Podcast | Digging Deeper – GDP Data Revision: Reading the fineprint in the year of the general election

February 12
23:29 2019


Moneycontrol Contributors

It’s a bird! It’s a plane! Or perhaps it is the lightning fast Vande Bharat Express! Or the zooming Indian economy that is slated to become the world’s second fastest by the year 2030, as per one government claim. Or perhaps it is India’s revised GDP data that makes us all look in the pink of economic wellbeing.

In the glut of politically driven messaging that is out there in the run up to the election, it is often hard to tell fact from fiction. What one side debunks, the other claims as masterstrokes of a mastermind. What one side hails, the other is quick to dub #fail.

There is understandable urgency on both sides because the stakes are high in the year of the big, fat Indian elections where 30 million first-time voters are up for grabs, and it is after all a battle of perception that must be won by any party hoping to run the country for the next five years.

Claims and counter-claims aside, on this edition of Digging Deeper with Moneycontrol, we will focus on a subject we have revisited in the past. That of GDP revision, but before we get there, just a brief – and we promise you water-tight brevity – look at some of the controversies pertaining to data that have erupted all around us in recent times. My name is Rakesh Sharma, and a revision of the GDP number revisions is what we offer on this class of Digging Deeper with Moneycontrol.

Claims and rebuttals

Speaking of trains, since we began with the mention of one, Devangshu Datta of   Scroll recently linked the imagery of a train to the idea of a national economy, with a “let’s pretend that the train is moving” game thrown in. He recalls how statisticians of the erstwhile Soviet Union spent much of their time cooking up fantasies about high growth economy, while concealing telltale signs of stagnation. One can argue that it’s much the same tactic at work in present day North Korea, or dare one say it, Russia. Datta thinks the present regime in India is doing pretty much the same. Not to bring in the Third Reich for no reason, but it is not so much the ‘Big Lie’ we need to worry about as much as a profusion of little, regular ones. To strengthen his argument, Datta cites the example of the last two independent members of the National Statistical Commission – PC Mohanan and JV Meenaxi – who resigned due to problems with the government’s approach to statistical data.

We quote further, “The new ‘back series’ of GDP data was released without the commission signing off on it (the version prepared by the commission was rejected). The government’s refusal to release the National Sample Survey Organisation’s Periodic Labour Force Survey, or PLFS, for 2017-’18 was apparently the last straw for these two statisticians. According to them, the report was readied in December.”

The leaked PLFS as has been reported widely, has shown that unemployment hit a 45-year high of 6.1% in 2017-’18, with joblessness among the “youth” being far higher.

And who or what was to be blamed?

The PLFS conducted between July 2017 and June 2018 puts both demonetisation (declared in November 2016) and the Goods and Services Tax (launched in July 2017) squarely in the dock, according to the piece. The government, the piece informs, has also stopped the Labour Bureau’s Quarterly Employment Surveys, removing another useful data source.

More controversies and confusion over GDP revision

James Wilson, a civil engineer and data cruncher, recently wrote in The Telegraph about how the chief of the Economic Advisory Council, Bibek Debroy, declared recently that a new national sample survey would be conducted by the government to show there is substantial job creation, thereby countering the latest report by the National Sample Survey Office (NSSO).

He also recalls how NITI-Aayog CEO Amitabh Kant and vice-chairman Rajiv Kumar struggled to pull off a satisfactory damage control.

But what has left many observers confused is the tweaking of Gross Domestic Product or GDP rates.

Wilson writes, “In 2015, the Central Statistical Office (CSO) announced two changes in the way GDP is calculated — the base year for all calculations was revised from 2004-05 to 2011-12, and the methodology was changed. These made the GDP growth rate of 2014-15 jump from 5.5 per cent to 7.4 per cent, surpassing even China’s growth rate.”

This, said The Telegraph piece, bewildered leading economists because globally, whenever such changes are done, past data are also revised concurrently. Not only had CSO failed to publish the GDP growth figures for the years prior to 2011-12, the demand was ignored for three years. In August 2018, when a draft report on the GDP back series was released by a committee set up by the NSC, it indicated that growth figures under the UPA were higher than the old methodology indicated. The report was swiftly removed from its website.

