On India's latest GDP numbers

  • Blog Post Date 25 March, 2017
  • Perspectives
  • Print Page
Author Image

Abhijit Banerjee

Massachusetts Institute of Technology

India’s latest GDP data shows that the economy grew at 7% in the previous quarter, despite demonetisation. In this article, Abhijit Banerjee discusses why the new numbers are likely to be an overestimation.

There is much discussion in the press about the surprising ‘fact’ that the Indian economy grew at 7% this last quarter, despite the demonetisation. Economic Affairs Secretary Shaktikanta Das, never one for understatements, declared: “The numbers completely negate the kind of negative projections and speculations which were made about the impact of demonetisation.”

And yet the answer to why the GDP (gross domestic product) numbers came out the way they did was already anticipated and nicely explained in the excellent Economic Survey that came out last month from the same ministry that Mr. Das works for. The report actually highlights a number of model-based computations that are consistent with the now confirmed official view that demonetisation will have a small effect on measured GDP growth, but then quite bluntly declares: “Recorded GDP growth in the second half of FY2017 will understate the overall impact because the most affected parts of the economy — informal and cash-based — are either not captured in the national income accounts or to the extent they are, their measurement is based on formal sector indicators. For example, informal manufacturing is proxied by the Index of Industrial Production, which includes mostly large establishments. So, on the production or supply side, the effect on economic activity will be underestimated. The impact on the informal sector will, however, be captured insofar as lower incomes affect demand for formal sector output, for example, two-wheelers.” The survey goes on to show evidence that the demand for two-wheelers crashed.

This is spot on. The problem is that there is no good way to collect data from the hundreds of millions of establishments that constitute the production sector of our economy — the farms, the one-man plumbing or shoe-polishing businesses, the many, many, many kirana1 stores. It would just take too long and cost too much, especially since most of these establishments keep no records. So what the Central Statistical Organisation does instead is to start by computing total the Gross Value Added based on output figures from the larger, registered establishments (the formal sector) where all the relevant data, these days, are in a computer file, and are therefore easy to access. To get the equivalent number for the agricultural sector it uses a number of rules of thumb, such as projections of agricultural output based on the weather and some informed guesses about the planting of various crops. For manufacturing and services, the basic rule is to assume that the output of each informal sector moves in some fixed proportion to formal sector output. That proportionality factor is updated from time to time, based on sample surveys and market estimates, but that process is obviously too slow and cumbersome (you need to figure who to survey, where they are located, how to get them to give reliable answers to questions, etc.) to be done every quarter. So when the quarterly GDP comes out it is necessarily based on proportionality factors that are dated, and in particular don’t take into account the fact that certain policies, like demonetisation, have a disproportionate effect on the informal sector because much more of it is cash-based. This means our GDP numbers are likely to be overestimated right now, and given just how large our informal sector is, the overestimation could be very significant.

None of this is entirely new, but it seems to be totally missed by our leaders who are happily beating their chests about our (mostly made up) growth numbers. It is not that we know that growth is down a lot, we just don’t know.

In fact the most valuable contribution of the current Economic Survey is less the specific points it makes (which are also interesting and important), but its attitude. In this time of alternative facts, I am deeply grateful for its commitment to the old-fashioned view that facts are important and need to be tracked down. For example, to make a prima facie case for Universal Basic Income (UBI), the Economic Survey reports on a fascinating fact-finding exercise. It starts by highlighting that the budget features around 950 different central and centrally-sponsored schemes for the disadvantaged, many of whom no one has heard of, with obvious consequences for bureaucratic efficiency (some have offices and staff but no real budget) and coordination (some people manage to drink from many fountains while others go thirsty because they don’t know where to look). To add to that the programmes that actually have money are very badly targeted. None of the major programmes targeted at the poor spend even 40% of their resources in the poorest districts where 40% of the poor live (the median fraction for six programmes the Survey emphasises is 30%). There seems to be, if anything, some amount of targeting of anti-poverty programmes to the less-poor.

I also loved the use of satellite data to compute the potential property tax base for some of the bigger cities, to demonstrate that many cities are failing miserably in collecting taxes. Once again it delivers to us an important fact that we might have suspected but didn’t, in any useful sense of the word, know.

I was also struck by the candour with which it talked about the biggest challenge faced by the Indian economy, which is the ballooning NPAs (non-performing assets) of our banking sector, and what it means for the future of investment and growth — nothing good — and also by its candid admission that the government does not yet have a plan for dealing with it.

My hope is that the standard it sets for the way we should carry out our public debates will stick, and shame those in positions of power and influence who have embraced the culture of insinuations, braggadocio, and outright falsehoods.

This article first appeared in Hindustan Times:


  1. Kirana is Hindi for a a small, usually family-owned shop selling groceries and other sundries.
No comments yet
Join the conversation
Captcha Captcha Reload

Comments will be held for moderation. Your contact information will not be made public.

Related content

Sign up to our newsletter