Opinion| The need for restoring faith in the Indian statistical system

The Indian statistical system is perhaps facing its sharpest criticism in recent memory. The data released (or not released) by the government are being openly questioned. Two members of the National Statistical Commission—including the acting chairman—resigned last week, protesting the delay in the publication of employment data, among other things. But the jobs report was later leaked. A Business Standard report showed that, according to the National Sample Survey Office’s survey, the unemployment rate in 2017-18 was at 6.1 percent—the highest in 45 years. The unemployment rate in 201-12 was at 2.2 percent. Further, the defence put out by NITI Aayog vice-chairman Rajiv Kumar did more harm than good.

However, what is damaging for the Indian statistical system is that the controversy around the jobs report is not an isolated incident. For instance, the involvement of NITI Aayog in the release of new gross domestic product (GDP) back series made it unnecessarily political. Additionally, it showed that the economy did not grow as fast as previously believed during the United Progressive Alliance tenure. The government also refuted the back series calculated by the committee on real sector statistics headed by economist Sudipto Mundle.

In fact, the new series is being regularly questioned. In a recent article in the Economic & Political Weekly, a member of the monetary policy committee of the Reserve Bank of India (RBI), Ravindra H. Dholakia, along with R. Nagaraj and Manish Pandya, for instance, noted: “...the very basis of the change in the approach to data collection for estimating manufacturing GDP seems questionable. Hence the higher share and faster growth rate of manufacturing sector reported in the new GDP series seems to have little justification based on mere coverage of ASI [Annual Survey of Industries].”

The incoming GDP data and regular revisions have not helped economists and analysts a great deal. According to the new data, the Indian economy grew at 8.2 percent in 2016-17—the year of demonetisation. In fact, the Indian economy grew at a faster pace in the year of demonetisation, compared with the previous year, and again slowed in the following year. It will not be easy for economists to explain this and revise their forecasting models.

At a broader level, though recent controversies regarding jobs and GDP data made headlines, the statistical system in the country needs to be reassessed. Dholakia, in another article in the Economic & Political Weekly, for example, argued that the current practice of measurement maybe overstating inflation.

Why is it important to improve the statistical system?

As the C. Rangarajan commission, which recommended the formation of national commission on statistics, noted: “A good statistical system is a prerequisite for sound decision making and for the formulation and monitoring of public policies.”

For example, if the current system is overestimating inflation, it is likely that real rates are at a much higher level than actually warranted. This would affect growth and employment. Similarly, overestimation of GDP will also have policy implications. The government’s budget deficit would look smaller as a percentage of GDP but will increase the risk of crowding out private investment by draining higher financial savings. This will affect growth in the long run.

Aside from policy implications, it is important that data doesn’t end up damaging trust in financial markets. Capital is extremely mobile in today’s world and might prefer moving out of a place where the very basis of making investment is being questioned.

To be sure, the economy is changing rapidly and governments all over the world regularly update the way information is gathered and presented. However, it is important that apprehensions in the market, if any, are adequately addressed.

So, what should be done?

First, it is important to restore the full strength of the National Statistical Commission. Second, the government should address all the doubts that analysts and investors have about national statistics. Finally, it is important to make data public or explain the reason for delay.

However, all of these might have to wait at least until May 2019.