Cracking the GDP mystery

Cracking the GDP mystery The CSO has been consistent with its methods, allowing little room for suspicion of window dressing

Aarati Krishnan

Did demonetisation deal a knock-out punch to the Indian economy? Or was it just a mild tap from which it is already recovering?

This debate should have been settled with the latest second advance estimates  from the Central Statistics Office (CSO) which peg FY17 GDP growth at 7.1%. But commentators who believe that the economy has suffered a debilitating blow from the note ban are not willing to rest their case here. They have flagged a long list of issues with these GDP numbers, apart from hinting that the numbers are fudged. 

However, this is taking the criticism a little too far. A closer analysis of the CSO’s estimates suggests that, contrary to perception, they do factor in the impact of the note ban. And while India’s GDP estimation method could certainly do with improvements, the CSO has been both transparent and consistent with its methods, allowing little room for suspicions of window-dressing. 
Why so high?


Well, that is a wrong reading of the numbers. To gauge the actual level of economic activity, Gross Value Added (GVA) is the more pertinent number than GDP. The GVA measures the value of output created by different segments of the economy. Indirect taxes (minus subsidies) are added to it, to arrive at the GDP.

First, the sceptics ask, how did the GDP growth for FY17 turn out to be so high? The 7.1% number is unchanged from the CSO’s initial estimates and is also well above the 6.5%-6.8% growth estimated by most private forecasters. Is the CSO implying that vacuuming up 86%  of cash in circulation had no impact on the economy?
The GVA for FY17, as per CSO data, does show a dent from demonetisation. At 6.7%, it has registered a sharp decline of 110 basis points from 7.8% (revised estimate) for FY16. 

Commentators cite some key indicators to ‘prove’ that economic activity shrank in the note ban months. For instance, two-wheeler sales collapsed by 22% year-on-year in December, banks reported anaemic loan growth at 5%, cement despatches fell by 9% and realtors saw a 40% dip in home sales.

While GVA growth is pretty close to private forecasts, what lifted the GDP is the strong 12.3% surge in indirect taxes that the CSO estimates for this fiscal. This is a plausible number, given that the Centre’s indirect tax collections already surged by 25% in April-December 2016, powered by higher excise duty on fuel and service tax. 
There is no obfuscation here, because assessing the GVA and adding back net taxes is the global prescription for GDP estimation by the output method.

Too mild?


But making up for these was the 6% rebound in agriculture (2.2% shrinkage last year), 6.8% increase in electricity, gas and water supply and a bumper 11.9% hike in ‘public administration, defence and other services’ which lifted the GVA.
So the CSO does admit that economic activity has been impacted by the note ban. But isn’t it estimating too mild an impact, with the Q3 GVA growth at 6.6%, against 7% last year?
But given that the economy is made up of literally hundreds of products and sectors, it is well within the realm of possibility that the economy did well even while these indicators slowed. For instance, for the same December month, steel output grew by 15%, power generation surged by 6% and refinery output expanded 6.4%. If bank credit slumped, companies doubled their borrowings from the bond market.

Agricultural output bounced back due to a good monsoon after consecutive drought years. Electricity generation was up on better coal availability. ‘Public administration’ reflects higher government payouts on salaries and pensions after the Seventh Pay Commission. 
Also, CSO estimates do show that some sectors of the economy took it on the chin in the demonetisation quarter. Manufacturing saw its GVA growth slide from 12.8% in Q3 2015 to 8.3% in Q3 2016. Finance, real estate and services saw growth collapse from 10.4% to 3.1%. Construction weakened from 3.2% to 2.7%.

Informal left out?

December quarter results from listed companies also provide independent confirmation that the big picture wasn’t much dented by the note ban. A Business Line analysis of over 1,700 listed companies showed that they just reported their best quarterly performance in three years, with sales growing over 9% and profits expanding 20%.
Commentary from listed firms suggests that urban discretionary purchases bounced back quickly as consumers switched to digital payments. Commodity industries, helped by global price rebound, did very well this quarter. In some sectors, business shifted from the unorganised to organised players due to digital payments.