The GDP figures were then finally released by NITI-Aayog vice-chairman Rajiv Kumar on November 28, 2018. And the piece recalls that they indicated that the revised average GDP growth between 2005-06 and 2013-14 — all UPA years — stood at 6.7 per cent, against 7.35 per cent, as recorded in the first four years of the NDA government.

There were of course naysayers. Like the former chairman of the NSC, Pronab Sen, cited in the piece, who questioned the NITI-Aayog’s unusual move of releasing GDP data, and pointed out that it is the prerogative of the CSO. We quote, “Sen also stated that a proper study of the methodology used to deduce real growth in each segment of the economy is warranted. The latest upward revision of GDP growth was done for 2016-17 and 2017-18 on January 31, 2019.” The revised figures for 2016-17 are 8.2 per cent, showing that maximum growth happened during the demonetisation year. That has stretched the bounds of collective credulity.

In a previous Money Control report, the legendary Mark Mobius had also expressed shock at the recent GDP revision and said that the Indian government should be transparent and open all its books to the public. Referring to the revision of the FY18 gross domestic product (GDP) growth rate to 7.2 percent from 6.7 percent, Mobius told CNBC TV18 that the revision of the government data had, well, shocked him.

He said, “The current growth rate of 8 percent makes sense for India. We have to take a lot of these statistics with a grain of salt but I think it is quite a shock at this stage to see this happen in India.”

He did also say though that a slight deviation in data does not matter and that the Indian numbers are good and credible despite revision. He also foresaw The Reserve Bank of India (RBI) and the government infusing capital to bail out troubled banks.

Data crunching continues in the meanwhile

The authenticity of data released by the government has however repeatedly been questioned. The government for instance was called out during and after demonetisation for contradictory and revisionist claims that were seemingly made to restore the faith of the prospective voters and diehard supporters, and to silence habitual critics.

The Telegraph recalls how after eight-and-a-half months, the RBI Annual Report of 2016-17 disclosed on June 30, 2017, that the RBI had received Rs 15.28 lakh crore of demonetised currency — in other words more than 99 per cent of the demonetised currency.

Claims that demonetisation would turn India into a cashless, digital economy and a more tax-compliant society, have also been debunked.

In a latest Business Today report though, a claim has been made that mobile wallet transactions in India increased 40 times in five years and that the digital economy in India was growing 1.5 times faster than the global average. Who made this claim? It was made by the National Cyber Security Coordinator in the PMO, Gulshan Rai. That is however a story for another day and another time.

As for the claim about Demonetisation making India a tax complaint country, as The Telegraph piece said, an examination of the Central Board of Direct Taxation data available in the public domain reveals that there was an 11.6 per cent growth in the number of income tax payers in 2013-14. It then fell to 8.3 per cent and 7.5 per cent in the next two years, increased to 12.7 per cent in 2016-17, but again fell to 6.9 per cent in 2017-18.

The report also cites tragically inadequate farmer suicide data. The last NCRB report published was in 2016 and showed that till 2015, in India, every hour a farmer committed suicide. But the Union agriculture minister, Radha Mohan Singh, told Parliament that the government had no data on farmer suicides since 2016. Among these stories of revisionism, we will concentrate on just the one for today – GDP numbers.

What the government has to say

A recent report in Khaleej Times cited India’s Chief Economic Adviser Krishnamurthy Subramanian who said that he is not worried about the upward revision of fiscal deficit target for 2019-20 to 3.4 per cent as he believes it will actually come down to 3.1 per cent after the revised GDP numbers are incorporated.

Khaleej Times said, “Subramanian also expresses satisfaction with the efforts taken by the government towards fiscal consolidation and said the country remained fiscally disciplined despite greater devolution of 42 per cent to states and implementation of the Seventh Pay Commission.”

In a post-budget interview, he told IANS that the budgeted estimate for fiscal deficit (for FY19) was 3.34 per cent. “The actual number has come to be 3.36 per cent. And as is the practice, it is rounded off to 3.4 per cent. So, actual slippage is 0.02 per cent which is just a few thousand crores.”

