Pitfalls of pegging repo rate to CPI

The CPI was never an important statistic for policy purposes until the Urjit Patel Committee recommended that it be used when targeting inflation. It was otherwise understood that the WPI would be appropriate as any control of money supply by restricting credit practices through interest rate action lowered growth in credit and hence demand-pull inflationary forces.

The CPI was more a secondary piece of information used; there were three indices in place which captured the price effect on industrial workers, agri labourers and unskilled manual workers. These indices varied because the composition of goods in each basket was different. However, as there was a divergence in the WPI and CPI inflation numbers, the central bank often fell back on the CPI to justify its action; finally the wisdom which prevailed was that the CPI would be the right indicator to target.

The CPI has three parts — for rural and urban areas and a consolidated one — so as to capture differing consumption patterns. While there can never be a perfect index, given that prices vary sharply across regions in the country for both food and non-food products including services, it has become fairly controversial at the time of interpretation as relating to these numbers has become difficult leading to several anomalies. As it is the most critical part of the MPC decision-making process, understanding the nuances becomes important.

First, CPI targeting has its own set of issues in the context of conducting monetary policy. Almost 90-95 per cent of the index is actually not affected by interest rates as the amounts spent by households are impervious to rate changes. This includes food products (which have a weight of 48 per cent in the index), housing or fuel expenses which are virtually fixed costs from the point of view of households. Therefore, monetary policy while targeting CPI can be said to be more a ‘reaction’ to inflation rather than a cure. Indeed, monetary policy action seems to be preserving a real interest rate regime, which is broadly defined as the difference between the interest rate and the inflation rate. Hence, a repo rate of 6 per cent and CPI inflation rate of 2 per cent yields a real rate of 4 per cent which will come down to 2 per cent in case inflation moves to 4 per cent, which appears to be the trajectory today.

Second, the concept of CPI inflation may still not make sense to the household. A number of 2 per cent does not sound convincing to a housewife who believes that prices have gone up by a much higher rate than what is officially conveyed by the index. Hence, there is disenchantment whenever the RBI lowers rates on grounds of inflation coming down, as households do not get the same feeling. This is because the way prices are perceived differently by an individual. An average person sees inflation as the increase in prices over the previous month rather than a year, which is the way the CPI inflation rate is reckoned. A good example to take here is tomatoes.

Tomatoes have been in the news for the large-scale variations in prices in the last month or so. The table presents information on the change in prices in two centres, Mumbai and Delhi, in the last three months on a month-on-month basis, and juxtaposes the same with what the CPI shows.

These numbers reveal that there is a divergence between price rise on a month-on-month basis and that presented by the CPI index. Households are averse to accepting that retail inflation has fallen, as their focus tends to be on specific commodities. Even when it comes to, say, pulses, the fact that prices have come down is not soothing as they had increased to an all-time high before declining and the comparison would always be with the earlier normal. Hence when prices of tur increased from ₹40 a kg two years back to ₹200/kg, and then declined to ₹70, the comparison is over the last low rather than high.

Now on a year-on-year basis, the CPI number presented for July for tomatoes would actually show an increase of 50 per cent (an index of 214.9 in July 2016 and 322.2 in July 2017). The same at the market level was 140 per cent in Mumbai and 82 per cent in Delhi. While the CPI calculation would be taking into account several markets, the numbers still do not add up to the whole.

Third, of late, the issue of housing income has come to the forefront. It has been argued that the HRA allowance paid to central government employees would tend to raise inflation with the new revisions brought in by the Pay Commission.

However, the rent which appears on the pay slip of an employee gets deducted automatically if the employee is residing in government accommodation. Alternatively if the employee does not stay in accommodation provided by the Government, then the pay slip amount goes up. Hence, an increase in HRA may not fully translate into higher retail demand or cost of living, even if the CPI goes up. This is a factor that should be taken into consideration in any decision on the repo rate.

Therefore, the CPI number will continue to perplex all segments as it is interpreted differently by the householders, policymakers and government employees. The market will always look at the number and make its demands accordingly, while the MPC too perforce has to follow the same path. But the ability of monetary policy to limit this number is weak.

At the same time, it may have to take a different stance when components such as housing move up numerically. At present, the inflation number is stable and low, which makes it easier to lower rates. But what if the price of tur increases sharply again like it did a couple of years’ back and the CPI number inches up towards 10 per cent? Unlike tomatoes or onions where there are multiple crops coming in, the impact of any shortfall in pulses reverberates for a year until the next harvest comes in. Will it lead to 25-bps increase in repo rate every time the MPC meets?

The writer is the chief economist at CARE Ratings. The views are personal