GDP growth: Over to the RBI

GDP growth is almost an all-consuming statistic for policymakers, academicians and market participants. However, when the new GDP growth figures were released for the quarter of April-June 2015 (hereafter 2015:2), there was more darkness than light. Not entirely unexpected, since the new GDP data have been met with scepticism(doubtful,संशयवाद). I have not been a sceptic, neither am one today; while the growth figures are confusing, there is a logical, consistent explanation. This time though, don’t blame the CSO for the controversy, blame the IMF.

What happened? GDP growth for the April-June quarter came in at 7 per cent versus market expectations of 7.4 per cent and my own estimate of close to 9 per cent. Many analysts were stunned by the low estimate, especially since the finance ministry and other officials were publicly touting(advertising,प्रचार) an 8-10 per cent GDP growth path. So, several officials (and commentators) have egg on their face(it means you have made a big mistake,बड़ी गलती). Or have they?

What did the GDP data released by the CSO show? The new data estimates for both gross value added (GVA) and for GDP, and the two are related to each other by this simple identity: GDP = GVA + DITS, where DITS is the difference between indirect taxes and subsidies.

Some simple facts about the three variables: First, DITS is normally positive and accounts for approximately 7 per cent of GDP. Second, GDP and GVA growth rates closely parallel each other. Normally. But for 2015:2, the two diverge, and diverge by a huge amount. In real terms, both grew by a near identical rate of 7 per cent in 2015:2. But in nominal terms, GDP grew at 8.8 per cent while GVA grew by only 7.1 per cent. In other words, inflation as measured by the GVA deflator was zero per cent, while inflation as measured by the GDP deflator was 1.8 per cent.

So the explanation for the difference in growth estimates is all in the computation of the deflator.

But which of the two deflators is likely to represent the correct inflation rate in 2015:2? Before this quarter, they are near identical; between 2011:2 and 2015:1, the aggregate GDP deflator increased by an average of 6.5 per cent a year, the GVA by 6.4 per cent.

One indirect check on what the deflator should be is yielded(return,मुनाफा) by the close synthetic proxy for the GDP deflator — a weighted average of WPI and CPI inflation. Historical tests suggest an approximate weighting of two-thirds for the WPI and one-third for the CPI. Between 2014:2 and 2015:2, CPI inflation was 5.1 per cent and WPI inflation minus 2.4 per cent. This yields a 0.1 per cent estimate for the implicit price deflator, strikingly close to the actual GVA deflator of zero per cent!

That deflator inflation is close to zero is also supported by the implicit price deflator for the three supply-side sectors — agriculture (4.5 per cent), industry (minus 1.1 per cent), and services (minus 0.5 per cent). Weighted by shares in GDP, this also yields a deflator close to zero. The chart documents y-o-y quarterly inflation according to four different indicators. The services deflator inflation is introduced to emphasise that inflation decline to close to zero levels is for all sectors. So the real puzzle is the very high GDP deflator inflation of 1.8 per cent — how can this number be so large when all inflation numbers suggest 0 per cent? The answer is the very high implicit price deflator for DITS. Actual nominal growth of DITS is 33.6 per cent, so real growth in DITS is only 6.3 per cent, which yields an implicit deflator of 27.3 per cent.

It’s not the CSO’s fault, blame it on the IMF. The IMF recommends/ mandates that official quarterly estimates of GDP be based on “historical” annual estimates of real DITS. At the end of the fiscal year, all data are available and “true” estimates can obtained. So notice what the CSO has done — it has assumed real DITS growth to be close to the historical norm of 6 per cent (as per IMF recommendations), and allowed the very large residual to be the price change!

The GDP deflator puzzle is solved. GVA deflator accounting for 93 per cent of GDP has a deflator of zero; and 7 per cent (DITS) has a deflator of 27.3 per cent. Aggregate GDP deflator inflation of (0.93x0 + 0.07x27.3) of 1.9 per cent. Eureka, puzzle solved, QED.

So what is the real GDP growth in 2015:2? Two methods: First, the nominal growth in GDP (CSO figures) in 2015:2 was 8.8 per cent; with a correct GDP deflator of zero per cent, this direct estimate of real GDP growth would be 8.8 per cent. Second, with DITS, one obtains (0.07x27.3) or 1.9 per cent as the extra GDP growth missed out by the CSO/ IMF estimate, that is, GDP growth was (7 + 1.9) per cent. So you can take your pick: GDP growth, correctly estimated, in 2015:2 was 8.8 or 8.9 per cent — not the reported 7 per cent.

An additional question in search of an answer: Why has DITS grown at such an enormous(big,large,विशाल) rate this year? Mostly oil, and also some reduction in subsidies. Have you noticed that while the imported oil price is less than half of what it was last year, the domestic price of petrol has stayed virtually constant? I exaggerate(overstated,बढ़ाचढ़ाकर कहना), it is down by about 10 per cent. The government (both Centre and states) has pocketed almost the entire fall in oil prices in the form of higher taxes.

Oil tax revenue adds to growth and with little extra expenditures, the extra tax revenues translate into a reduction in the fiscal deficit. Even if there are no additional gains, this 2 per cent gain in one quarter translates into an annual reduction in the fiscal deficit of around 0.5 per cent of GDP.

But does growth feel like 9 or 7 per cent? Absolutely the latter, as also indicated by the growth in value-added estimates of agriculture, industry and services. So what’s the meaning of 9 per cent growth? Assume for a moment the government suddenly discovered gold worth 2 per cent of GDP and sold it and used all of its revenues to pare down debt. Then, GDP growth would be 9 per cent but feel like 7 per cent.

This points to an important difference between the BJP and the Sonia Gandhi-led UPA. The UPA received a windfall gain (low oil prices) in 2009 and squandered(waste,नष्ट) it, possibly in the interests of winning an election. The BJP has been a lot more responsible and seems to be dedicated to doing the right thing for the economy — bringing down inflation, reducing the fiscal deficit, even if it means sacrificing some short-term growth in GDP and jobs.

When and how will India see a GDP growth close to its potential of 8-10 per cent? I mean real GDP growth of 9 per cent, accompanied by growth in jobs and economy-wide growth — and not by the decline in oil prices. There are no magic wands(stick,छड़ी), but there is a time-tested policy: Make real policy rates competitive with the rest of the world. And 25 bps will just not do it. It’s now over to the RBI.