He added the 3.36 per cent figure was derived based on the old GDP estimates and had not used the revised estimates.

To refresh your memory a bit, the government had revised the GDP growth rates by 110 basis points from 7.1 per cent to 8.2 per cent for 2016-17 and by 50 basis points from 6.7 per cent to 7.2 per cent for fiscal 2017-18.

About the alarming job deficit numbers, he said that problem was a discrepancy between supply and demand and told IANS, “It’s done through skilling, which, by its very nature, doesn’t happen overnight. It takes time. That’s where Skill India campaign and investments will start showing their results.”

So just how good are we at research and data gathering?

Since most controversies we seem to be addressing in this space, begin with how data is gathered, it is a valid enough question. Are we good at research? This is a detour, but an essential one to understand the landscape in which we are operating.

A recent Moneycontrol piece reported that while India ranks 8th in the quality of private and public research institutions across 112 countries, according to the World Economic Forum’s Global Competitive Index 2018, it scores just 0.42 out of 5 in the quality of research institutions index.

The primary reason for low-quality researches, according to the piece, is the low spending on research and development (R&D). India spends just 0.62 percent of GDP on research, according to R&D Data Release 2018 by UNESCO Institute for Statistics.

We quote from the piece, “Interestingly, India’s spend on R&D has been rising but the percentage of GDP devoted towards it has been declining. The low spends answer why India lacks the resources required for conducting high-quality researches. Time and again, questions have been raised on the quality and authenticity of the research output in India and policymakers have long been aware of the dismal state of research.”

The scarcity of compensation or funds for scholars, says the piece, however, is not the only reason behind the research crisis in the country. The woes of India’s education system are rooted in early schooling.

We quote, “The quality of education imparted to students in government, as well as private schools, has come under the scanner after the Annual Status of Education Report, 2017 stated that about 57 percent of the country’s youth aged 14-18 years were unable to solve a basic division problem.

The survey, which focused on the quality of education at various institutes, said that teens couldn’t successfully perform simple activities like counting money, adding weights and telling time.”

The government policies, informs the piece, too play a major role in the publication of a large number of bogus journals and researches and India needs to support researches and development within the country, bring in laws to check the loopholes that exist in the laws and set up a strict monitoring body to check the research quality.

The backstory of the GDP tweaks

Because numbers can be baffling and because they often get lost in the din of excessive information, let us just break this story down for you, repetitions and all.

The controversy over GDP tweaking resurfaced this year when in January 2019, the government revised its forecast for GDP growth for 2017-18 to 7.2% from the earlier estimate of 6.7%. It also revised the actual growth rate in 2016-17 to 8.2% from the 7.1% estimated earlier.

The revisions were made by the Ministry of Statistics and Programme Implementation in its First Revised Estimates of National Income, Consumption Expenditure, Saving and Capital Formation, 2017-18. The numbers painted the demonetisation year of 2016-17 in a rosy light and showed a strong growth in sectors widely agreed to have been badly hit by the exercise.

The Hindu quoted DK Srivastava, chief policy adviser at EY India who said, “For 2016-17, this hike of 1.1 percentage points in GDP growth comes as a surprise because it was the Demonetisation year. If you look at the demand side, the main factor for this is the increase in private final consumption expenditure, which has also increased 1 percentage point. That is inconsistent with the idea of people having less cash to make purchases.”

He further said in the piece that the main driver of the upward revision on the output side in 2016-17 was the construction sector, which has been revised upwards by 4.7 percentage points. “Construction is also a sector which has a large informal sector component and all earlier analyses had indicated that demonetisation adversely affected the informal sectors,” Srivastava said.

What did the government have to say? Well, it said, “The estimates of GDP and other aggregates for 2015-16 and 2016-17 have undergone revision on account of the use of the latest data available on agricultural production, industrial production, and government expenditure (replacing the Revised Estimates with Actual for 2016-17) and also more comprehensive data available from various source agencies like the MCA and the NABARD and State/Union Territory Directorates of Economics and Statistics.”

