Why it's time to challenge the math on inflation targeting

Inflation targeting framework is now enshrined as a formal agreement by the government and the RBI; thus, it may seem that we are flogging a dead horse, says Soumya Kanti Ghosh.

The global experience of inflation targeting has not been impressive, with the number of countries facing negative inflation increasing in the last few years. The question is: Are such price declines purely the result of iflation targeting or a by-product of global disinflation?

Recently, the government notified a four per cent inflation target (Consumer Price Index or CPI) with a range of plus/minus two per cent for the next five years under the monetary policy framework agreement with the Reserve Bank of India.

Undoubtedly, this is one of the most crucial and path-breaking monetary policy reform till date. This policy change is also being touted as an adoption of a modern monetary policy framework.

Since it was first implemented by New Zealand in 1990, inflation targeting has been adopted by  28 countries. India being one of the largest economies in the world is the latest addition to it.

There is a large debate on the pros and cons of the inflation targeting framework; we don’t want to dwell on the specifics again.

However, what is interesting is that in recent times, there has been more sensitivity in the market regarding any constructive discussion/criticism on inflation targeting - so much so that I could not trace a single paper on the International Monetary Fund website that was critical of inflation targeting. This paper was there even in 2014 when the Urjit Patel committee had recommended inflation targeting for India! On a lighter note, am I getting old then?

Let us first discuss the relative merits and demerits of inflation targeting in India. However, before going into the discussion of a four per cent target, let us take a cursory look at the anatomy of CPI inflation in India.

CPI is broadly classified into six categories: food and beverages (weight: 45.9 per cent), pan, tobacco and intoxicants (2.4 per cent), clothing and footwear (6.5 per cent), housing (10.1 per cent), fuel and light (6.8 per cent) and miscellaneous (28.3 per cent).

Among these categories, food and beverages is the most important and difficult to target in India. Food inflation in India is highly volatile and impacted by various internal and external factors.

To gain a clear insight into food inflation in India, we looked at data for the last 45 years (1971-72 to 2015-16) for both WPI (Wholesale Price Index)-Food and CPI-IW (Industrial Workers) food inflation, as historical CPI (combined) data is not available. Even the Patel committee back-cast the CPI series before 2010 by looking at the CPI-IW data.

The WPI-food average is 7.4 per cent for the period 1953-54 to 2015-16 (63 years), ranging from minus 11.3 per cent to 26 per cent; the CPI-IW food inflation average for the period 1971-72 to 2015-16 (45 years) is 8.3 per cent, with the minimum at minus 7.3 per cent and the maximum at 28.3 per cent.

In both the series there are nine observations/years when food inflation was negative but in most cases it was because of base effect and/or tail events. Negative inflation largely pertains to the years before 1970; the present structure of the economy has since changed significantly.

Based on this data, the distribution curve for CPI-IW food inflation for the last 45 years (1971-72 to 2015-16) indicates a normal distribution. The number of observations when food inflation is more than eight per cent is 50 per cent of the time, while on close to 80 per cent of the occasions, food inflation is more than four per cent.

WPI-food inflation, on the other hand, does not follow any distribution - it works in a cyclical manner. Of the 63 years under consideration, WPI-food inflation is in the range of four to six per cent in four years only; in most other observations it is in the range of six to nine per cent of food inflation.

Thus, if we purely consider history, it is clear that food inflation has tended to be on the higher side. Even if we assume four per cent as the best case for food inflation, core CPI should be at 2.2 per cent.

If we take the mid-point of four to eight per cent for food inflation, average core CPI is estimated at 1.2 per cent, if we were to hit the four per cent CPI target.

Considering these numbers, it is hard to believe that sufficient homework has been done for arriving at the CPI target of four per cent. Remarkably, since 2012, average food CPI has stayed at 8.4 per cent, which sums up the nuances of the four per cent arithmetic.

In the terminology of noted economist Thomas Sargent, hasn’t the government and the RBI decided to navigate an unpleasant CPI arithmetic for the next five years?

Now, a word on the global experience of inflation targeting in recent years. Since 2013, the percentage of economies facing negative inflation has been increasing, with actual inflation being below targets in 15 countries (out of 18 inflation targeting major economies) in June 2016. Seven of those countries have inflation rates below one per cent.

The obvious question is therefore: Are such price declines purely the result of inflation targeting or a by-product of global disinflation?

The inflation targeting framework is now enshrined as a formal agreement by the government and the RBI; thus, it may seem that we are flogging a dead horse.

To me, it seems the arithmetic is challenged little beyond the assumptions that have gone into it. Also, in a developing country such as India, it is obvious to be apologetic about inflation. But isn’t it a crime to not be apologetic about employment?

Soumya Kanti Ghosh is chief economic advisor, State Bank of India. Sumit Jain and Tapas Parida contributed to this article. The views expressed are personal.

Photographs: Rupak De Chowdhuri/Reuters and Sailesh Andrade/Reuters.