How fast is India growing?

With corporate earnings not meeting expectations and automobile sales plunging in July 2019, the story could get uglier

The title above repeats the one for my article in this newspaper of April 9, 2015, nearly four-and-a-half years ago, written a couple of months after the Central Statistics Office (CSO) brought out estimates of national income and output based on the new 2011-12 base. Some of the doubts and issues I raised then, along with other analysts and even the Economic Survey for 2014-15, were: 

• The sharp upward revision of gross domestic product (GDP) growth in 2013-14 to 6.9 per cent, implying a 2-percentage point acceleration in the GDP growth rate amid prevalent “policy paralysis”, a mini balance of payments (BoP) crisis and a 300-basis point hike in the policy interest rate;

• The high average growth rate of 7 per cent for 2013-14 and 2014-15, which correlated poorly with weak performance of high frequency indicators (HFIs) such as the Index of Industrial Production, corporate earnings, bank credit, tax revenues, and investment and exports;

• The major revisions of sectoral shares of manufacturing (upwards) and wholesale/retail trade (downwards) for 2011-12 and 2012-13, along with remarkably high growth of these two sectors in 2013-14 and 2014-15, despite lacklustre trends in HFI;

• Problems emerging with the new MCA-21 database on company accounts;

• The absence of a back-cast series on the 2011-12 base.

With the passage of time, new data releases (and associated anomalies like the peak GDP growth estimate of 8.2 per cent for the year of demonetisation, 2016-17), the publication of two alternative back-cast series last year (one under the auspices of the National Statistical Commission and the other by the CSO) and numerous commentaries on the new series, the controversy has rumbled on. It rose to a crescendo last month in the two-day annual India Policy Forum conference organised by the National Council of Applied Economic Research (NCAER) and the (American) National Bureau of Economic Research. Two full sessions and much of an evening lecture were devoted to the subject. The contributions by stalwarts such as Arvind Subramanian, Sudipto Mundle, N R Bhanumurthy, R Nagaraj, Pranab Bardhan and Sebastian Morris can be readily perused on the NCAER website. Much of the critical questioning of the 2011-12 series and its sources and methods is valid. Particularly compelling is the paper by former chief economic adviser Subramanian (2014-2018). Let me give its flavour.

Subramanian’s paper is a follow-up of his June 2019 Harvard working paper and includes pretty convincing rebuttals to criticisms of that paper levelled by some analysts, including official sources. He divides India’s growth performance (according to the official 2011-12 base series) into two periods: pre-2011, meaning 2002/3-2010/11, and post-2011, meaning 2012/13-2016/17. The cut-off is 2011-12 because that is when the new base started. He, then, focuses on comparing the trend in the growth of output (GDP) to trends in the growth of key aggregate demand variables, notably investment, exports and credit; less persuasively, he also throws in imports. 

The comparison between growth trends in these variables over the pre-2011 and post-2011 variables is shown in the figure and is striking. Pre-2011 is the boom period when GDP was growing at nearly 8 per cent, driven by similar buoyancy in real (inflation-adjusted) investment (13 per cent), real exports (15 per cent) and real credit (17 per cent). Everything is hunky-dory. Then come a series of negative shocks: A stalling of global trade after 2011, the UPA government’s policy paralysis in 2012-2014, the associated mini-BoP crisis of 2013, the growing twin balance sheet crisis, the droughts of 2014 and 2015, and demonetisation in 2016. These far outweigh the only positive windfall, the oil price decline after 2014 (since partially reversed). As expected, the growth of macro demand indicators collapsed post-2011. Investment, exports and credit plummeted to 3 per cent growth from the high double-digit levels of pre-2011. One would have expected GDP growth post-2011 to have also fallen sharply. It didn’t!  Subramanian’s “central puzzle” is that GDP growth only dipped 0.8 percentage point, from 7.7 per cent to 6.9 per cent. 

Subramanian rightly infers that there is something wrong with the GDP growth estimates post-2011, as measured by the new 2011-12 base. He goes on to show that in no country has periods of 7 per cent  economic growth been accompanied by such low growth of investment, exports and credit as we have experienced in the post-2011 period. Conversely, he marshals the international experience to show that when these macro demand indicators are as sluggish as they have been in India post-2011, no country has grown faster than 5.5 per cent, only four have grown at 4.5-5.5 per cent, 13 at 3.5-4.5 and 33 at below 3.5 per cent. There is much more in his paper and in the others presented in the conference.  All raise serious doubts about GDP growth estimates post-2011 and call for a major reassessment of national income estimates by the NSO.  Among the possible reasons suggested for the systematic and significant upward bias in the current official estimates of GDP growth post-2011 are problems with the MCA-21 database on companies and the choice of price deflators used by the current GDP estimates.

So how fast is the Indian economy growing? Clearly, significantly below the rates shown by the official series. How much below? The truth is one can only make educated guesses based on the kind of detailed analysis carried out by Subramanian in his two recent papers.  His papers indicate a “discount” of at least 2.5 percentage points. That is, if official estimates indicate 7 per cent growth, then the “real” growth is likely to be 4.5 per cent or less. My own more conservative guess at the discount would be a trifle less, in the range of 1-2 percentage points. So, if the most recent official estimates (for Q4 of 2018-19) show that India is growing at 5.8 per cent — let’s say rounded up to 6 per cent — then the “real” growth rate is more likely to be in the range of 4-5 per cent. Not a pretty story. But with what we know of 2019/2020Q1 developments in HFI and with automobile sales plunging 30 per cent in July 2019, the story could get uglier.

The author is Honorary Professor at ICRIER and former chief economic adviser to the Government of India. 

Views are personal