Looking at 2017-18, analysts told The Hindu that the government’s explanation does not seem credible because the two main drivers of the upward revision — the mining and quarrying sector and the public administration sector — both have data that is compiled by the government itself and so should not have undergone such a vast revision. Madan Sabnavis, Chief Economist at CARE Ratings, said, “It is not something we are able to explain. The revisions where the growth rates are going up by 1 percentage point and 0.5 percentage points are difficult to reconcile with the ground-level facts.”

What the earlier revision was all about

Let us also recall one of our earliest podcasts on the subject in November 2018, when the NITI Aayog and the Central Statistics Office’s (CSO) released the ‘back-series’ of India’s gross domestic product (GDP) data from 2005-06. This was done, as we have mentioned repeatedly, using a new methodology, which changed the base year to 2011-12. The result? The new data released on November 28, downgraded UPA years’ GDP growth rates by shaving off the previous estimates by a few percentage points in several years.

On the surface, as Gaurav Choudhury reported for Moneycontrol, the report conveyed that the Indian economy did not grow at the pace it was earlier made out to be during the UPA regime. Those double-digit growths, the government said, did not happen.

Claims that this methodology to gauge past GDP growth rates was much better than the previous one were made by Rajiv Kumar, Vice Chairman of government’s policy think-tank NITI Aayog.

He said, “The new series (2011-12) is far more superior and has made some significant methodological improvements.” He also added that the government had consulted experts such as member of the monetary policy committee Chetan Ghate, former chief statistician TCA Anant, among others.

Contentious numbers

At the heart of the controversy caused by the revised numbers was the debunking of a previous estimate that India had clocked double-digit growth of 10.3 percent in 2010-11. The new estimate has revised this to 8.5 percent. And as Money Control reported, the revision has reduced GDP growth rates of 9.3 percent, 9.3 percent and 9.8 percent in 2005-06, 2006-07, 2007-08 respectively, to 7.9 percent, 8.1 percent and 7.7 percent respectively. As mentioned in the Moneycontrol report, in 2008-09, growth declined to 3.9 percent due to the global financial crisis. However, new estimates show that the growth rate was overestimated and economy actually grew 3.1 percent.

Finance Minister Arun Jaitley had also stepped in to defend the revised GDP numbers for the UPA era and said CSO was a credible organisation. He said, “I don’t think any service is being rendered by people who choose to discredit a highly credible organisation like the CSO because its data is based on facts and the revised formulations which is a continuous exercise because every time you try and improve upon the formulations to make them more representative of the real state of economy.”

Moneycontrol also cited government officials involved in the compiling of the new data who say that the earlier series did not capture financial services value addition in its entirety. It captured only a few mutual funds (primarily UTI) and estimates for the Non-Government Non-Banking Finance Companies as compiled by RBI, apart from banking and insurance activity.

CSO said in a releas, “The methodology for preparing the back-series estimates for the years 2004-05 to 2010-11 is largely the same as the methodology followed in the new base (2011-12). In certain cases, owing to the limitations of the availability of data, either splicing method or ratios observed in the estimates in base year 2011-12 have been applied.”

Chief Statistician Pravin Srivastava was quoted by various news sources as saying that the revision of GDP back series data is a robust one, with the methodology now being comparable to how other countries crunch data.

In this series, the coverage of financial sector was expanded by including stockbrokers, stock exchanges, asset management companies, mutual funds and pension funds, as well as the regulatory bodies, SEBI, PFRDA and IRDA.

We wrote on Moneycontrol, “This probably explains the sharp drop in growth rate to 3.1 percent in 2009-10, mirroring a deeper impact of the global financial crisis of 2008 on the Indian economy that previously thought. In January 2015, the CSO had released new set of national accounts, revising the base year from 2004-05 to 2011-12. In the new series, the GDP at factor cost was replaced with Gross Value Added (GVA) at basic prices was adopted, as per international standard.”

A credible method?

This method, said multiple news sources, adopted a gross value added (GVA)-based approach as compared to a predominantly volume-based calculation.

