There is a consensus on the need to change the base years for our economic variables. The present base year for IIP, WPI and GDP is 2011-12 while it is 2012 for the CPI.

There are two issues with the base year being outdated. The first is that there is a concern over indexing all numbers to a base which is over 13 years old.

The second, which is probably more important, is that the composition of production and consumption of goods has seen sharp changes every three or four years. To capture these changes, revisions are called for, about every five years.

Time period

The choice of the time period is crucial while changing the base year. While it has to be as contemporary as possible, it needs to satisfy several criteria. This becomes a challenge because unlike the phase of Great Moderation which lasted till the Lehman crisis, the world economy has been volatile in every sense even though the emerging markets have been more insulated from these shocks. Covid was a universal phenomenon which affected all countries.

So choosing a year which was too good or too bad has ‘hangover effects in reverse’ where subsequent years would look disproportionately high or low. The normal year also has to be one in which where the variables did not witness any shocks.

Today with global crude oil prices having the potential to skew inflation, the base year must be free of such disturbances.

If these principles were to be adhered to, which year would be most suitable base year? While it is tempting to pick up a contemporary year like 2021-22 or 2022-23, there is a huge decline bias still in the economy.

A quick way to understand this bias is to look at the real GDP numbers. During 2014-19 there was an increase of around ₹42-lakh crore in real terms. In contrast, post 2018-19 when Covid struck and a lockdown was imposed in March 2020, the real increase has been of the order of just ₹33-lakh crore (FY19-24).

Hence while the GDP growth rates of 9.7 per cent, 7 per cent and 8.2 per cent (as per the provisional estimates for FY24 released on May 31) in the last three years are impressive, they come over low number of 3.9 per cent and minus 5.8 per cent in FY20 and FY21. We are clearly not yet back to normal. This being the case, FY19 looks a better year (6.1 per cent growth).

Inflation factor

Also inflation has been unusually high for the last four years which will tend to depress the numbers going ahead. The inflation numbers were 4.8 per cent, 6.2 per cent, 5.5 per cent, 6.7 per cent and 5.4 per cent running up to FY24.

In FY19 it was 3.4 per cent, within the MPC target. Therefore 2018-19 fits the bill here. In fact ever since data was released on household consumption expenditure, there has been discussion on whether households are spending less on food and more on elements of core inflation.

Two related issues come up when revising these base years. The first is whether the stock indices too should be changed. While the Sensex and Nifty are privately constructed indices, their base years are 1979 and 1995 respectively. This realignment will make it more complete.

The second pertains to the composition of indices. Currently the WPI is a producer’s price index and does not include services. With services having a larger share in GDP, the WPI tends to only partly capture price changes.

While the government tracks a services price index, ideally it needs to be included in the WPI. The CPI already has several elements of services like education, health and recreation. But sectors such as logistics need to be included in the WPI.

A related issue to the collection of data pertains to both GDP and IIP. There are large elements of the unorganised sector that need to be included. This pertains to the MSME segment which dominates both services and manufacturing. There definitely needs to be better representation of this sector with less recourse sought to imputations.

Here the GST data provides useful clues and can be used to impute more accurate values on the value addition in this sector. It must be remembered that when the base year of 2011-12 was introduced, there was no GST.

In case of those units outside GST, there should be surveys carried out by the MSME Ministry to get an idea of the value addition. Covid has definitely skewed the process of selection of a new base year for these indices. Choosing a base year post FY21 can lead to bias that can impact policymaking.

This holds especially for monetary policy where the inflation numbers could get a downward bias due to the high inflation trajectory in the last three years.

In fact, the household consumption survey data, if used, can also skew the weights as food has a lower share based on the data for FY23. The pent up demand witnessed post-Covid has led to households spending a lot on travel and tourism as well as consumer goods which has lent this bias to non-food items. Therefore, care must be taken when selecting the normal year.

Under these unusual circumstances there could be a case for deferring the exercise until a new normal year closer to the present day emerges. But this could be fraught with uncertainty.

Choosing a pre-Covid year of FY19 may be a bit dated as there has been a drastic change in how business is conducted post-Covid. A compromise solution is to update the base year to FY19 but look out for a new one which can be FY25 or FY26. While using FY19 as a base some of the weights can be aligned to the present when reckoning some of the indices. This would be a pragmatic way of going about the exercise.

The writer is Chief Economist, Bank of Baroda. Views expressed are personal

Related Topics