 Hello and welcome to NewsClick. Today we have with us Prof. Abhijit Sen who has been former Chairman of Agriculture Commission and also a member of Planning Commission for a very long time. Abhijit, we have had this controversy about the GDP back series. Can you tell us in layman's language what is the GDP forward series if we have a back series and what is the back series, what does it really mean? Well, I mean the GDP is calculated every year by the Central Statistical Organization and usually what they do any year is use new numbers, apply an existing method and bring out a number for that year and this goes on for as long as the method that they're using remains constant. It's just the new numbers which come in. And it's on a particular base year. It's on a base year that base year is really the year at which you start this methodology that you're going to follow for the next few years and it also gives you the year to which the constant prices that is you're fixing prices when you need a real series as distinct from a nominal series. Nominal series, the face value of the nominal series is simply the income of the nation at current prices. Now that is simply if you had everybody's income or every sector's income, you just add it up. The real series is slightly more difficult because you've got to deflate everything by the inflation and that inflation rate might vary from sector to sector and that's where a complication lies because that's not the nominal income and that depends on the base year quite a bit. So where are you starting your inflation measurement from? Now this process of using new numbers and applying an existing method to it goes on as long as the method itself doesn't change. So it also gives a year to year comparison of the growth because you're doing essentially pegging it to a base year. So it's not just the nominal rate of growth but the actual rate of growth. The real rate of growth deflated or compensating inflation in some sense. Now the point really is that existing methodology. Now this existing methodology can last for a fairly long time or for a shorter time. Now what happened in the case that we're talking about is that the methodology was changed somewhere in 2015 to move from a base of 2004-5 to a new base of 2011-12. So there had been an old methodology or the previous methodology before that you've got other methodologies which had been there from 2004-5 to 2012-13. Now you've got a new methodology coming in from 2011-12 onwards. So the existing GDP series is the 2011-12 GDP series and there has been a controversy regarding the change in methodology and essentially that change in methodology had to do with the shift from two types of shifts. One the major shift which is the shift in the sources of data used was that they moved from a whole set of different sources to a common source which is the company law. The Ministry of Company Affairs has balance sheets and they added up all of these balance sheets irrespective of the sectors and that gave you a nominal number because balance sheets are always in current terms. Nobody deflates anything for inflation in that and to that you applied whatever inflation thing. Now before that what you had was different sectors were treated differently. Many of these sectors were treated differently. In some cases you did use balance sheets so you went from nominal to the real. In others you actually use measures of the real itself to actually get the real thing first and then multiplied it by a price index to give you a nominal thing. So the way you came were different and this created a problem because for example politically you had people like Yashman Sena and others claiming that the new series was overestimating growth by roughly 2% per annum and that was based on a comparison of the old series for 12, 13 and 11, 12 compared to the new series estimates of those same 2 years. So it seemed that it seemed that tweaking to improve the growth. It is not tweaking. You could either say tweaking that you could apply a political motive to it or you could say okay you had this change in methodology which somehow gives you a higher number and the demand then was that if this methodology gives you a higher number why don't you apply this methodology back to see whether it gives you higher numbers in the previous things as well. The CSO has always calculated the back series where it does a change. CSO has not always calculated a back series but certainly since 93, 94 they have been giving back series every time they have changed the methodology and the base here. They did it in 2004, 5. They had done it before in 99, 2000. They had even done it in 2003. Before that you don't really have it but the point is that you didn't have as large a change in the data source as you did this year and in addition to that there was a second set of changes which happened in 2011's 12 series is that you moved to a complete synchrony which something called the SNA. It's a United Nations classification. What that meant was conceptual changes. They don't add up to too much but they are conceptual changes. What is GDP? What is GDP? How do you define a GDP as distinct value added? We used to have something called GDP at market prices and GDP at factor cost. The idea of GDP at factor cost is not there in the SNA so they moved to something called the GVA at basic prices. Now all of these require some tweaking so all of that tweaking also took place. Now what came out of the wash is as I described something which for the two years that they were common gave you a higher growth rate by 2% in the new series as compared to the old series and this started off a row about let's get a back series. Let's get a back series. Why haven't you done it? You've done it in the past. Why aren't you doing it now? Now the statistics ministry or the CSO tried to do some of this but at some point and used to promise to everybody that AI is going to come, it's going to come but at some point they threw up their hands and said it cannot come because we don't have the data. Dr. Pradamsen said a deus clique that actually they had a back series quite soon after this deus series was done and actually the head of DT Iogh, Professor Pangariya at that stage actually stopped it but