are TIPS expensive? or are they cheap?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
MtnBiker
Posts: 178
Joined: Sun Nov 16, 2014 4:43 pm

Re: are TIPS expensive? or are they cheap?

Post by MtnBiker » Fri Jul 12, 2019 11:27 pm

Latest from Swedroe: https://www.advisorperspectives.com/art ... tips-cheap
The bottom line is that TIPS look cheap relative to both Treasuries and CDs.

Topic Author
grok87
Posts: 8772
Joined: Tue Feb 27, 2007 9:00 pm

Re: are TIPS expensive? or are they cheap?

Post by grok87 » Sat Jul 13, 2019 6:46 am

MtnBiker wrote:
Fri Jul 12, 2019 11:27 pm
Latest from Swedroe: https://www.advisorperspectives.com/art ... tips-cheap
The bottom line is that TIPS look cheap relative to both Treasuries and CDs.
thanks
in the linked article Larry cites the philly fed survey as being at 2.20% for 10 year CPI
the 5 year CPI figure is 2.10% which may be more appropriate for the 5 year tips analysis.

personally i think that figure is too high and my number more like 1.8-1.9%
RIP Mr. Bogle.

User avatar
Kevin M
Posts: 10921
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Kevin M » Sat Jul 13, 2019 3:18 pm

MtnBiker wrote:
Fri Jul 12, 2019 11:27 pm
Latest from Swedroe: https://www.advisorperspectives.com/art ... tips-cheap
The bottom line is that TIPS look cheap relative to both Treasuries and CDs.
Once again, the problem with Larry's analysis is in the conflicts between these observations:
The liquidity premium (which depresses yields) and the risk premium for unexpected inflation (which increases yields) work in opposite directions and may cancel each other out. If that is the case, the TIPS-to-nominal-bond spread is a good indicator of the market’s aggregate view of expected inflation.
Thus, ignoring the liquidity premium, instead of having to pay a premium to hedge the risk of unexpected inflation, investors are actually being paid a premium (0.79%).
I'm not sure that it's correct to categorize this as ignoring the liquidity premium, but the main point is that there is no justification given for ignoring either premium, and there is no analysis at all about the premiums. The observation about the premiums possibly canceling each other out is simply ignored.

To clarify what Larry is saying in the first quote--the one where he describes the premiums and mentions that they may cancel each other out ...

First, he is talking about effects of the two risk premiums on the nominal Treasury yield. So Larry is characterizing the premiums as either increasing or decreasing the nominal Treasury yield relative to a fixed TIPS yield, and thus increasing or decreasing the BEI rate (nominal Treasury yield minus TIPS yield) relative to the market's expected value for inflation.

The liquidity premium he refers to is a nominal Treasury liquidity price premium. The liquidity of nominal Treasuries increases demand, which increases price, which "depresses" yield. This premium will tend to make the BEI rate lower than the market's actual inflation expectation. Or in other words, expected inflation is higher than the BEI if this premium is the only factor influencing prices/yields.

The "risk premium for unexpected inflation" is a nominal Treasury yield premium. Since it's a yield premium, it tends to increase nominal Treasury yield. This premium will tend to make the BEI rate higher than the market's expected value for inflation. In other words, expected inflation is lower than the BEI if this premium is the only factor influencing prices/yields.

Since Larry (and Grok) are arguing that expected inflation is higher than the BEI (which is what they think makes TIPS cheap), I would characterize it as ignoring the risk premium for unexpected inflation, or at least that the liquidity premium is larger than the unexpected inflation risk premium, and that difference is the difference between expected inflation and the BEI. The lack of any analysis about these premiums in the article seems to me to be a bit of a non sequiter in the analysis.

Alternatively, if we "ignore the liquidity premium" as Larry says, meaning that we assume that it's 0%, then we are saying that the nominal Treasury unexpected inflation risk yield premium is negative. The implication is that not only are investors not demanding a nominal Treasury yield premium for unexpected inflation risk, they are actually accepting a yield discount for unexpected inflation risk. It seems to me that the implication of this is that investors, in aggregate, are more concerned about unexpected deflation than unexpected inflation. Thoughts on this?

Now, Larry may have done an analysis of the two premiums that's not shared in the article, and may still conclude that TIPS are cheap based on this analysis. I've seen Larry chime in on the forum a bit lately, so maybe he'll chime in here to clarify.

We've seen references to models that estimate one premium or the other, but I don't recall seeing a reference to a model that estimates both premiums, including data as to what they currently are. That would be an extremely useful addition to this conversation.

I would anticipate that the argument is something like: we can't observe the premiums, and there are no trustworthy models for both premiums, so we'll just use a survey and/or TIPS swaps for our expected inflation estimates. My counterargument to using surveys is: why should we trust surveys more than the market? Regarding TIPS swaps, I recall seeing a reference to liquidity issues with them as well, so I'd like to see more analysis about using TIPS swaps.