As the cash situation normalises, these one-offs should get ironed out, moderating GVA growth for the March quarter. But this doesn’t imply that the Q3 GDP numbers are suspect.
Analysts also suspect that, in some cases, companies mopped up demonetised notes from their distribution channels and pumped them with inventory instead. (This would show up as ‘sales’ in the company’s books and as ‘output’ in GDP estimates).


A third criticism of the CSO estimate is that it fails to capture the performance of the informal economy, which clearly bore the brunt of the note ban.
This criticism is partly valid. 

More accurate estimates of what really transpired in the Indian economy post-demonetisation will be available when the CSO publishes its first revised GDP estimates, with more  ground-level data, in January 2018.

Over 40-45% of the Indian economy is informal and hardly any data points relating to it are available at a quarterly frequency. Therefore, what the CSO does to arrive at its quick estimates of the GDP is to take the available data from the organised sector and extrapolate it to infer informal activity. Thus, the GVA for agriculture is guesstimated based on kharif and rabi crop prospects. The GVA for services is inferred from sales tax collections, deposits and credit, telephone connections and so on. Manufacturing GVA uses the index of industrial production and listed company filings.
Owing to such guesswork, it is quite likely that the quarterly GVA estimate, which mainly uses data from the formal sector, painted a rosier picture of growth than the ground reality. But then, if the CSO — with its access to multiple data sources — has no way to estimate the quarterly performance of the informal sector, neither does anyone else.
 



As long as the CSO consistently follows the same method for measuring the informal sector and publicly discloses it, this is the only estimate we have to gauge economic activity. Both the methodology for estimating informal sector performance and GDP revisions are well-documented and disclosed on the Ministry of Statistics and Programme Implementation website. 
Until then, critics must follow Keynes’s tenet — when facts change, it is best to change your mind.

aarati.k@thehindu.co.in











Getty Images/iStockphoto







Did demonetisation deal a knock-out punch to the Indian economy? Or was it just a mild tap from which it is already recovering?

This debate should have been settled with the latest second advance estimates  from the Central Statistics Office (CSO) which peg FY17 GDP growth at 7.1%. But commentators who believe that the economy has suffered a debilitating blow from the note ban are not willing to rest their case here. They have flagged a long list of issues with these GDP numbers, apart from hinting that the numbers are fudged. 

However, this is taking the criticism a little too far. A closer analysis of the CSO’s estimates suggests that, contrary to perception, they do factor in the impact of the note ban. And while India’s GDP estimation method could certainly do with improvements, the CSO has been both transparent and consistent with its methods, allowing little room for suspicions of window-dressing. 
Why so high?


Well, that is a wrong reading of the numbers. To gauge the actual level of economic activity, Gross Value Added (GVA) is the more pertinent number than GDP. The GVA measures the value of output created by different segments of the economy. Indirect taxes (minus subsidies) are added to it, to arrive at the GDP.

First, the sceptics ask, how did the GDP growth for FY17 turn out to be so high? The 7.1% number is unchanged from the CSO’s initial estimates and is also well above the 6.5%-6.8% growth estimated by most private forecasters. Is the CSO implying that vacuuming up 86%  of cash in circulation had no impact on the economy?
The GVA for FY17, as per CSO data, does show a dent from demonetisation. At 6.7%, it has registered a sharp decline of 110 basis points from 7.8% (revised estimate) for FY16. 

Commentators cite some key indicators to ‘prove’ that economic activity shrank in the note ban months. For instance, two-wheeler sales collapsed by 22% year-on-year in December, banks reported anaemic loan growth at 5%, cement despatches fell by 9% and realtors saw a 40% dip in home sales.

