> Many economists, mind you, believe CPI doesn't correct quite enough, and suspect that it overstates the inflation it hopes to measure by around 1%. This is be

> Many economists, mind you, believe CPI doesn't correct quite enough, and suspect that it overstates the inflation it hopes to measure by around 1%. This is because it tracks the actual prices of a certain "market basket" of goods with specific products in it, and that basket gets out of date, as people substitute products. These substitutions are tricky, if hypothetically a consumer can move from eating fresh local produce to preserved canned produce then it's likely they will make the switch under price pressure when fresh produce increases in cost by 2x. You could calculate CPI based on the new realized purchasing patterns - or you could calculate it based on the desired purchasing pattern. Basing CPI on realized purchasing behavior will lead to errors in how inflation is perceived or where consumers are trading quality for cost. From a monetary policy perspective ignoring this consumer tradeoff could lead to sudden shifts in CPI when consumers run out of quality substitutions. I'd argue we've seen this in housing in the major cities where first home prices were excluded for rental equivalent, then rental quality fell in both the amount of space available in a unit as well as the overall quality of the unit. Eventually you hit the wall where quality can't be traded off any longer and you're left with many people who can't legally house themselves.