Is WPI useful in India anymore?

Prior to the introduction of the all-India Consumer Price Index, popularly known as CPI combined (rural plus urban), the Wholesale Price Index (WPI) was the most useful price index in India. It measured the weekly rhythm of price movement in the country.

Since 2009, WPI has been computed on a monthly basis, similar to other price indices. The Reserve Bank of India (RBI) primarily used WPI inflation for the formulation of monetary policy under monetary targeting framework as well as under multiple indicator approach (MIA)— although inflation measured by other indices was also monitored/ analysed. Moreover, the Central Statistics Office (CSO) has been predominantly using WPI to deflate GDP at current prices to arrive at GDP at constant prices. Where only volume data are available, the CSO uses WPI to convert volume to value to arrive at GDP at current prices.

WPI inflation does not receive as much attention as earlier for several reasons. Following the Urjit Patel Committee recommendations, the RBI Act has been amended and flexible inflation targeting (FIT) has been put in place with CPI inflation as the nominal anchor. During the period of transition from MIA to FIT, at least WPI inflation was mentioned along with CPI inflation in the monetary policy documents.

Under the FIT, as the RBI has been mandated to achieve price stability measured in terms of CPI inflation, the use of WPI inflation has been completely done away with. All projections relating to inflation are currently done in terms of CPI.

As of now, WPI is predominantly used for converting GDP/GVA at current prices to the same at constant prices. Several items of services sector GDP that are not included in the WPI basket are deflated by WPI to compute real GDP of this sector. In fact, the GDP deflator, which is defined as a ratio of GDP at current prices to GDP at constant prices multiplied by 100, closely tracks WPI inflation.

The GDP deflator, which is often argued as the true indicator of inflation, has been at the focal point of discussion while analysing the real GDP growth in India during the recent period. The sharp decline in the GDP deflator and the dramatic decline in WPI inflation coincided. This contributed significantly to real GDP growth in India notwithstanding modest rise in production during recent years. Compilers of GDP argued that fall in the input prices has led to large value addition and hence high GDP growth in real terms. However, ground realities have been somewhat less optimistic, particularly in the manufacturing sector.

While the base year of the current GDP series is 2011-12, the base year for the WPI series has till recently been 2004-05. Since April 2017, this anomaly has been overcome following revision in the base year of WPI from 2004-05 to 2011-12 to align it with the base year of other macroeconomic indicators. The new base 2011-12 is already five years old. The next revision of base year is being discussed. In this context, a general suggestion would be to switch over to the chain-based method.

One of the striking features of the new WPI series is that the item level averaging is being done by using geometric mean, instead of the arithmetic mean used earlier. This is as per international best practice and similar to the practice adopted for the CPI. Data available for the overlapping period with old base and new base indicate that the WPI inflation as per new base is consistently lower than the same as per the old base. Several reasons are responsible for this outcome. The geometric mean itself has significantly moderated WPI inflation, besides other factors such as change in the composition of basket, weighting diagram, and so on. Moderation of WPI as per revised base has pushed up real GDP/GVA considerably during recent years.

To remove the influence of fiscal policy, indirect taxes have been excluded from the quotations used to compute new WPI. This will make the new WPI conceptually closer to the producers’ price index. Exclusion of excise duty from the computation of WPI has also partly contributed to lower WPI inflation during recent years, which in turn has pushed real GDP up to some extent.

A high-level Technical Review Committee has been set up to dynamically review periodically the articles to be included/ excluded from the revised basket. Moreover, the new WPI series is computing food inflation – a combination of food products and food articles. Improvement in the compilation of WPI is unlikely to address the problems with the GDP deflator and its attendant effects on GDP computation.

A better way to estimate GDP accurately is to deflate input and output prices through separate indices, popularly known as double deflation. Unfortunately, several developing countries such as China and India use single deflation method to compute real GDP/GVA as separate indices for input and output are not available. When output prices move relatively faster than the input prices, the single deflation method overestimates GDP/GVA and vice versa.

In order to ensure accuracy, it is high time to discard the single deflation method to estimate GDP/GVA by using WPI as a deflator. What is therefore needed is to compute both input and output price indices for each sector and adopt double deflation method at the earliest as suggested by the System of National Accounts (SNA) 2008 – the standard compilation manual of the IMF for the estimation of GDP. Separate services sector input/output price indices are required to deflate services sector GDP for which WPI is anyway not appropriate. The attention of authorities concerned is called for, to resolve issues relating to the GDP deflator in India, as WPI has lost some of its usefulness.

The writer is former Principal Adviser and Head of the Monetary Policy Department, RBI