While GVA growth is pretty close to private forecasts, what lifted the GDP is the strong 12.3% surge in indirect taxes that the CSO estimates for this fiscal. This is a plausible number, given that the Centre’s indirect tax collections already surged by 25% in April-December 2016, powered by higher excise duty on fuel and service tax. 
There is no obfuscation here, because assessing the GVA and adding back net taxes is the global prescription for GDP estimation by the output method.

Too mild?


But making up for these was the 6% rebound in agriculture (2.2% shrinkage last year), 6.8% increase in electricity, gas and water supply and a bumper 11.9% hike in ‘public administration, defence and other services’ which lifted the GVA.
So the CSO does admit that economic activity has been impacted by the note ban. But isn’t it estimating too mild an impact, with the Q3 GVA growth at 6.6%, against 7% last year?
But given that the economy is made up of literally hundreds of products and sectors, it is well within the realm of possibility that the economy did well even while these indicators slowed. For instance, for the same December month, steel output grew by 15%, power generation surged by 6% and refinery output expanded 6.4%. If bank credit slumped, companies doubled their borrowings from the bond market.

Agricultural output bounced back due to a good monsoon after consecutive drought years. Electricity generation was up on better coal availability. ‘Public administration’ reflects higher government payouts on salaries and pensions after the Seventh Pay Commission. 
Also, CSO estimates do show that some sectors of the economy took it on the chin in the demonetisation quarter. Manufacturing saw its GVA growth slide from 12.8% in Q3 2015 to 8.3% in Q3 2016. Finance, real estate and services saw growth collapse from 10.4% to 3.1%. Construction weakened from 3.2% to 2.7%.

Informal left out?

December quarter results from listed companies also provide independent confirmation that the big picture wasn’t much dented by the note ban. A Business Line analysis of over 1,700 listed companies showed that they just reported their best quarterly performance in three years, with sales growing over 9% and profits expanding 20%.
Commentary from listed firms suggests that urban discretionary purchases bounced back quickly as consumers switched to digital payments. Commodity industries, helped by global price rebound, did very well this quarter. In some sectors, business shifted from the unorganised to organised players due to digital payments.

As the cash situation normalises, these one-offs should get ironed out, moderating GVA growth for the March quarter. But this doesn’t imply that the Q3 GDP numbers are suspect.
Analysts also suspect that, in some cases, companies mopped up demonetised notes from their distribution channels and pumped them with inventory instead. (This would show up as ‘sales’ in the company’s books and as ‘output’ in GDP estimates).


A third criticism of the CSO estimate is that it fails to capture the performance of the informal economy, which clearly bore the brunt of the note ban.
This criticism is partly valid. 

More accurate estimates of what really transpired in the Indian economy post-demonetisation will be available when the CSO publishes its first revised GDP estimates, with more  ground-level data, in January 2018.

Over 40-45% of the Indian economy is informal and hardly any data points relating to it are available at a quarterly frequency. Therefore, what the CSO does to arrive at its quick estimates of the GDP is to take the available data from the organised sector and extrapolate it to infer informal activity. Thus, the GVA for agriculture is guesstimated based on kharif and rabi crop prospects. The GVA for services is inferred from sales tax collections, deposits and credit, telephone connections and so on. Manufacturing GVA uses the index of industrial production and listed company filings.
Owing to such guesswork, it is quite likely that the quarterly GVA estimate, which mainly uses data from the formal sector, painted a rosier picture of growth than the ground reality. But then, if the CSO — with its access to multiple data sources — has no way to estimate the quarterly performance of the informal sector, neither does anyone else.
 



As long as the CSO consistently follows the same method for measuring the informal sector and publicly discloses it, this is the only estimate we have to gauge economic activity. Both the methodology for estimating informal sector performance and GDP revisions are well-documented and disclosed on the Ministry of Statistics and Programme Implementation website. 
Until then, critics must follow Keynes’s tenet — when facts change, it is best to change your mind.

aarati.k@thehindu.co.in