India s official data, whether on jobs or GDP, has come in for severe criticism in the recent past, not just because there seemed to be a political angle to it witness how NDA-2 growth trumped that in UPA-1 and UPA-2 as a result of the back-series but also because the data seemed to fly in the face of common sense. How could jobs, for instance, be growing at the pace the government claimed if private consumption expenditure growth was slowing so fast? And how could GDP growth in NDA-2 average 7.7% as compared to UPA-2 s 6.7% if bank credit grew at roughly half the pace (8.6% in NDA-2 vs 16.3% in UPA-2) and investment levels also fell to around 29% of GDP from around 33.2% in the same period? The latest blow to the integrity of India s data system comes from a study by former chief economic advisor Arvind Subramanian; the former CEA has argued that, as compared to the official estimates of around 7% per year, GDP is likely to have grown by just around 4.5% per year between 2011-12 and 2016-17.
Given how the period Subramanian has chosen spans both the UPA as well as the NDA, of course, his analysis cannot be seen as a critique of the statistical system only under the NDA though many will interpret it as such. Indeed, even the original rebasing exercise this is what led to the first controversy over GDP estimates was started during the UPA period and the methodology was finalized before the NDA came to power; the results may have benefitted the NDA, but that was coincidental. Subramanian comes to his conclusion by, first, establishing that the GDP estimates are not kosher and, having done that, he constructs several alternative indices such as power consumption, vehicle sales, petroleum consumption etc and runs regressions with them to see how well they correlate with GDP. As Subramanian asks, how GDP growth can be roughly the same in 2001-02 to 2011-12 (7.5%) and 2012-13 to 2017-18 (6.9%) if other indices are so different? Exports of goods and services few 14.5 percent before 2011 but just 3.4% thereafter while the growth for imports were 15.6% and 2.5% respectively! Production of commercial vehicles, similarly, grew at 19.1 percent before 2011 and minus 0.1 percent after 2011; a host of such examples can be given.
The first concern for the re-elected government, then, has to be getting India s statistical system back in order; if even the data is faulty, how can any government or the RBI even conduct policy? If the government knew that, as Subramanian says, growth was doing so badly, it would have been faster to address both the farm as well as NPA crisis with faster bank recapitalization; RBI too would have been quicker to cut repo rates if economic distress was so palpable. On the flip side, while the numbers suggest monetary policy was too tight, it shows fiscal policy was too loose since, with a lower GDP number, the fiscal deficit will look even higher. While part of the reason for the data being problematic has to do with the collapse in inflation, a larger reason could due to the use of MCA-21 data in place of volume-based data like the IIP or proxying informal sector growth by formal sector activity. Though recent events make it look as if the statistical system has been undermined by the NDA, as Subramanian points out, the problem is more deep-seated.