We quote Moneycontrol, “Earlier, the index of industrial production (IIP) served as the primary metric for gauging manufacturing and trading activity. This limited the calculation to only number of units produced, and did not distinguish between a sedan and an entry-level car. Organised industrial activity was based on IIP and it used to get updated two years later based on data coming in from the Annual Survey of Industries (ASI), which captures goods’ value at the factory gate only for firms registered under the Factories Act. Now, the corporate affairs ministry’s MCA 21 records, a comprehensive compendium of balance sheet data of about 5,00,000 firms.”

Further, as the piece explained, the base year of the national accounts is the year chosen to enable inter-year comparisons. It is changed periodically to factor in structural changes in the economy. The new series, as we told you before, changed the base to 2011-12  from 2004-05.

Moneycontrol again, “The base year is important in GDP calculation as factors such as purchasing power and enables calculation of inflation-adjusted growth estimates. Labour income is calculated differently in the new method, where different weights are assigned on whether one was an owner, a hired professional or a helper using a method called “effective labour input, Importantly, the new series uses National Sample Survey Organisation’s  2011-12 establishment survey to calculate trading income, compared to the 1999 survey data used in the earlier series.”

The criticism

The recalibration of the previous government’s economic report card just before elections has had its share of critics and the opposition, primarily Congress accused the government of manipulating the GDP data of previous years for political gain and in a desperate attempt to undermine India’s growth story over the last 15 years. Reactions came in hard and harsh.

Congress chief spokesperson Randeep Surjewala termed the step a “classic” case where “the operation is successful but the patient is dead.” Said Randeep Surjewala, “Modi Government and its puppet Niti Aayog want the people to believe that 2+2=8.

Such is the gimmickry, jugglery, trickery and chicanery being sold as ‘back series data.” Calling the new GDP, as “Gimmickry Data Product’, the Congress leader said the economic mismanagement created by demonetisation, and what he termed as a “flawed” GST and “tax terrorism” had led to economic anarchy. And this had led to what he called was a malicious and fraudulent jugglery of GDP to hide the enormous body blow caused to India’s economy.

The real blame he said could be attributed to, “Modinomics plus a Pakoda economic vision.”

Just goes to show how frayed political tempers are at this junction with the general election only a nudge away.

Surjewala also alleged that since May 2014, there had been a determined effort to disparage the record of the two UPA governments under Dr Manmohan Singh during 2004-2009 and 2009-2014. Changing the base year from 2004-05 to 2011-12, was part of the same strategy, he said.

He continued, “Congress had demanded that the back series data of GVA and GDP should be calculated and released so that meaningful comparisons can be made.” He added that the back series data as released by National Statistical Commission (NSC) in August 2018 is available in public domain and numbers prove that the truth cannot be suppressed. He also claimed that UPA-1 and UPA-2 delivered the highest decadal growth of 8.13 percent at factor cost since Independence and during this period 140 million people were lifted out of poverty. Adding also that by the time the UPA left office, economic growth had recovered to 6.39 percent at market prices in 2013-14, the fiscal deficit had been reduced to 4.48 percent and the current account deficit contained at 1.7 percent.

He said that the revised ‘New Series’ by the Ministry of Statistics and Niti Aayog had undermined the NSC (National Statistical Commission), the autonomous body for deciding data transparently and in accordance with global standards of calculating GDP as per Market Price Linked Methodology.

Surjewala again, “Today, the Modi government deceived the country by surreptitiously changing the methodology in order to manipulate GDP data. This is called ‘Excel Sheet Management’ i.e deciding the result first and filling the marks later. The new methodology is GVA based, which does not take into account tax+ subsidies and consequently is extremely flawed to sub serve the myopic interests of the Modi government.”

Senior Congress leader and former Union finance minister P Chidambaram also said that the revised GDP numbers released by the Niti Aayog were a ‘hatchet job’ and termed it a ‘joke’.

The NITI Aayog Vice Chairman Rajiv Kumar though stated the government had no intention to mislead or do something purposefully that did not reflect the reality.

