Understanding the CPI-WPI divergence

A development that has attracted wide attention in the recent period has been the growing divergence between inflation as reflected by the wholesale price index and that by the consumer price index. This has implications for the conduct of monetary policy. The purpose of this article is to examine the factors responsible for this divergence.

Widening gap Table 1 presents the divergences in inflation according to the two indices. While CPI inflation has always been higher than WPI inflation, the divergence has been growing. While the difference in 2012, 2013 and was only 2.2, 3.6 and 3.3 percentage points, it widened to 7.8 percentage points during the period January 2015 to October 2015. While in this period, the CPI inflation remained positive, the WPI inflation turned negative. It is also seen that the divergence in food inflation between the two series is much lower than in non-food inflation. Between January and October 2015, the divergence in food inflation between the two indices is 2.2 percentage points. But the divergence in non-food inflation is 9.5 percentage points.

Food inflation Traditionally, the major reason for the CPI inflation being higher than WPI inflation has been that food articles had a higher weight (48.3 per cent) in CPI than in WPI (24.3 per cent). This factor plays an important role, whenever the primary trigger of inflation is food inflation. It is to be noted that food inflation in the last few years is being spurred more by items such as vegetables, pulses and eggs, meat and fish than cereals. For example, in the period January to October 2015, inflation in cereals in CPI was 2.0 per cent while that in pulses was 20.3 per cent, in vegetables 5.3 per cent and in egg, fish and meat 5.1 per cent (Table 2).

The same pattern is seen in WPI food inflation also. However as pointed out earlier, the divergence in food inflation between the two series is still within narrow limits. Even in 2015, up to October, food inflation was positive according to both series.

Non-food inflation Non-food inflation according to WPI and CPI have very different composition. Fuel and power category has a much bigger weight in WPI (14.9 per cent) than in CPI (6.8 per cent). What is included in this category also differs very much.

That is why in the recent period when crude prices fell, inflation in this category fell by 13.7 per cent in WPI during 2015, whereas CPI inflation was positive at 5.3 per cent (Table 2).

In fact, this divergence is inexplicable. There are certain items which figure in CPI but do not figure in WPI. These may be broadly treated as ‘services’. Table 3 lists the items. They have a total weight 28.3 per cent in CPI. In 2015 up to October, inflation in these items rose by 3.4 per cent. Medical care inflation went up by 5.2 per cent. Education went up by 6.8 per cent. Only travel and communication came down.

This analysis shows that a striking divergence between CPI and WPI inflation is seen only in the last eight months.

The divergence has been more due to non-food items, particularly services, than food inflation which normally has been responsible for higher CPI inflation. Medical care and education are two items where inflation has persistently remained high. For example, inflation in ‘education’ has remained around 7 per cent since 2012. These two items have a total weight of 10.8 per cent in CPI.

Implications for policy With more than one price index, a critical question is: what should be the focus of policy makers? In the past in India, policymakers, including the monetary authority, paid more attention to WPI for two reasons. First, there was minimal time lag in the publication of data and second the index was frequently updated with new bases and weights.

However, things have changed. CPI data are available with the same lag as WPI. CPI is also being updated with a new base frequently. Since the control of inflation is aimed at minimising the impact of rising prices on the people, the primary focus of policy makers and particularly monetary authorities should be on CPI. In almost all countries which have adopted inflation targeting, either formally or informally, this has been the case. This is what is proposed in the new monetary framework signed between government and RBI. What then is the relevance of WPI?

Of course there are some concerns with the way WPI is computed. The prices chosen in the case of some items are ex-factory prices. In some others, they are selected differently. All the same, there are some insights which a comparison of the two indices can throw. Take for example, a situation in which for the same item, inflation is higher in CPI than in WPI. This points to increasing margins and gives some indication where the policy makers should act. This apart, in the case of monetary policy when CPI inflation is above the target level but within the permissible limits, the authorities have to decide whether to tighten policy or leave it as it is.

Other considerations such as behaviour of GDP enter. In this situation, the monetary authorities must take into account not only the real growth rate but also the behaviour of WPI. If WPI is behaving in the same way as CPI, there is no additional information. But if it behaves differently, it is an important input in decision making. A negative WPI inflation with a CPI inflation staying higher than the target rate as is happening now is a rare situation. Policy makers cannot however afford to ignore the behaviour of WPI. WPI is not simply a reflection of international commodity prices. It also reflects domestic conditions.

Rangarajan is Chairman, Madras School of Economics and former chairman, Prime Minister’s Economic Advisory Council. Kannan is a professor at MSE