Since TIPS swaps rates are market driven, I think I'd place more confidence in them than in surveys if I understood more about any risk premiums that might be influencing the TIPS swaps rates relative to expected inflation rates. More info on this would be a useful contribution to the conversation.

Also, is there a reliable, easy to access source for current TIPS swaps rates? I recall doing a little web searching for this, but found the results confusing, and I wasn't persistent enough to spend much time on it.

Finally, let's not ignore this:
As long as investors are willing to sacrifice the conveniences of a fund (such as the ability to automatically reinvest interest), they can buy individual certificates of deposit (CDs) instead of a Treasury bond or a Treasury bond fund. FDIC-insured CDs typically carry significantly higher yields than Treasury bonds of the same maturity – which is why my firm, Buckingham Strategic Wealth, generally buys them instead of nominal Treasury bonds when building laddered portfolios.
<snip>
As I write this, the yield on a five-year CD in the secondary market (where CDs trade just like bonds) is 2.40%.
<snip>
Note that itis often possible to find higher yields in the primary market by shopping around at such sites as this, which would make CDs relatively more attractive.)
To "shop around", we can simply go to https://www.depositaccounts.com/cd/5-year-cd-rates.html, and observe that the top yield on a 5-year direct CD is 3.30% APY, so 90 basis points higher than the brokered CD rate of 2.40% that Larry quotes, and about 145 basis points higher than the 5-year Treasury yield of about 1.85%. If one is willing to "shop around", TIPS no longer look cheap, even based on an analysis that lacks any estimation of the risk premiums, and relies on Fed surveys or TIPS swap rates (or Grok's personal inflation estimates).

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

Angst
Posts: 2019
Joined: Sat Jun 09, 2007 11:31 am

Re: are TIPS expensive? or are they cheap?

Post by Angst » Sat Jul 13, 2019 8:59 pm

Kevin M wrote:
Sat Jul 13, 2019 3:18 pm
MtnBiker wrote:
Fri Jul 12, 2019 11:27 pm
Latest from Swedroe: https://www.advisorperspectives.com/art ... tips-cheap
The bottom line is that TIPS look cheap relative to both Treasuries and CDs.
Once again, the problem with Larry's analysis is... [Snip... Big Snip!]
Kevin,
I really enjoyed reading your post. I'd never even thought about how the pricing of CD's might be considered in regards to trying to discover a BEI rate. I suppose though, it would sure be nice if there were also also a market including inflation adjusted CD's. If it existed, might not this differential you came up with all come out in the wash, so to speak? Interesting to think about regardless - thank you for your post.

User avatar
Kevin M
Posts: 10921
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Kevin M » Sat Jul 13, 2019 10:32 pm

Kevin M wrote:
Sat Jul 13, 2019 3:18 pm
Finally, let's not ignore this:
<snip>
Note that itis often possible to find higher yields in the primary market by shopping around at such sites as this, which would make CDs relatively more attractive.)
To "shop around", we can simply go to https://www.depositaccounts.com/cd/5-year-cd-rates.html, <snip>
I hadn't noticed that Larry provided a link to the same depositaccounts.com URL in his article that I referenced (I underlined the word in the article quote above that included the link).

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

larryswedroe
Posts: 15955
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: are TIPS expensive? or are they cheap?

Post by larryswedroe » Sun Jul 14, 2019 7:46 am

Since Larry (and Grok) are arguing that expected inflation is higher than the BEI (which is what they think makes TIPS cheap),
Kevin, don't know about Grok but I've made no such argument and don't know how anyone could draw that conclusion.

As to the Swap method of determining inflation, that is one measure of market's view, (though even that has counterparty risk in it) and I would argue that a consensus of 80 of the country's leading economists as financial institutions is a good measure of estimated inflation as those forecasts influence investor expectations.

As to the two premiums, liquidity and unexpected inflation, there is simply no way to parse those two risks. Which is why I have not done so. The liquidity premium is typically fairly low, accept in crisis (we are not in one), and risk premium is likely low when you have very stable inflation for long time, as we have had. And larger when it is been more volatile. IMO it is a reasonable to assume with two likely low premiums that they roughly offset.

What I have recommended is simply doing the breakeven analysis and deciding if you are willing to take the risk of unexpected inflation based on the size of the premium. For some 30bp would be enough, perhaps workers with stable jobs whose income tends to rise with inflation, and perhaps 50 or even higher would be enough for a retiree with low tolerance for unexpected inflation. There is obviously no right answer, just one right for each person

Primary CDs almost always have a higher yield, sometimes it small advantage and sometimes large. But one should be careful and be sure they know the credit rating of the institution when buying longer term CDs because while there is no risk of principal loss if under FDIC limit, there is risk of losing the high yield if bank defaults/gets taken over. The bank taking over has right to change the rate. Also you lose liquidity with primary CDs. On other hand they have advantage of put option if rates rise, typically the cost of X months interest.

Hope that is helpful
Larry

Angst
Posts: 2019
Joined: Sat Jun 09, 2007 11:31 am

Re: are TIPS expensive? or are they cheap?

