substitution is fully included. So, for example, if because of high beef prices, people switch to turkey necks and other lower-priced kinds of meat, and just plain eat less meat, this shows up in chained type indexes as a slowing of meat price increases or even a reduction in meat prices. That's why I hate the PCE (known as the "pissy" ) and, instead feature the CPI (the "sippy" ) which has lot less of substitution effects.

I suspect the PCE is the Fed's favorite inflation gauge because it's generally lower than the CPI, and therefore easier to hit the 2% target with it.

Just so people know what's behind all the media bubbly boo about the PCE being the Fed's preferred index.

Another thing to watch out for is the way super-over-emphasis on a 12-month figure (aka "year-over-year", "past year" ), which is full of old data, more than half more than 6 months old. That's fine, for history-loving types.

Also to be noted is that if the latest month-over-month increase is greater than the month-over-month increase from 13 months ago that drops out of the 12-month window, then the 12 month figure increases. Otherwise it decreases. The number leaving the window is just as important as the number entering the window. So when the media bubbly boos about the 12 month figure going up (or down), realize that is all that it boils down to - a comparison of the most recent one-month increase with the one-month increase from 13 months ago.

I prefer to know what recent and current inflation is doing. I prefer the 3 month average for that, annualized (just like they do for GDP).

Here's the rest of the harangue as well as the most recent CPI graphs
https://www.democraticunderground.com/?com=view_post&forum=1002&pid=20666711

so I assume the numbers are still accurate but they can be a bit misleading depending on how one interprets them.

grinding the numbers systematically the same way they always have (or for years and years anyway), so at least they are comparable to previous numbers. Maybe that's kind of an extended definition of honesty, but it's the world we live in

One reason for the quotes around "accurate", aside from statistical sampling error that is always present --

House prices, mortgage costs, cost of home insurance, property taxes, home maintenance and other costs of owning a home are not included in the CPI. Instead they use something called "owners' equivalent rent" -- where they ask the owners what they think they could rent their house for if they moved out. This is a proxy for the foregone rent that homeowners lose by occupying their house.

Specifically on house prices, they consider a house an investment rather than a consumption item. When a house goes up in value and sells for $100k more, the owner gains $100k more, while the buyer pays $100k more, so it's a wash, doncha know? Well, if the former owner now buys an equivalent home, they pay $100k more so they really haven't gained anything in that scenario.

https://www.bls.gov/cpi/factsheets/owners-equivalent-rent-and-rent.htm

As for the PCE, I haven't looked into it, but I think it's much the same re: home ownership costs. Although one thing I read is that shelter has about twice the weighting in the core CPI than in the core PCE.

and his attempts to fire Powell will make people skeptical of future data, even if it is accurate.

Krasnov wants a weak dollar, cheap money.