Decoding the complexity

For those feeling confounded by the numbers in the controversy, Moneycontrol’s own Gaurav Choudhury broke down some of the terms being used in the arguments. And we present that piece partially.

What is GDP?

Gross Domestic Product or GDP represents the total value of all the final goods and services that are produced within a country’s borders within a particular time period, typically a year or a quarter.

Real and nominal GDP?

Nominal GDP is calculated at current prices. Real GDP is GDP adjusted for inflation.

What is a “base year”?

The base year of the national accounts is chosen to enable inter-year comparisons. It gives an idea about changes in purchasing power and enables calculation of inflation-adjusted growth estimates.

The new series changes the base to 2011-12  from 2004-05. Every national accounts dataset gives GDP calculations for two years: 2011-12 and the current year.

When was the new series launched?

A decision to change the GDP calculation method was taken during the UPA-II years. The NDA government launched the first set of data, giving out levels of GDP and growth rates from 2011-12.

Back series

The back series data serves as a link between the old and new formulae. The back series is aimed at calculating/updating national accounts using the new formula to help allow inter-year comparisons and enable better economic forecasting.

Varied perspectives

Vivek Kaul, the author of Easy Money Trilogy wrote a piece in Firstpost to address the controversy. He said, “The biggest takeaway from this exercise is that the economic growth between 2004-2005 and 2011-2012, and the economic growth between 2011-2012 and 2017-2018, was more or less the same.  Between 2004-2005 and 2011-2012, the Indian economy grew at 6.89 percent per year. Between 2011-2012 and 2017-2018, it grew at 6.87 percent per year, which means there was barely any difference. For anyone who has lived through these two periods, this just doesn’t feel right.”

Another point he made is that if the Indian economy grew at more or less the same rate between 2004-2005 and 2011-2012, as it did between 2011-2012 and 2017-2018, then this should reflect in other ‘real’ economic indicators as well. “The GDP number is ultimately a theoretical construct and that being the case, the way it goes, should also be reflected in the high speed ‘real’ economic indicators like car sales, two-wheeler sales, tractor sales, commercial vehicle sales, non-food credit, corporate tax, personal income tax and a whole host of other economic indicators.”

He cited multiple indicators to claim that the economic growth between 2004-2005 and 2011-2012 should have been much higher than the economic growth between 2011-2012 and 2017-2018. But as per the new GDP series, this is clearly not the case. “Given this, the new GDP series does not pass this basic smell test.”

What does make sense is that excessive lending between 2004-2005 and 2011-2012, led to a huge amount of bad loans in the years to come, especially for public sector banks. This to some extent has held growth back, post-2011-2012, he said.

He cited former Reserve Bank of India governor D Subbarao who had written this in his 2016 book “Who Moved My Interest Rate,” perhaps foreshadowing this controversy and we quote, “The poor quality of data is compounded by frequent and significant revisions, especially in data relating to output [GDP] and inflation which are at the heart of monetary policy.” Vivek also recalls former RBI governor YV Reddy’s quip, ” everywhere around the world, the future is uncertain; in India, even the past is uncertain.”

Another view is that needless controversy could have been avoided if only the Central Statistical Office (CSO) had not organised a joint press conference with NITI Aayog representatives especially because the NITI Aayog should have no role in calculating the GDP numbers, as it is the job of CSO.

Former chief statistician Pronab Sen told PTI, “the integrity of CSO has been dented in the eyes of users. We have always had a system that data CSO brings out is completely removed from the political interference. Even the Prime Minister would get to know of the numbers just before they are released. Now to do that alongside NITI Aayog, which is a political institution like the (previous) Planning Commission was, is essentially diluting the integrity of the CSO.”

And as Times of India cited Sen, “When a political institution releases national statistical data, it puts a huge question mark on the credibility of the data and the political independence of the statistical agencies.”

Regardless of what projections and data revisions proclaim, finally it is the citizenry that will decide whether the train to prosperity moved fast enough to take them to the Promised Land or stayed stuck in reams of economic jargon. The voter who casts the ballot will have the final word and will soon decide the winners and the losers of the ongoing war of perception.

Related Articles