Search interesting materials

Tuesday, April 20, 2010

Measuring inflation better

In India, inflation measurement is commonly done using year-on-year growth rates. The change in a price index from March 2009 to March 2010 shows the average rate of change over the 12 intervening months. But the most important thing to focus on is the point-on-point change seen in January, February and March. Computing this requires first seasonally adjusting the price index, so as to avoid being confounded by seasonal fluctuations. We do this at with updation every Monday.

Another big issue in inflation measurement in India is the problems of food and fuel. In both cases, there's an element of administered prices, so a jump in the point-on-point seasonally adjusted price level might just be a month in which the government raised a price. This is not a statement about the deeper inflation in the economy. In any case, fuel prices are greatly influenced by the global oil price. Food prices often jump around reflecting a good harvest or a bad harvest. In order to understand the deeper `core' inflation in the domestic economy - the stuff that domestic monetary policy should care about - it's useful to look at the WPI excluding food and fuel. Starting with yesterday's release, we do this also.


  1. Prof. Shah,

    Why do you think assets and hence asset price changes are not counted in for measuring inflation.

    My limited empirical enquiry tells me that while land prices in most places have jumbed 10-20 times in the last 10 years, price of food items, oil, even cars, white goods etc, have eitehr remained same (cars etc) or gone up 2 - 3 X times

    what could be the reason for this phenomenon.

  2. also sir, it is often advocated that cpi is a better measure of inflation than wpi, which we use in much as i know, a proposal to use cpi is lying with the goi since middle 90s...what is going wrong in the implementation that it is taking more than a decade to change the index?

  3. Index of industrial production (Capital Goods) sharply went up to %304 in dec '09 ... why is it so?

  4. Naveen, inflation measurement is about the price of the basket of goods & services purchased by the average household. Asset prices are not in it directly. They are, of course, in it indirectly. E.g. CPI has to have an imputed rental cost, and that goes up when land prices go up.

    Hiranya, CPI is indeed a good idea. The trouble is, CPI measurement in India is less transparent than WPI and involves bigger delays. And, a big chunk of the CPI is in food, and food prices tend to bounce around in ways that have little to do with monetary policy. So someday we'll move to a good CPI but right now we're stuck.

    Hardik, it's 304% annualised. I.e. if it kept going up at that rate for 12 months, it'd yield 304% growth over the year. But even after taking that into account, this is a huge value that I just don't understand. Makes me wonder if there's some data glitch there.

  5. ok proffessor
    but if we look at the present indian scenario, where inflation is attributable to the shortage of food supply, are we really going to be salvaged by monetary tightening? or in otherwords are you trying to say that monetary tightening will not help much in the event of cost push inflation?

  6. Prof. Shah,

    The second part of my question is related to the first.

    Can we say that non-inclusion of asset prices ( land is the best example) is a strong indirect reason for the phenomenon I mentioned. land prices multiplying 10X times compared to those commodities in the basket mutiplying 2-3X times in the last 10 years.

    What would be your explanation for such a phenomenon, if you agree that such a phenomenon is indeed observed.

  7. RBI is now 75.. in terms of their responsibilities, having a good cpi must have been somewhere near the top of their list throughout their lifetime... but no resolution?

    Naveen, I'm not sure what your question really is but if you are questioning why something moved 10X vs something else moved 3X I'm not sure why you would ask that. You don't expect all assets to move at the same rate, do you?

  8. This comment has been removed by the author.

  9. MG... What I have observed is that while most FMCG products, white goods , commodity prices, automobiles etc have either gone down (A/c's) remined more or less same (cars, sugar and wheat) or moved inconsistently, land price in particular has seen consistent increase over the last 10 years. My data from

    How do we explain

  10. Naveen, econ 101: demand/supply on a per asset basis? and investor perceptions?

  11. Ajay,

    Even if its confusing because of the jumps, you have to include diesel fuel and food.

    The average household doesn't care about the 75 (or whatever other are) basket constituents.

    The average household cares about transport, food, electricity, entertainment, communication. Number crunchers like to invent new indicators or proxies because they don't like uncertainty in their data. Common people have to deal with it, so you deal with it too. Data should represent reality, tweak your model, not the data.

  12. This is a very interesting post, and the comments are also fantastic to read. I’ll have to have a little re-think about my own contact form on our new website


Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticism is perfectly okay; uncivilised language is not. We delete any comment which is spam, has personal attacks against anyone, or uses foul language. We delete any comment which does not contribute to the intellectual discussion about the blog article in question.

LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.