But is Obama using the right numbers? Chained CPI is by definition not an apples-to-apples index. It tries to correct for what is called substitution bias, the idea that if prices are rising in apples, the basket of goods used to calculate inflation should be adjusted to include more of a substitute rising less fast. A truer measure of inflation would count the increase in the price of apples based on how many apples people were eating prior to the price increase, not just assume away the increase based on the assumption that people will switch to oranges. I like oranges. But if the price of apples is soaring, inflation figures should reflect this.
So chained CPI is a fudge, and a slippery slope. Taken to its logical conclusion if the price of steak is soaring, but the price of pink slime (or, to give it its euphemistic name “lean finely-textured beef”) remains cheap then consumers may be assumed to substitute pink slime for steaks. That isn’t measuring the cost of living. That’s just an austerity fantasy.
Trying to appease the Washington Post editorial board is no substitute for sound economic principles. When we measure inflation, we should use the best data available. As far as I can tell, that’s MIT Billion Prices Project which indexes by far the largest range of prices. More data means more accuracy.
According to their methodology, the CPI is very slightly underestimating the level of inflation, not overestimating it: