India's impressive GDP data has some puzzling elements

After posting year-on-year growth of 2.2 percent in October-December, private final consumption expenditure did pick up some pace in January-March, but only to 2.8 percent. According to Rahul Bajoria, managing director and head of emerging markets Asia (ex-China) economics at Barclays, the rise in private consumption "appears weaker" than what high-frequency indicators suggest.

Further, this weakness "does not reconcile with the robust value-added growth of consumption-led sectors like Trade, Hotels, Transport and Communication services", Madhavi Arora, lead economist at Emkay Global Financial Services, said in a note.

The 'Trade, Hotels, Transport, Communication, and Services related to broadcasting' segment posted gross value added (GVA) growth of 9.1 percent in January-March, second only to the 10.4 percent expansion reported by construction.

GDP-GVA divide

If GDP growth sharply surprised on the upside, GVA growth outperformed even that, coming in at 6.5 percent in January-March.

GDP is the sum of GVA and indirect taxes, net of subsidies. As such, higher GVA growth is suggestive of a contraction in net indirect taxes.

However, the latest data from the Controller General of Accounts (CGA) shows the Centre's expenditure on major subsidies was Rs 1.8 lakh crore in January-March—up only 3 percent compared to the same quarter last year. Meanwhile, indirect tax collections increased by 7.3 percent over the same period. On a net basis, indirect taxes were 6 percent higher last quarter. It is only if Integrated Goods and Services Tax collections are excluded that the indirect tax mop-up is lower by 1.4 percent in January-March on a year-on-year basis.

Rising discrepancies

A curious subhead of the GDP data is 'discrepancies', used to explain any difference between the GDP calculated through the income and expenditure methods.

In January-March, the discrepancies amounted to a negative Rs 1.28 lakh crore, implying that the GDP arrived at from the expenditure side is greater than that from the income side. Whether this is indicative of demand being overestimated is anybody's guess.

What can be said without uncertainty is that the size of discrepancies, whether negative or positive, is rising, with its absolute value now at a six-quarter high. This leaves plenty of room for future revisions in the headline GDP growth number.

Constant revisions

This conveniently brings us to where we started: How did all economists not read it right?

"The reason everybody got this wrong is the alarming regularity with which the data is getting revised," noted Kunal Kundu, India economist at Societe Generale.

"The release of October-December data saw a revision of data for the previous quarters over the past three years. The latest release once again saw a revision of recently revised quarterly data, making forecasting a rather hazardous task," Kundu added.

Constant revisions have indeed been the bane for policymakers as well as economists. After data for October-December, released in February, showed weak private consumption growth in the quarter, V Anantha Nageswaran, the chief economic adviser to the government, was at pains to point out that "data revision to the prior years has made a 6 percent growth rate come down to 2 percent".

In July 2009, D Subbarao, then the governor of the Reserve Bank of India, said that while data revisions are understandable, they are more frequent and commonplace in India.

"Most economies have to contend with an uncertain future; here, in India, we are having to contend with an uncertain past as well," Subbarao had quipped.

Forecasters may have got it wrong this time. But who knows, maybe revisions will align the data closer to what they had expected.