The RBI Governor, Dr D. Subbarao, was right in observing a week ago that the quality of macroeconomic data left much to be desired. Estimates of GDP, industrial output and inflation, on the basis of which the RBI formulates monetary policy, are subject to constant and major changes. This adds to the risk of the RBI and the government making the wrong moves. For instance, the growth estimates for 2009-10 changed from 7.4 per cent towards the end of that fiscal to 8 per cent by January 2011. Six months from now, we may be looking at a very different figure for GDP growth in 2010-11, making the present policy steps seem inappropriate. And even that ‘final' number is likely to be wrong, thanks to the inability to capture developments across large segments of the economy.
To be fair to the government, it has revamped the WPI and the IIP, enlarging the representation of services in the former and including about 100 new industries and weeding out 42 old ones in the case of the latter. But this can only be the beginning. GDP estimates are unable to get a fix on the unorganised sector, believed to make up over half the economy. Enterprise Surveys carried out by the National Sample Survey Organisation, are read along with the NSSO's surveys on employment and the Census to arrive at the number of workers and units in the unorganised space. But it is anyone's guess whether the universal or sample surveys are reliable. Estimates of gross value added (GVA) per worker are considered suspect, more so when the output is intangible, as in services. It is, however, hoped the output estimates of organised manufacturing will improve with the introduction of the new IIP, which can be used to extrapolate data collated by the Annual Survey of Industries. Input-output tables for various sectors of industry need updating from time to time. Due attention must be paid to services sector data, believed to be the most unreliable of all. Studies suggest that the GDP deflators used for the services sector have not kept pace with inflation after the 1990s, leading to an overestimation of output.
The RBI Governor is not the first to voice concerns about data quality. The former Finance Minister, Mr P. Chidambaram, aired his doubts over IIP data soon after industrial output was reported to have grown at 1.3 per cent in August 2008. However, most policymakers bring up the subject only when the numbers don't suit them. Rather, they should focus on institutional reforms in data-gathering. Industry bodies can be roped in to plug data gaps and independent research agencies asked to supplement the government's efforts. Unless this is done, economic projections would amount to whistling in the dark.