GDP revision a move in right direction, but questions remain: Pronab Sen

The GDP revision improves measurement, says former chief statistician Pronab Sen, but raises questions on double deflation, consumption surge and fiscal maths

India’s new gross domestic product (GDP) series, with 2022-23 as the base year, has introduced major methodological changes — from double deflation to improved coverage of the informal sector and segregation of multi-activity firms. While the government has called this a step towards better accuracy, former chief statistician of India, Pronab Sen, says the changes warrant careful scrutiny. In a video interview with Abhijeet Kumar, Sen talks about what matters, where doubts remain, and what ordinary citizens should take away. Edited excerpts:

 

Which methodological changes under the new series are crucial and why?

 

The issue of double deflation is important. This means you are using different price indices to deflate inputs and outputs. Earlier, we used single deflation, with the price of the final product being used to deflate the whole value added. Here, you are not doing that.

 

 

To my mind, it is an internationally accepted methodology. However, this creates a problem because you will only see the price of the final product. You will see a difference between the deflator used for output and the deflator used for value added. It is very difficult to figure out what that means.

 

The second major change is the segregation of multi-product firms. Earlier, we classified firms by their major product. So, if a manufacturing firm was doing services, all the value of those services went into manufacturing. If these have been separated properly, it is a big improvement. It clarifies where growth is occurring. But it is unlikely to be totally comprehensive.

 

For a non-economist, why does double deflation matter?

 

If you think about GDP, one straightforward way of measuring it is to add up the market value of all finished goods and services. If you are only looking at final products, you can simply deflate those values using the price of the finished product to arrive at constant prices.

 

However, when you measure GDP through value added, that is, by calculating the contribution of each activity and then adding those up, it becomes more complex. In this approach, you are not taking the total value of the product; you are taking the value added at each stage of production.

 

My concern is that, although everything is expressed in monetary terms, you are effectively combining values that have been adjusted using different price measures. In that sense, you are adding apples and oranges.

 

The new series uses survey data like the Annual Survey of Unincorporated Sector Enterprises (ASUSE) and the Periodic Labour Force Survey (PLFS) to estimate the informal sector. Does this capture informal activity better?

 

Earlier, we had surveys once every five years. We took productivity numbers from there and assumed they remained constant. Then we multiplied them by the estimated number of units. Now, with ASUSE, you are getting productivity numbers annually. That is a huge improvement. But the bigger problem is estimating the number of units. The informal sector has very high birth and death rates for units. To have a precise estimate at any given point is difficult.

 

We have not had an economic census for long. That creates a problem. We know many unincorporated units died after Covid, and some new ones are coming up. But a lot of this is still guesswork.

 

Whenever GDP metrics are revised, there is a trust question. Does this revision strengthen credibility or complicate comparability?

 

They have released three years of overlapping data. What we now need to do is sit down with the old series and the new series and check where the differences are prominent and whether these differences are stable. If the differences keep changing, it creates a problem. It is a large exercise because you have to go sector by sector to understand this.

 

The nominal GDP for FY26 is lower than earlier estimates under the old series. How do you see this difference?

 

I think it is because of price. If you look at the new series, the overall inflation taken is 0.9 per cent. That sounds on the lower side. I don’t think any of us feels inflation has been 0.9 per cent. It is practically not there.

 

What in these numbers surprised you most?

 

Consumption has risen sharply. Earlier, it used to grow at about 6 per cent. Now it has grown by over 7.5 per cent. That comes as a surprise. But remember, this is an advance estimate. These numbers undergo revisions. We need to wait for revisions around April or May. Those will be interesting.