Thread by @shrikanth_krish

Two days ago I compared Per-capita GDP of all states relative to India PCI for 2004-05 & 2017-18

Since then conversations with @PoliteVersion and links to an RBI source permit me to make some changes to the analysis I had shared

Two Tweaks to earlier exercise

1. Use of 2017-18 numbers for all states (earlier used 17-18 numbers for most states, but 2016-17 for some)

2. Comparison over time using both Constant prices & Current prices

Now 1 is definitely something we should do.

So let us first make Change 1 and see how results change (while still using "current" prices)

The basic narrative does not change. Telengana remains the state with the biggest jump. But Gujarat now does much better than before

After fixing 1, let's discuss 2.

2 involves using the same price level to make the relative comparisons for both years. (Constant prices)

Previously I had used the prices of 2004 for 2004 numbers and the prices of 2017 for 2017 numbers (Current prices)

Using current prices entails the assumption that inflation rates shouldn't impact the relative standing of states w.r.t "All India" average. We assume that inflation should impact each state similarly

However that need not be the case. Hence the motivation to use constant prices

By using constant 2004-05 prices for the whole period, we get a different set of results

Now Gujarat and Delhi top the list with the biggest "relative jumps"!

Telengana remains close to the top at No 3

Punjab is at the bottom among major states, with either approach

Whether to refer to "current price" based comparison or "constant price" based comparison can be debated.

Intuitively using "constant prices" may make more sense.
But it has its own complications -

a) Consumption baskets can change over time

b) Getting results using "constant 2004" prices" is non trivial, as it involves mapping two different GDP series- 2004-05 & 2011-12

c) Also I have seen conflicting data points in the "constant price" data for 2011-12 series

Even a straightforward comparison of State A and State B as of Year X - should yield similar results regardless of whether we use current Year X prices or "constant" prices from Year X-10

But that's not always the case. So "constant price" results has its qualifiers

I have published both results, and in my view, both lists are worth perusing

Thanks to @PoliteVersion again for providing more data to do the "constant price" analysis

Next step is to expand the scope of the analysis and do a comparison of the PCI of states over time from 1990 till date

That will be for another day.

Sources

1. m.rbi.org.in/Scripts/Annual…

2. niti.gov.in/content/capita…

3. niti.gov.in/content/capita…

4.mofapp.nic.in:8080/economicsurvey…

5. pib.nic.in/newsite/PrintR…

6. mospi.gov.in

PS: For states where 2017-18 numbers weren't available I've extrapolated based on 2014-2016 growth rates

Postscript : I am getting some queries on why do we even need to look at "current price" ranking....Why don't we just go by "constant prices".

There is a reason I shared both. Because of my personal discomfort with the "constant price" data -

Let's take an example -

2013-14 PCI using 2004-05 prices
Gujarat - 63,168
Karnataka - 46,012

2013-14 PCI using 2011-12 prices
Gujarat - 102,589
Karnataka - 101,918

So for the same yr (2013-14), we note that Gujarat and Kar have same PCI using 2011 prices

But 2004 prices lead you to believe otherwise

Hence the rider, and the creation of two lists...

I think this "constant price" conundrum needs to be debated more widely