leaving that aside you know the question is that there could have been real problems but there could have also been other issues. I mean I'm not privy to what happened within the CSO but it's certainly true that TCA Anand who was then the chief statistician of the country for a time he was said that you know we'll be releasing it soon and then at some point he said we won't be releasing it at all. Okay, so that's there is some basis to say it. Now however in July this year the National Statistical Commission which is really supposed to actually monitor it's not doing the work it's supposed to monitor the work of the statistical system actually brought out a paper which was on a different issue altogether to the real sector measurements in which one chapter had a back series. Dr. Mandel's committee. It was Dr. Mandel's committee this particular work was done by the NIPFP, Bhanu Murthy did it but it was the first which came as a back series of some kind following a certain methodology and they're careful enough to say that there can be an alternative methodology is possible this is one among at least three possible methodologies. Now what that did was essentially bumped up the growth rate a little bit not as much as 2% but almost 1% and essentially it got those would be saying that you know the growth rate in the new series is overestimating it compared to the old series it obviously supported them and also made the NDA-UPA comparison a little shall we say in weighed in favor of the UPA years because it bumped up those growth rates a bit. It bumped up the growth rates from 2004 to 2011 which were all UPA years and therefore essentially an UPA NDA comparison which even without if you simply took the old series and the new series the UPA was looking better and fairly significantly better and this made it look even more significantly better. Now the matter could have rested there that a set could have been shall but obviously there was a certain amount of churning within the statistical department possibly but the way this was really suggestive not the statistical department which started the churning it started from NETI-IO. So, you have this suddenly one morning nobody expecting it you get a new set of back series released under the chairmanship of the back chairman of the NETI-IO whose has simply no business to be there. The statistical system reports to itself NETI-IO can look at it after all NETI-IO needs to have numbers but that number comes to it. It does not generate. It does not generate numbers. It can sit and does sit on working groups which lead to methodologies but those working groups must be there they are part of the statistical system and NETI-IO if represented there is represented as part of that system. It is usually chaired by someone within the statistical system and in any case the reporting authority for the statistical system is not the NETI-IO is the National Statistical Commission and the National Statistical Commission had actually released the previous thing which was Mandelstaff. So, this was a bit of out of the hat coming from NETI-IO again sitting so I mean questions of political propriety come along and I think it been expressed by a lot of people and it is important because the whole reason the statistical commission was set up was to ensure that statistics was above manipulation by government. So, keep it isolated or autonomous of the political system. The government would like to show itself in a better light and that is natural always every government and opponents might want to show it in a worse light but to actually keep it as basically purely expert technical body which establishes what the facts are on the ground. So, you had a statistical commission which was independent both of government as well of the statistics ministry and reporting only to parliament to set up to do that thing. So, it was National Statistical Commission reports to the parliament not to the government. That is what it was supposed to be. So, you got this new series suddenly which was thrust upon by thrust upon or announced by the vice-German of the NETI-IO and lo and behold what does it show? It shows that the rate of growth instead of being about a percentage higher than the old rate of growth from the old series as calculated by Mandel and reported by the statistical commission was actually to almost 2 percent less than the old series now. And this obviously has raised a huge set of questions regarding what the hell is happening to the credibility of our statistical system and that is where it stands because I think like a lot of other things of lot of other institutions which should be independent in a sense of government itself are slowly being seen to becoming much more to quote the Supreme Court pay a cage parrots. I mean they say exactly what you like them to say. Now, be it as it may be the argument put forward by the after this by again the vice-German of the NETI-IO. In fact, then you went on to say this is official that hold that old stuff which was the Mandel stuff is not really kosher. So, that's where we are so that you've got a huge defense of one as against the other and they are actually starting with the old series one is taking it up for them and the other is taking it down one is making the one was making the UPA look better the other is making the UPA look much worse. Now, to actually make sense of what is happening and I think much of what has the newspapers have actually reported on the skirmish on the politics side of it. But they have also I think reported on the content of it. I mean the numbers content of it and what almost all those who have commented on the subject have pointed out is that what seems to be happening in the new back series that is the one recently released is that it is roughly the same on the nominal. So, that if you take the nominal stuff that is GDP by expenditure or value added by sectors they seem roughly the same what seems to have changed the only thing seems to have changed is that you have used different inflation indexing. So, that when you deflate it going back you are actually in the new series assuming a higher rate of inflation than was there implicit in the old series. And so if the money numbers remain the same a higher