India is not the fastest growing big economy

A closer look at recent data on GDP shows that the numbers are flawed and recovery is incomplete

The Provisional Estimates of Annual National Income in 2021-22 just released show that GDP grew 8.7% in real terms and 19.5% in nominal terms (including inflation). It makes India the fastest growing major economy in the world. Further, the real economy is 1.51% larger than it was in 2019-20, just before the novel coronavirus pandemic hit the world. In nominal terms it is higher by 17.9%. These numbers imply that the rate of inflation was 10.8% in 2021-22 and 16.4% between the two years, 2019-20 and 2021-22.

This picture implies almost no growth and high inflation since the pre-pandemic year. So, the tag of the fastest growing economy means little. If an economy drops sharply and then rises equally fast to reach its earlier level, that cannot be taken as an indication of a rapidly growing economy.

The quarter to quarter growth currently may give some indication of the present rate of growth. In 2020-21, the quarterly rate of growth increased through the year. In 2021-22, the rate of growth has been slowing down. Of course in 2020-21, the COVID-19 lockdown had a severe impact in Q1 (-23.8%); after that the rate of growth picked up. In 2021-22, the rate of growth in Q1 had to sharply rise (20.3%). Ignoring the outliers in Q1, growth rates in 2021-22 have sequentially petered out in subsequent quarters: 8.4%, 5.4% and 4.1%. True, the last quarter (January-March 2022) data was impacted by the Omicron-related lockdowns in January and February. It was further impacted in March by the supply disruptions following the war in Ukraine and the severe COVID-19 lockdown in China. Going forward, while the lockdown in China is over, the war-related impact is likely to persist since there is no end in sight. Thus, price rise and impact on production are likely to persist. The rapid rise in prices will impact demand from the vast majority of citizens who are losing out. That will reduce growth further.

More worryingly, the issue is about correctness of data. The annual estimates given now are provisional since complete data are not available for 2021-22. They may be better than the second advance estimates released three months ago as more data become available. There is a greater problem with quarterly estimates since very limited data are available for estimating it. So, the data for the fourth quarter of 2021-22 released now is even more problematic

The first issue is that during 2020-21, due to the pandemic, full data could not be collected for Q1. Further, for agriculture, quarterly data assumes that the targets are achieved. But in Q1, a lot of fruits, vegetables, flowers, milk and poultry products could not come to the market, and rotted and wasted. This is more than 50% of the agriculture output. Thus, the growth rate of agriculture was certainly less than the official figure of 3%.

Agriculture is a part of the unorganised sector. Very little data are available for it but for agriculture — neither for the quarter nor for the year. It is simply assumed that the limited data available for the organised sector can be used to act as a proxy. In other words the non-agriculture unorganised sector is represented by the organised sector. The data for the full organised sector are also not available so ‘high frequency’ data (listed in the press note) are used. For instance, Goods and Services Tax (GST) collection data are used. But, it is well known that GST is collected almost entirely from the organised sector. In brief: very little data are available for quarterly estimates; and even less is available for the unorganised sector. Since the same method is used to estimate the annual growth rate the errors get repeated.

If better data became available after the shock of the lockdown, and it got used, there should be substantial revision in the previous year’s quarterly data. But if one compares the Q1 2020-21 data in the latest release with the data released in May 2021, the change is 0.3%. Does this imply that the high frequency data used is very well able to predict quarterly GDP? This is unlikely to be the case when a shock is administered to the economy which changes the parameters of the economy. The data remaining largely unchanged implies that the same error is being carried forward.