up to at least another Rs 2 lakh crore of reduction and I see they have captured only half of it.Pronab Sen, Former Chief Statistician Of India
Watch | Former Chief Statistician Pronab Sen dissects the Q4 GDP numbers.
Here’s the edited transcript of the interaction.
What do you make of the 3.1% growth number? It’s not that bad as some had expected but it’s a very early and partial read of the impact Covid-19 is having on the economy.
That is correct. I think this is an over-estimate and they have come out and pretty much said so. So I think one should be careful. My own sense is that this particular figure that they have quoted in the current crisis is high by about Rs 2 lakh crore. One should knock that off.
Why would you assume that? Could you please explain that reasoning?
My personal estimate of the loss that took place in March is about Rs 4 lakh crore. We had one week of total shutdown which at the average GDP of last year comes to about Rs 2 lakh crore. And prior to that for three weeks we had various states that went into lockdown. The tourism sector has been wrecked right across the country. So all of that would total up to at least another Rs 2 lakh crore of reduction and I see they have captured only half of it.
And this could be because of the inability to collect data ?
The bulk of the data for these estimates comes from the quarterly returns that listed companies file with SEBI. Now the government has given an extension to the reporting date and as a result the CSO probably does not have the full set of data because companies have held back reporting and will do so later. And then these numbers will undergo change depending on what comes up.
Analytically, what do we read into this data that can give us useful information about how the economy progressed in the quarter, particularly towards the end of the quarter? Do we look at sectors which were vulnerable to Covid, such as construction and manufacturing, and assess those?
Yes, that is certainly one way of reading it. But the other way which I think you should also be looking at is what is happening to the final expenditure pattern.
So if you look at things like consumption for instance. The consumption data suggests that, at least up to this point, consumption has held up. And so has fixed capital formation. And I am quite surprised so has the export of goods and services. Well, not entirely it has slipped a bit but not a whole lot. All of these are going to start looking down because in the initial phases of the lockdown, you would have people who try to keep up their consumption by drawing down their savings. But that effect is going to peter away very quickly.
Maybe over a couple of quarters, it will disappear and then you will be left with the pure reduction in income and consumption will start declining then. That’s really the big worry. So one really has to wait to see that extent to which that happens.
Now, much of that kind of analysis depends upon what the government is doing. At the moment, the government has announced two packages, one which was announced on March 25 and the other which was announced in mid-May. Now if this is all that the government is going to do in terms of fiscal support, then I am afraid the effects are going to be negative and large. Because the household consumption is going to slide and if the government consumption does not step up then the GDP will have to slide downwards.
Moreover, we know that exports have done badly in April and are doing badly in May. So you are going to see another large downside for exports. So one will have to wait for this data to come in. Unfortunately, there is so much time lag in data collection because the next quarter’s data which will give you really a good feel for what has happened during this period and what we can expect to look forward, that’s not going to come till the end of August.
I do want to expand on the points you were making on the expenditure side data. So you know in Q4 the private final consumption growth came down to 2.7% and gross capital formation was -6.5%. So you are saying they will partially capture what is happening?
They will get worse. They partially capture an initial indication of what is in store.
And will we see the government’s final expenditure data go up further?
Well it should. But the important issue is, at the moment, what we have are announcements. There has been some release of funds particularly for the first package, which was the March 25 package, in terms of transfers to Jan Dhan accounts and farmers. But the second package, a lot depends on when the money hits the ground. Because CSO will count it not at the time of announcement but at the point of expenditure. So how quickly that money comes in is really of concern.
So, for instance, the cash transfer would have hit perhaps somewhere in April and on the assumption that it got spent, that impact would show in both government and private final consumption expenditure in the April-June quarter. Am I right?
That is correct. Although it shouldn’t show up in government consumption. Because anything which is in the nature of a transfer, like the Jan Dhan deposits or PM-Kisan, these are not counted as government consumption. Government consumption only includes actual expenditure on goods and services.
What are your thoughts on the trends we’ve seen on the industry side. There was a strong 4.9% in agriculture and a decent performance in mining. If we were to exclude those two, assuming they weren’t impacted by Covid-19 at the start, should we look at industry and services up ex-agriculture and ex-mining and then see what the conditions were?
That’s right. If you want to get a sense of how much of the economy shut down, what you really need to do is exclude sectors which were exempted from the lockdown. Agriculture and mining were two. Unfortunately, in industry there was a lot which was exempted. Like a lot of petrochemicals and steel were exempted. Then one would have to drill down to the product level. But it gives you a fair idea of the proportion of economy that was impacted. And if you can get that then you can extrapolate the effect for the next two months, which is pretty much the first quarter of the new fiscal.
What do you expect policymakers to be looking at when they judge this data to try and see what the nature of the response they should be designing?
Frankly, this data is not particularly useful when you are trying to handle a crisis. It is very useful when you are trying to handle cycles. But cycles and crises are different.
In cycles, you can be slightly behind the curve and you can do an upward adjustment to take account of the fact that you are behind. But in a crisis you cannot do that. Because then things could get out of hand and you would not be able to intervene in a large enough manner to take care of the problem.
So for crises, you need data which gives you at least the sense of what is going on, practically in real time. And the only data that I know which can enable that is the GST data. And I think people underestimate the richness of that data. If you use that intelligently, you can actually locate which industries have seen what kind of drop in GST payments. That is a good sign of the slowdown in their production. That will also track recovery as they come back and restart production. It is not very good for estimating the level of GDP but it is excellent for tracking change.
Then perhaps it is inadvisable that they deferred the release of the last round of GST data.
Yes, it is very sad. I think they are using it. Whether we get it or not I think is of less importance. Whether the government is using it intensively or not is really of the moment.
Do you have an estimate of how badly the entire year will play out and what the final picture will look like?
I really don’t know as of now.