Why does TIPS follow CPI?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
CorporateFinGuy
Posts: 8
Joined: Wed Sep 09, 2020 4:25 pm

Why does TIPS follow CPI?

Post by CorporateFinGuy »

I have read from several sources that CPI in the past has overestimated the rate of inflation. If this is true, does this mean that the spread between CPI and the actual inflation is considered a "free lunch"? As far as I know the only "free lunch" is diversification, so this threw me off.

Another question is that if the CPI is not accurate in measuring expected inflation, why is CPI used for TIPS? Is it because while CPI is not perfect, it is the best option?

Thank you,
CorporateFinGuy
User avatar
David Jay
Posts: 11681
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Why does TIPS follow CPI?

Post by David Jay »

Not following your “free lunch” language. Short term TIPS are selling for well under 0% so even adding in CPI it is unlikely that the return will exceed inflation.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
alex_686
Posts: 9061
Joined: Mon Feb 09, 2015 2:39 pm

Re: Why does TIPS follow CPI?

Post by alex_686 »

CorporateFinGuy wrote: Tue Sep 14, 2021 8:37 am I have read from several sources that CPI in the past has overestimated the rate of inflation. If this is true, does this mean that the spread between CPI and the actual inflation is considered a "free lunch"? As far as I know the only "free lunch" is diversification, so this threw me off.
You can't measure inflation. If you could measure inflation it would be a ordinal number (1st, 2nd, 3rd, etc.).

Take a look at cars today verse 40 years ago. They are more expensive. Right? We can just compare the prices between a new car over 40 years. Except that would be plain wrong. Cars have radically changed over 40 years. They have become more larger, more dependable, safer, higher fuel efficiency more features, etc. You have to account for all of these in adjustments. These adjustments tend to overstate inflation.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
User avatar
jeffyscott
Posts: 9942
Joined: Tue Feb 27, 2007 9:12 am
Location: Wisconsin

Re: Why does TIPS follow CPI?

Post by jeffyscott »

CorporateFinGuy wrote: Tue Sep 14, 2021 8:37 amI have read from several sources that CPI in the past has overestimated the rate of inflation.
Is this, perhaps, based on CPI vs. PCE?

I would not consider it a free lunch because when market participants set the price in the market or at auction, they are fully aware that the coupon payments and principle are based on CPI.

The inflation indexed security that is currently a free lunch would be I-bonds, this is just because they do not sell them with a real yield below 0.
The two greatest enemies of the equity fund investor are expenses and emotions. ― John C. Bogle
User avatar
dodecahedron
Posts: 5809
Joined: Tue Nov 12, 2013 12:28 pm

Re: Why does TIPS follow CPI?

Post by dodecahedron »

CorporateFinGuy wrote: Tue Sep 14, 2021 8:37 am Another question is that if the CPI is not accurate in measuring expected inflation, why is CPI used for TIPS? Is it because while CPI is not perfect, it is the best option?
CPI has always been acknowledged as inherently flawed (as have pretty much all operational economic concepts, such as GDP, money supply, risk-free rate of return) but it represents the best efforts of the consensus of the economics profession.

Over the years, the Bureau of Labor Statistics, which is charged with responsibility for collecting the data and computing the CPI, has continued to refine the methodology. It started out much cruder than today, for several reasons. (Technology has improved massively from the early pre-computer days. The data series starts in 1913! The modern economics profession was also in its methodological infancy at the time.)

I started studying economics in the early 1970s, as inflation was starting to heat up to levels that were high compared with recent history. I remember reading in my textbooks at that time about the limitations of the methodology, with suggestions for addressing some of those limitations. Some of those suggestions have since been adopted.

But it can never be perfect. How do you deal with adjusting for massive quantum improvements in product quality or entirely new products that never existed before? No easy answers to some of the methodological issues.

You can find some discussion of recent methodological changes here:
https://www.bls.gov/cpi/additional-reso ... ucture.htm

Edited to add: the CPI is not a monolithic concept. There are various "flavors" of CPI, e.g., CPI-E, with a basket of weights reflecting typical purchases of elderly, CPI-W with a basket of weights reflecting typical purchases of workers, CPI-U with a basket representing urban purchasers, etc. TIPS and series I bonds use CPI-U while Social Security uses CPI-W. Social Security discusses why they don't use CPI-E here.

Further edited to add: while one can certainly argue that the CPI concept (in all its varieties) retains flaws, I am grateful that we no longer live in times like when I first began studying economics in the early 1970s. At that time, SS was not indexed for inflation at all, nor were the parameters of the tax code (which made bracket-creep a big issue at the time.) And of course, neither TIPS nor I bonds existed back then.
Last edited by dodecahedron on Tue Sep 14, 2021 4:53 pm, edited 1 time in total.
dbr
Posts: 37019
Joined: Sun Mar 04, 2007 9:50 am

Re: Why does TIPS follow CPI?

