[AngrySkillzz]

From a programmer's perspective, you are right that there are no technical barriers to constructing an arbitrary index that delivers whatever value of inflation you want. But of course that is the case for most human things. Which is why institutions are important - hopefully while we are over here being honest and thinking hard about programming, there are people over there being honest and thinking hard about the consumption basket. But then a lot of us work in adtech so maybe that honesty is questionable...

And then there are the other even more severely complicated parts. If next year's iPhone is better but costs the same amount, that's deflation. You got more for the same amount of money. How do we quantify that? Those are called hedonic improvements. How do you account for free services? What is Google worth? What if a new thing comes out that has never existed before? How do you compare that to the previous consumption basket? It would be relatively easy to argue that the preponderance of brand-new, incomparable, improving every year, cheap/free tech stuff that we are living in a relatively deflationary environment. You get more, better stuff every year for less money!

[jjoonathan]

Hedonic improvements like the ones used to multiply up Intel's revenue in order to claim that "US manufacturing is doing better than ever!" while we were shipping our entire industrial infrastructure to China?

https://qz.com/1269172/the-epic-mistake-about-manufacturing-...

I found that episode informative and convincing, but in the exact opposite direction. I stuck my neck out on the basis of those numbers while arguing with conservative family members and it turns out the numbers were rigged after all. For a very reasonable definition of rigged.

In the face of shenanigans, I tend to put more faith in simple concepts like "assets hedge inflation" than in indices.

[AngrySkillzz]

I'll take the other side of that argument. It's an interesting article! But it doesn't make me think that the numbers are "rigged," just that they have been misinterpreted. Semiconductors are getting way better, and have been for a few decades, and that counts as an awesome productivity improvement in manufacturing. If you just look at the top-line number, it says "manufacturing is getting more productive," which in an aggregate sense is still true! But it seems a lot of people though that meant "manufacturing is getting more productive because we have a lot of robots" even though manufacturing employment is declining, rather than "manufacturing is getting more productive because our semiconductor fabs are awesome and the rest of the sector is bleeding out." Which is what it actually meant.

The "US manufacturing is super productive" narrative is unfortunate, especially if policymakers were making decisions based on it. But it doesn't really call into question the data or methodology, just that people have misinterpreted it. But the people who are actual experts in this area are writing papers about that! This paper is from 2014[1], and she's been writing about this since 2010! With economists from the Federal Reserve system, no less.

edit: I guess my point is, the point that article makes isn't, "the game is rigged, inflation is made up, we should use shells for currency." The point it makes is, "don't just read the top-line number and then write an opinion article, make sure you read the breakdown first."

Also, assets do hedge inflation. Most of them, anyway. I never argued otherwise. Just that asset price increases are not the same thing as inflation.

[1] https://research.upjohn.org/up_workingpapers/209/

[jjoonathan]

I'll grant you that it's entirely possible for this sequence of events to have happened spontaneously, without any malice, on the basis of misunderstandings. It could also have happened through the selective promotion of convenient metrics (which I'd argue does constitute rigging, just by a different party), or indeed through actual malice, though I tend to share your doubts on that front.

My invocation of the term "rigged" was not charitable but that's pretty far from saying it was actually incorrect. If I had to bet, I would bet on the middle possibility: metrics are built honestly but selectively promoted in accordance with political agendas in a manner that I would be comfortable to characterize as "rigged."

In any case, assigning blame is one thing and designing portfolios robust to this kind of mishap is another. Fortunately, the difficult parts of assigning blame are completely irrelevant to portfolio design. No matter who was responsible, the correct response is to avoid metrics that are easy to misunderstand (or rig!). Which brings us right back to favoring "assets hedge inflation" over "TIPS hedge inflation."