inflation simply means that the real thing is less. But again the higher rate of inflation is not born out but the actual rate of inflation if you look at the inflation rate itself. No, actually here is where the problem really lies. The problem lies in the following and I think this is where the meat of the thing becomes true. Any GDP can be looked at either as demand or it can be looked at as supply. If you want to look at it as demand then you look at ok there is a certain income how is it being spent? And you have basically some major sectors private consumption expenditure, government consumption expenditure, investment and then exports imports and then you throw in a certain discrepancy amount. Now, what is interesting about the new series is that with the exception of private consumption expenditure almost all the other expenditure components are not only roughly the same in nominal terms their rates of growth are even the same in real terms as the back series. So, the only thing which seems to have changed on the expenditure side by any considerable amount is private consumption expenditure. So, it is the only the inflation for that for private consumption expenditure which seems to have been jacked up ok. Whereas for investment, for trade, for government consumption expenditure all of that seems to be roughly where it is it is a simple splicing. You are just keeping the rates of growth same and just adding new numbers to change for the shift in base. For private consumption expenditure the inflation rate goes up by about 2 percent and therefore, that rate of growth goes down the way 2 percent and that is where all the kale is coming from when you look at it on the expenditure side. Now, that is a bit weird ok. I mean there could be arguments for it and the real reason why it is a bit weird is when we turn to the supply side of GDP. The supply side of GDP says where is the income coming from not how is the income being spent, but where is the income coming from. So, it divides it up into sectors and ask which sector is growing at what rate. Now, here if I look at there is a lot of sectors and they are doing different things. So, the difference between the old series and the new back series are different in different sectors. Now, it would be too long to actually and I would not even have the data in my head to tell you what is happening in each sector, but there is one motor bath which comes through which is that the main difference between the old series and the new back series that has just been released relates to only two very major sectors and these two very major sectors are the sector which includes communications trade and storage not trade storage transport communication that sector, but within that sector specifically it is communication. Communication basically means telephone and the other big sector is what is called fire which is finance real estate and business services. And within that sector there are really two real sectors which are making the difference one is finance and the other is that real estate and along with it professionals business services including IT. So, the big differences are coming in these two sectors you know communication on the one hand which is really cell phones which is the big story of the year. Finance which is banking insurance and non banking financial corporations and the stock market whatever they add and the third is we cannot quite divide it up, but in that real estate and professional services which includes the IT sector and the differences are huge. In fact, if you take the GDP coming from the supply side, if you take these three or four items out they are roughly again the sum although not the components necessarily is roughly the same as the back series. But for these it is a huge decline for example communications in real terms was growing at over 20% per annum in the old series and it is down to about 5% per annum in the new series. Which is very difficult to understand with the growth of the telecom sector. You see the telecom sector roughly started growing from 2001 onwards. In fact the sharp thing was 2003 onwards. The sharp thing was really 2003 onwards and given that your base was negligible. It's a huge bump. It's a huge bump but 20% growth rate seems reasonable. In fact I would say that it's roughly comparable if not less than the number of cell phone users that we are getting per year. And the amount that they are spending and what makes it even more peculiar is that suppose you take the period after this. It's gone down from 23% to about 6% and after this it's gone up to 11%. So it comes up. It's gone up in the NDA years again. Now first of all by the time NDA year starts the base is big. Correct. There the base was small and you are having only 5% growth and here you are now having 11% growth and you've got all these telecom majors making huge losses consolidating now and they were actually coming into business. As we know they are in crisis, NPAs and so on. It simply doesn't make sense and that's one big sector. The other big sector is finance. 2004 onwards is for that till the world financial crisis breaks. That was a big time when portfolio capital started coming into the country. Like it or not that portfolio capital was getting parked in either the stock exchange and having huge repercussions on the financial system as a whole. You would have expected it to grow and the old series shows it not a great deal but a fair amount about a 15% rate of growth. Suddenly you cut that down to about 9%. Now unlike in the case of communication this 9% is still slightly higher than what you get now with the show at around 6%. But nonetheless it seems odd that the financial sector which was doing bloody well then and is doing bloody badly now is the other sector which accounts for much of this change. And the third is and I'll tell you what the reasons for these two sectors finance and communications are. It is the inflation rate. What happens in the communication sector which is most interesting is that in the old series what you showed was the prices of communications that is call charges was coming down. So that the 2011 average price was about half of what it was in 2004-5 which is not unbelievable. We all know we bought