RBI's inflation indicator

RBI's inflation indicator
The basket of goods whose prices are measured by CPI inflation is constant. Hence, the CPI reflects the prices of goods and services consumed by domestic households. It is the best indicator

The chief economic advisor, Dr Arvind Subramanian, has recently suggested that it may worthwhile revisiting the issue of which inflation indicator the RBI should use in conducting monetary policy. His basic argument rests on comparing the somewhat contrasting recent trajectories of inflation as measured by the consumer price index (CPI), the wholesale price index (WPI) and the GDP deflator. While the CPI is currently still around five per cent, the WPI inflation and the GDP deflator numbers are close to zero. Consequently, if one measures the stance of monetary policy by the repo rate minus the inflation rate, CPI inflation suggests that monetary policy is very accommodative while the other two suggest that it is very tight. Which of these measures is most informative for policy?

The choice of indicator rests crucially on the requirement that the chosen indicator reflect current inflationary pressures in the economy as accurately and as quickly as possible, since monetary policy needs to respond in a timely fashion to unfolding developments. One of Subramanian's proposed candidates is the WPI. This measure is wholly unrepresentative of the economy since it doesn't include the service sector which comprises over 60 per cent of output and around 40 per cent of employment in India. Any measure that fails to include the service sector will fail to provide a reliable indicator of the true state of the economy.

Another of his candidates is GDP deflator inflation. This measure has at least two problems. First, the GDP deflator includes prices of the country's exports, which can often have very little to do with the prices being paid by consumers domestically as also the types of goods being consumed locally. Consequently, it is an incomplete indicator of domestic inflation, which makes it of relatively limited usefulness.

Second, the GDP deflator is subject to large uncertainty due to its dependence on GDP estimates from the national income accounts. As is well known, GDP numbers are subject to large and significant revisions over time. The first estimates of quarterly GDP that are put out by the CSO are necessarily based on quick estimates which subsequently get revised as more data for the previous quarter come in. These revisions often continue for over a year. India is no exception in this. In the USA, the revision of quarterly GDP growth numbers have averaged nearly 1.5 percentage points since 1965, a period when average US GDP growth has been 2.7 per cent. Clearly, inflation based on the GDP deflator is a rather poor reflector of the true contemporary state of the economy.

In view of these limitations of other measures of inflation, central banks the world over mostly tend to target either headline or core CPI inflation (core inflation excludes some items like food and fuel). The CPI is not prone to large revisions. Moreover, the basket of goods whose prices it measures is constant. Hence, the CPI reflects the prices of goods and services consumed by domestic households. It is presumably for these reasons that the RBI has chosen to target headline CPI inflation. It is the best option amongst the available alternatives.

One concern with the CPI is that it may not accurately reflect prices faced by producers since it measures prices paid by consumers. The difference between producer prices and consumer prices is typically the distribution cost of getting the good from producers to households. It is conceivable that CPI inflation might be higher than inflation in the producer price index (PPI) due to increases in distribution costs. Unfortunately, the PPI does not yet exist in India though the recently constituted Goldar committee is currently working on it. Once it becomes available, the RBI could consider enhancing the information set on which it takes decisions to include the PPI. But this has to be clearly announced to the public.

The last point about clear announcement brings me to a much more fundamental problem with Subramanian's suggestion of revisiting the CPI as the inflation indicator. A key component of inflation is the private sector's expectation regarding inflation. Thus, firms set prices and labour negotiates wages based on their expectations of future inflation. Clarity regarding the inflation indicator that the RBI is using as well as the target level of inflation are both key for private agents to anchor their inflation expectations. This is precisely what the recent agreement between the RBI and the government on a four per cent long-run target for CPI inflation was meant to achieve. For the CEA to the government to now suggest that this may be worth revisiting is unfortunate, since it risks undercutting the precise goal of the agreement that the government itself signed.

The sub-text underlying Subramanian's suggestion is of course the government's desire that the RBI lower interest rates more aggressively. Subramanian justifies this by implying that high interest rates are stretching firm balance sheets and thereby causing further problems for banks that have lent to these corporates. This is disingenuous at best, since it conveniently omits to mention that the deterioration in firm and bank balance sheets occurred over the past few years when real interest rates were hovering close to zero by all measures of inflation. The causes of this balance sheet fragility are clearly elsewhere.

It is perhaps best for the government to focus on dealing with the structural issues facing the economy instead of trying to undercut some of the better articulated operating procedures like CPI inflation targeting for some purely short-run expediency.These statements are more likely to just confuse markets