Quote
Originally posted by: Campion
If you're a numbers geek, maybe you want to take a look at this (mostly regarding CPI but a little on GDP). It's by a friend of a friend. I won't pretend that I understand all of it. I would be curious to hear what you smart f*ckers think
CPI accuracy?
The real question about CPI is "How are you going to use that statistic?". I agree with the premise that CPI tends to understate inflation because of hedonic adjustments and substitution effects which are difficult to measure. I've got no problem with excluding volatile food and energy costs...although maybe they could be annualized and adjusted backwards after 12 month period. If we hadn't excluded energy costs, the last 12 months would have registered deflation with the cost of oil decreases.
But...If those hedonic adjustments weren't included and the increased CPI was used for real life decisions that are tied to the CPI (Like SSI or pension payments), it could feed its own cycle of a general increase in prices. It's never a good thing when your measurements are actually impacting the values you're trying to measure. No Heisenberg.
What we can say about the CPI is that it's a valuable tool for making comparisons in the rate of change of inflation. Whether it represents an accurate point in time measurement is up for debate. It's one of the better measurements we've got.
What might be valuable from a policy perspective is to create a price measurement that looks at what poor people actually spend their money on.....(medical care, fast food, rental housing, cigarettes, used cars, gasoline, electricity, check cashing services & whatevers) and then create a basket of goods for the poor....a PPCPI Poor Persons CPI. I think we'll find that will easily double the CPI.
