The following piece by Prof. SG Badrinath, Visiting Faculty & Chair, CCMRM, IIM-B was featured on Monday, 15th February 2015 in the Opinions section of the Mint newspaper.

In the last week, there have been a number of pieces talking about the dramatic revision in Indian GDP estimates. On January 30, 2015, the Ministry of Statistics and Programme Implementation released details in a note which economists will be poring over in the coming months. The estimate change is substantial, a 50% increase in GDP from 4.7 % to 6.9% for 2014. The change in methodology is significant, two most notable are a) the use of market prices instead of factor costs and b) a much more comprehensive coverage of the corporate sector. The former change is consistent with global practices. As Pronab Sen, Chairman of the National Statistics Commission notes, the MCA database now used carries information for nearly 5,00,000 firms as opposed to 2500 in the earlier Annual Survey of Industries. The former change will probably increase estimate volatility while the latter has to make the data more reliable.

The media reaction both domestic and overseas is predictable, the Wall Street Journal laughs pointing to reports that even India’s central bank governor Raghuram Rajan is confused. A Financial Times headline screams: “ India: GDP growth rate up, confidence in statistics down? The Financial Express quotes India’s Chief Economic Advisor Arvind Subramanian as puzzled by the high growth rate. Out-of-power politicians are quick to point out that the revised, higher GDP growth rates should mean that economic recovery started during their term. Indian press notes that the current government is hoping to reap the benefits too. The Chinese government pooh-poohs the notion that India may actually grow faster than China in the coming decades. The obsession with growth is so pervasive that the oxymoron of “negative” growth even merits a definition in Barely heard amidst all this blather are acknowledgements of the difficulty of estimating economic activity in a population of over a billion people. Measured, thoughtful responses that this release is not consistent with other data on bank credit and corporate earnings, that the information contained in it should be studied more carefully, that even this revision may eventually be revised are all relegated to the fine print. In a country where nearly half of the domestic product comes from “informal” sources, the difficulties in measurement should be self-evident.

Let us also not forget that bashing government statistics is a time-honored pastime in most domiciles. The Wall Street Journal routinely knocks economists from the Bureau of Economic Analysis in the US for leaving out oil and food from CPI calculations. If we don’t eat and don’t drive they say, we won’t have inflation, forgetting of course that the very reason for excluding these components is the volatility of prices in their underlying markets and that monetary policy pronouncements based on such numbers would be imprudent.  Follow sites like Shadow Government Statistics ( and you will see well-reasoned complaints that the US government dramatically understates unemployment figures.  Other publications hint that if the US Government were to reveal “true” inflation estimates, then the size of their entitlement payouts would increase.

The purpose of this piece is not to get into debates about estimation methodologies, political motivations, cross-country comparisons or to point fingers at a certain kind of journalism. Indeed, some journalists have valiantly attempted to go inside the numbers in this press release while others like A.S. Panneerselvan of The Hindu more generally stress the value of “data journalism. The purpose of this piece is to draw attention to a thinly veiled, somewhat pervasive, anti-intellectual attitude towards data and statistical analysis of which the furore about GDP numbers is just one more illustration. More pernicious is the feeling that by torturing data long enough, economists can get it to confess to anything! All around us are the often-annoying manifestations of data-mined marketing drives to get consumers to buy things they don’t want and may not even need! Data sometimes does not fit into the categories that our mental accounting processes allot. Data does not frequently conform to the elegance and precision that science teachers have taught us to prize. Data may not always cooperate in the quest for clarity, for certainty in an increasingly uncertain world. Rather than belittle efforts to present better data, shouldn’t we as a society adopt a more positive mindset? Shouldn’t we support nascent efforts like the National Data Sharing and Accessibility Policy, which promotes the dissemination of government data within the constraints of national security and privacy? Otherwise, hasty inferences from one-off events such as a GDP revision will reinforce the notion among already-skeptic decision makers that data and statistics are unreliable, that data providers are suspect, that efforts to collect quality data is futile. One fallout from such perceptions will be to make information gathering and dissemination less thorough and transparent. This in turn, would hamper the quality of the prescriptions that become possible through evidence-based policy making, clearly not a desirable outcome in the big data world we now inhabit.

S.G. Badrinath, Chair, Centre for Capital Markets and Risk Management, Indian Institute of Management-Bangalore.

The views expressed in this commentary are personal.