our cell phones and you know we started getting I mean you had to pay for receipts now you don't you know all. Of course call charges have gone down while the number of calls have gone up so that's really. Whereas in the new thing what they've done is they've deflated it by an index of transport and communication. A price index which is dominated by transport at a time when oil prices were going up. So what you're picking up is as against a halving of the price you've got the prices of communications actually going up by about 50%. And that's what causes that huge difference difference in the communication side is a bit similar in the banking side. It's again largely what you are actually doing in terms of the deflator. Now the banking deflator is an animal I still don't understand what are you deflating it by what are the services being offered. So you'll have to find someone more knowledgeable but it's the deflator which is playing the thing. And it is doing this at a time we know that the banking sector was doing extremely well and you're then actually cutting it down. That growth cutting it down to size. The third sector which is really this business services and real estate and dwellings. Now as far as the real estate dwellings and professional services part of it about half of it is this rent your paying yourself if you are living in your own house. The other half is other stuff which includes real estate activities as well as some professional activities including IT. Now in this sector unlike in the other the main hit is not higher inflation. It seems even nominal incomes haven't grown as fast as the old series was showing. And again in this case what you have is a cut from something in the neighborhood of 10 above 10 to 12% to down to 6%. That is the old series was showing something which is almost double the 6% of the new back series is showing and the present is about 11%. Again we get the bump up. When the real estate prices are actually going lower. Where real estate prices are down, rents are in fact down in some cases and hardly up in any. And IT is actually doing much worse than I mean in now than it was doing then. So here again it's simply not explicable by what we think the economy should be. What we think the sector was happening let's put it what should be or not should be let's not bring in value judgments. Anyone's essential instinct to say usala chhatha isala chha hai. You say ne ne tapto you know things were booming in the real estate sector they aren't now. Empty flats lying around then things are. Now these are the three sectors which explain it all on the supply side and they are surprising in themselves. But what makes the whole thing even more surprising is that these three sectors are not necessarily all to do with private consumption expenditure. They affect other parts components of expenditure as well. And somehow when you go from the supply side to the demand side on the expenditure side is only the expenditure side which is picking it up. I mean only private consumption picking it up and everything is the same. So my own instinct is it's a bit of a rush job. I wouldn't ever say that however much under pressure the CSO wouldn't maintain certain standards. So they probably tried their best to reproduce in the new back series some of the elements which are contained in the new series. So the objection if any should be to releasing the new series rather than to the methodology. They couldn't apply the methodology completely simply because one data source wasn't there which is that ministry of company affairs data is only there in driblets. Before this. Before this they couldn't use it. So they had to make do with other stuff and secondly what is dubious and never done in past back series. They have used new inflation indices to deflate. So people must have been convinced or made to be convinced about the utility of doing something which never had been done before. Norms and ratios often used for the new series have then been applied backwards. They've clearly fallen very strongly in sectors which are growth sectors. Communications, IT, you know new fangled real estate and finance whose nature is changing. So norms and things taken from 2018 let's say or 2012 do not necessarily apply to the norms which were there in 2004. So there are issues about using these norms to go back there and I don't think those things were thrashed out in full. I'm also pretty sure that much of this was done in a hurry because if it wasn't done in a hurry you would have got much greater consistency between the expenditure side and the supply side. I mean you cannot get a deflator for consumption expenditure without working out the composition of final consumption expenditure, the private final consumption expenditure. The deflator only comes out once you've calculated that at real and nominal terms and at least the press note which is provided doesn't provide that. It does not provide any rational fitness. It doesn't provide that data at all and therefore one thinks that that's still a work in progress. Sometimes they will actually produce it but you actually then use a deflator which can come out only from that work in progress and rush into a press note that shows that you know put jaldi kvazi. Or there is political expediency to contain the damage from the mandal report shall be said. Well I don't know whether there was that much of a damage. At least from the Niti Ayog's point of view who seems to have become much more of a quote-unquote political animal than would be justified by being a think tank head. Well let's put it this way Niti Ayog unlike the planning commission has really very little impact on economic policy. If you want to find work for yourself you can find work of a different sort. Public relations could be the work for the government using statistics and economic data. Thank you very much Avijit for explaining a rather difficult exercise for most people. You know people blank out when they see numbers and when they see a series they're likely to blank out even more to make this explicable to the people. This is really a challenge for all of us. Thank you very much for being with us. We'll come back to you with more such difficult exercises when we come across them.