Post by dbr »

CorporateFinGuy wrote: Tue Sep 14, 2021 8:37 am I have read from several sources that CPI in the past has overestimated the rate of inflation. If this is true, does this mean that the spread between CPI and the actual inflation is considered a "free lunch"? As far as I know the only "free lunch" is diversification, so this threw me off.
The dilemma here is what is even the definition of "actual inflation" let along any means to measure it. Probably the idea of the above free lunch is in my category of notions that are "not even wrong."

In practical terms if the dollars needed to be spent by an individual to maintain a certain lifestyle increase less rapidly than CPI then a CPI indexed investment is a good deal for that individual. It is the opposite retro circum.

There are certainly debates which of many methods of measuring inflation are most appropriate or most fair to different sets of individuals.
alex_686
Posts: 9061
Joined: Mon Feb 09, 2015 2:39 pm

Re: Why does TIPS follow CPI?

Post by alex_686 »

This might be a good example.

To measure inflation you need to find 2 different baskets of goods that generate the same amount of utility.

In 1970 a LP cost around $5. 46 minutes of music.

2020 a 1 month subscription to Spotify cost around $10. Unlimited music.

Based on this, can we say that we have had 100% inflation? Well, only if the 2 baskets are equivalent - and they are certainly not. Owing a piece of plastic forever verse having access to a huge library of music. The fidelity quality of a LP - assuming there are no scratches and you still have vacuum tubes to run your stereo - verse the ability to play the music over your phone. Some people think that LPs are great and steaming is of poor value. Others think the other way.

Remember when I said inflation was ordinal? Well, the way CPI works is that the new basket has to be at least as good. This is the critical bit. As in the music example above you can see that the lines are blurry. So when steaming music is substituted for LPs the least favorable case is assumed. If you think that LPs are great you should be left in a neutral position. If you think that streaming is great the CPI has overstated inflation in your case.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
JackoC
Posts: 2629
Joined: Sun Aug 12, 2018 11:14 am

Re: Why does TIPS follow CPI?

Post by JackoC »

CorporateFinGuy wrote: Tue Sep 14, 2021 8:37 am 1. I have read from several sources that CPI in the past has overestimated the rate of inflation. If this is true, does this mean that the spread between CPI and the actual inflation is considered a "free lunch"? As far as I know the only "free lunch" is diversification, so this threw me off.

2. Another question is that if the CPI is not accurate in measuring expected inflation, why is CPI used for TIPS? Is it because while CPI is not perfect, it is the best option?
1. As was pointed out, any link we can find discussing flaws in CPI can be assumed as well known to buyers and sellers in the TIPS market as it is to us. Free lunch potential in the TIPS market would only be in the relatively unlikely case we can develop here an analysis of CPI that's superior to any in the market...and nobody else cottons on to it by reading the open forum thread. :happy

I-bonds are a 'free lunch' now because of the quirk of not allowing a variable rate below zero, when the 5 yr TIPS (I-bonds redeemable without penalty in 5 yrs) yields -1.7%. But it's nothing directly to do with the index.

2. It's the last, because it, CPI-U, is viewed as the closest proxy to 'inflation' for consumers. Not because anyone (reasonable) thinks there is an 'absolutely correct' overall measure of the change in the price level, let alone one which would simultaneously represent the increase in different baskets of goods and services that different people purchase, understanding there is no market mechanism which forces prices for all goods and services to increase at the same rate. Many criticisms of CPI-U by ordinary people come back to expecting one measure to be accurate for different baskets: impossible. There can be criticisms of more merit but within the mainstream of economics I think it's fair to say they are generally marginal, and economists who believe CPI-U overstates 'real' inflation appear to outnumber those who think the opposite. Among ordinary people it would seem the other way around, but again I believe a lot of that is the issue is which basket, and a good deal of it is also cherry picking and assuming a certain steady increase in economic welfare is the 'neutral' outcome. We tend to pay more attention to disturbing price increases (for some things) than lack of even 'normal' ones (for others). And we tend to feel eg. cars or computers being more capable is something we deserve, their makers are *finally* giving us more reasonable value for our money, not something that should count as reducing calculated inflation.
Topic Author
CorporateFinGuy
Posts: 8
Joined: Wed Sep 09, 2020 4:25 pm

Re: Why does TIPS follow CPI?

Post by CorporateFinGuy »

Thank you so much everyone, I understand now thanks to you guys. :)

Have a great day.

Best regards,
CorporateFinGuy
Post Reply