Here's a question about CPI in my microeconomics text (Pindyck), which I don't seem to understand:

The price of computers has fallen substantially over the past two decades. Use this drop in price to explain why the Consumer Price Index is likely to overstate substantially the cost-of-living index for individuals who use computers intensively.

In my opinion, the CPI is a universal thing, and the way we calculate the real price in some year with respect to a base year is also fixed, i.e. multiplying nominal price by the ratio of CPI of base year and the year in consideration. So how exactly can CPI overstate or understate the cost of living in any circumstance?

Please help me out, I'm an engineering student and I'm very new to Economics (I've taken up a little course since I find it interesting)

P.S. Here's the solution that's given (which I hardly can make sense of)

The Consumer Price Index measures the cost of a basket of goods purchased by a typical consumer in the current year relative to the cost of the basket in the base year. Each good in the basket is assigned a weight, which reflects the importance of the good to the typical consumer, and the weights are kept fixed from year to year. One problem with fixing the weights is that consumers will shift their purchases from year to year to give more weight to goods whose prices have fallen, and less weight to goods whose prices have risen. The CPI will therefore give too much weight to goods whose prices have risen, and too little weight to goods whose prices have fallen. In addition, for nontypical individuals who use computers intensively, the fixed weight for computers in the basket will understate the importance of this good, and will hence understate the effect of the fall in the price of computers for these individuals. The CPI will overstate the rise in the cost of living for this type of individual.


  • $\begingroup$ Exactly which part of the solution do you not understand? $\endgroup$
    – Herr K.
    Commented Aug 21, 2019 at 18:20
  • $\begingroup$ @HerrK. All of it, as I stated earlier I feel CPI is an exact measure and can't overstate or understate the cost of living. $\endgroup$ Commented Aug 22, 2019 at 2:09
  • $\begingroup$ I would suggest that you create a fictitious basket of goods in excel for two periods. Change the prices of some of the goods in period 2. Let's say you incrase the price of PCs by 20% in period 2, but assign a lower weight to PCs; reduce some of goods' prices for which the weights are higher. Keep the weights constant over the two periods. It will then become clear why CPI could over/understate the effects of individual items in the basket. $\endgroup$
    – london
    Commented Aug 23, 2019 at 6:51


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.