Year-on-year analysis of economic performance serves a very limited purpose | Business Standard Column

lipped from:

A year-on-year analysis is useful in a conceptual way as it accounts for the seasonal impact on the data, but it tells you next to nothing about the current state of the economy.

T C A Srinivasa-Raghavan

I was surprised a few days back to see a well-known, top economist tweet that the First Advance Estimates for GDP growth for 2021-22 has meant that India’s annual GDP growth over the last two years was a mere 0.6 per cent.

While he is technically correct, to give his interpretation importance would be quite unwise because measuring the progress of a post-crisis economy — against what it was pre-crisis — is not a great demonstration of neutrality.

After all, when you gauge the progress of the recovery from a heart attack, you do it from the day the person had the attack, not from the day before, when he or she was running around playing volleyball.

Be that as it may, the tweet reveals a larger malaise among those tracking economic data. They seem to have forgotten what they were looking for.

Take the stubbornness with which everybody measures the IIP, or the “eight core” sectors, not to mention inflation data on a year-on-year basis.

To be sure, a year-on-year analysis is useful in a conceptual way as it accounts for the seasonal impact on the data, but it tells you next to nothing about the current state of the economy.

Let’s take IIP. In April 2021, IIP growth was 134.6 per cent. What does this tell us? That economic activity in April was so much higher than it was during the lockdown month of April 2020? What does that mean? Is 134.6 per cent a good number or is it too low?

Would it not have been more meaningful to say that the IIP contracted nearly 13% in April 2021 compared to March 2021, which would have been one of the first data-based signs of the economic impact of the second wave?

Another thing is the core sector component of the IIP. It’s 40 per cent. But the labour input is very high and when that was suddenly reduced because of the virus, obviously core sector output declined and is picking up slowly.

Similarly with CPI inflation. You can say that inflation in December 2021 was 5.6 per cent compared with the previous December. Or you can say that inflation in December 2021 actually eased by 0.4 per cent compared to the previous month.

But which of these is a more meaningful statement as far as measuring the current situation goes, and which of the two is the man on the street going to care about more?

It must be said, however, that the man on the street probably doesn’t care about either number, as neither accurately captures the price movement in his local mandi. But you get the point.

The Department of Economic Affairs has taken a hybrid approach. In its monthly reports, it neither measures high frequency indicators on a year-on-year approach nor a month-on-month one. Instead, it rebases the data according to a set base month and measures progress against that month.

This is a useful method, especially if one has chosen the base month well. I would disagree with the February 2020 base month the DEA has chosen, as it again pegs progress against the pre-crisis period, but the overall concept does not fall into the trap of a blind year-on-year analysis. That’s good I think.

With a more reasonable base month of, say, June 2020 (long enough after the lockdown was lifted for there to be enough economic activity), such a method would actually provide a much clearer and more informative picture of how we are faring now.

According to this method, if IIP in the post-lockdown month of June 2020 was 100, it was 119 in December 2021, which gives you an immediate idea of the level of progress that has been made since the lockdown was lifted.

I am not saying do away with a year-on-year analysis of data. But, please, don’t cling to it blindly. It serves only a very limited purpose.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s