That has, quietly, changed. While few believe Indian statisticians are actively working to make growth numbers look better than they are, less and less data is publicly available, methods are less transparent, and the GDP figures in particular sometimes diverge puzzlingly from independent data.
The minister in charge of statistics recently told Parliament that the government planned to ask a new committee to recommend how it should update its national accounts. Official statisticians should seize this opportunity to overhaul how India’s GDP is calculated in order to win back trust.
The government’s justification for the update is that India’s data is still based on prices dating back to the financial year that ended in March 2012. Such “rebasing” is a chance for wholesale reform — particularly because the last time the GDP series was revised was exactly when questions first began to be asked about its reliability.
Statisticians have a much harder task in India than in the West or even in China. For one, the economy is dominated by services, rather than by manufacturing. It can be easier to value aggregate output in industrial sectors that produce a defined output with a clear price.
A bigger problem is that, unlike in more advanced economies, a large part of activity in India happens in the informal sector. That is, by definition, invisible to the government; official tax records, for example, are not going to tell you much about small businesses that traditionally haven’t paid taxes.The micro-sized enterprises where most Indians work are also very hard to survey. They go out of business quickly; they change their names and locations frequently.Official numbers get around this problem by looking at a sample of larger companies and extrapolating from that. If the performance of smaller enterprises is linked closely with how bigger ones are doing, such an estimate can work well.
But, if the formal and informal sectors are growing at very different rates, the calculation might lead to biased estimates of GDP. Some think that’s exactly what has happened in recent years.
The figures for value-added contributed by the private sector have been particularly worrisome. They are often out of sync with other macroeconomic indicators — corporate earnings, credit growth, the central bank’s estimates for capacity utilization in industry. Many India-watchers now supplement the official GDP numbers with other data points, such as sales of two-wheelers or the revenue reports of consumer-goods companies.
About a decade ago, when the “base year” for GDP calculations was last revised, statisticians also changed the data source they used to estimate private-sector output. That may have introduced unintended problems.
What really angered many economists, however, is that it seemed the “new series” of GDP behaved completely differently from the one they had been using for decades. Comparisons between macroeconomic performance in the 2010s and the 2000s, already politically controversial, became impossible.
The new series also seemed to be overly sensitive to how statisticians calculated price changes between one year and the next. Normally, such questions might be answered by official papers detailing the statisticians’ sources and methods. But those have stopped appearing as well.
If authorities are serious about fixing India’s problem statistics, they can do two things. First, better mine the enormous amount of available digital data, including the 500 million transactions a day recorded on state-supported payments infrastructure. If they don’t do so, some future AI bot will.
Second, they need to be much more open about what they are doing and why. Statistics are useless if they aren’t trusted. And only transparency builds trust.
India has wound up with official statistics that are unpredictable, don’t match up with other figures, aren’t robust to how price shifts are calculated, and provide limited comparability over time. That’s sub-optimal, to say the least. To promote growth, the government first needs to know how the economy is really doing.
(Disclaimer: The opinions expressed in this column are that of the writer. The facts and opinions expressed here do not reflect the views of www.economictimes.com.)