>His COTI (Cost-of-Thriving Index) is hand-wavy,

It's hand wavy in the sense that anything approaching an economic discussion in a twitter thread is hand wavy, but if you're looking for detail he links to it: https://www.manhattan-institute.org/reevaluating-prosperity-...

>namely, that CPI is somehow incorrect

He's telling you exactly why he thinks it's incorrect, not 'somehow.'

>means nothing to me, and (lo and behold) validates his hypothesis,

>inflation is fake news

Come on now, I think you owe a bit of generosity to the argument here.

EDIT: Look here, let's just break it down with the car example. Inflation as calculated currently is composed of two parts, let's call them the (1) monetary phenomenon and the (2) technological phenomenon.

So he says: while it's nice that the 2018 Grand Caravan ($26,300 in 2018) has many features the 1996 Grand Caravan ($17,900 in 1996) did not, you still face the problem that you need an extra $8,500 to buy one

and you say: $17,900.00 in 1996 is $28,647.64 in 2018.

But you're just combining 1 and 2 again, not separating them. Dodge is selling their 2018 Grand Caravan in 2018 for $26,300. How much would Dodge be able to sell a 1996 Grand Caravan for in 2018? Certainly not $26,000, maybe more than $17,900, maybe less. But that price, how much Dodge would be able to sell a 1996 Grand Caravan for in 2018 would give us the 'monetary' component of inflation, all the rest is based on technological improvements. And his point is that since there is no way to opt out of those technological improvements - because Dodge doesn't sell 1996 Grand Caravans priced at purely monetary inflation - it's stupid to put the # in an inflation calculator and then say it costs the same because it doesn't, it's more.

Have you looked into how CPI is calculated? It's not that it's "incorrect", just that it doesn't reflect the challenges that real people face trying to afford the basic expenses of life.

I find his COTI quite misleading/confusing. It seems he is counting 100% of health care expenses (100% of premiums) but not counting the ~70-80% of premiums an employer pays the median employee in median weekly earnings.

In the college field, he also uses a semester if college tuition for some reason. Since this is a 1 time cost, it should almost surely be accounted for differently.

In his article he defends all of the choices he made for the COTI calculation. 100% of the premium is used because most people in 1980 paid for the whole thing themselves. I guess it raises the question that the institution of employer sponsorship allowed the costs to balloon, but that is a whole different discussion.