Post by Angst » Sun Jul 14, 2019 9:33 am

larryswedroe wrote:
Sun Jul 14, 2019 7:46 am
Since Larry (and Grok) are arguing that expected inflation is higher than the BEI (which is what they think makes TIPS cheap),
Kevin, don't know about Grok but I've made no such argument and don't know how anyone could draw that conclusion.

[Snip...]
I'm dumbfounded, I just don't get it. I've read the article in full, twice three times now, and without claiming that Larry has explicitly stated that expected inflation actually IS higher than the BEI, I find it hard to see how that isn't at least strongly implied. He ticks off multiple sources of inflation predictions as examples of why TIPS are a the better choice over nominals...

Moreover, Larry's inability to "know how anyone could draw that conclusion" I find stunning.

larryswedroe wrote:
Sun Jul 14, 2019 7:46 am
What I have recommended is simply doing the breakeven analysis and deciding if you are willing to take the risk of unexpected inflation based on the size of the premium. For some 30bp would be enough, perhaps workers with stable jobs whose income tends to rise with inflation, and perhaps 50 or even higher would be enough for a retiree with low tolerance for unexpected inflation. There is obviously no right answer, just one right for each person
The spirit of the article is fairly bullish for TIPS and this forum posted (above) caveat is barely alluded to and really beside the point I'm addressing. Overall, the article comes across (at least to me) as an enthusiastic vote for (5 yr) TIPS over (5 yr) nominal Treasuries due to (various) expected inflation predictions exceeding BEI. Now I do recognize I'm not the sharpest tool in the shed and I probably need an empathetic advisor who is capable of putting him or herself into my shoes and imagining the root of my apparent misapprehension of what I've read multiple times, but clearly lacking that, I appeal to the BH community: Have you read the article? How do you interpret it? Am I stupid?

https://www.advisorperspectives.com/art ... tips-cheap

larryswedroe
Posts: 15955
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: are TIPS expensive? or are they cheap?

Post by larryswedroe » Sun Jul 14, 2019 10:38 am

Angst
I reread the piece and I repeat I don't know how anyone could draw that conclusion. All I've said is that based on market estimates of inflation TIPS look cheap against Treasuries (really cheap) and even against CDs. That's ALL I've said. I've said nothing in the article about expected inflation is higher than estimates. What I have said is that investors averse to risk of unexpected inflation should prefer TIPS unless the premium for unexpected inflation is high. And right now it is negative relative to Treasuries and very small vs secondary CDs. You are simply reading into the piece what is not there. I don't know why that is the case. It has nothing do with any view I might have about expected inflation (and I have none, other than the market's, my crystal ball is cloudy).
Larry

User avatar
Kevin M
Posts: 10921
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Kevin M » Sun Jul 14, 2019 1:23 pm

Thanks for contributing Larry--really appreciate it.
larryswedroe wrote:
Sun Jul 14, 2019 7:46 am
Since Larry (and Grok) are arguing that expected inflation is higher than the BEI (which is what they think makes TIPS cheap),
Kevin, don't know about Grok but I've made no such argument and don't know how anyone could draw that conclusion.
<snip>
What I have recommended is simply doing the breakeven analysis and deciding if you are willing to take the risk of unexpected inflation based on the size of the premium.
You're not arguing that expected inflation is higher than the BEI rate, or just that this is not what makes TIPS cheap (or both)?

Let's look at those one at a time.

Doing the breakeven analysis to calculate size of the premium is just comparing an estimate of expected inflation to the BEI rate.
From the article:
The survey’s second quarter 2019 estimate for inflation over the next 10 years is 2.20%, which is 0.79 percentage point higher than the breakeven inflation rate of 1.51%, obtained by comparing the yields on TIPS and nominal Treasury bonds. (The survey’s forecast for the period ending 2022 is 2.10%). Thus, ignoring the liquidity premium, instead of having to pay a premium to hedge the risk of unexpected inflation, investors are actually being paid a premium (0.79%).
You are calculating the TIPS premium as expected inflation minus BEI rate = 2.20% - 1.51% = 0.69% (correcting your arithmetic, but not important conceptually). For this premium to be positive, expected inflation (estimated by whatever means you choose) must be higher than the BEI rate. An inescapable conclusion, really just a tautology, is that you think that expected inflation is higher than the BEI rate, which is the first underlined part (in the quote above) of what I stated.

I don't know how anyone could come to any other conclusion.

Later in the article:
Given that investors in TIPS are not currently sacrificing any expected return (in fact, they have a higher expected return), they should greatly prefer five-year TIPS to five-year Treasury bonds.
This seems to be saying that TIPS are cheap (should be greatly preferred) because of the calculated TIPS premium (higher expected return). Since a positive TIPS premium requires that expected inflation must be higher than the BEI rate, the flow of the articles logic suggests that TIPS are cheap because they have higher expected return than nominal Treasuries, which is because the expected inflation rate is higher than the BEI rate.

I don't see how anyone could not come to these conclusions based on the flow of the article's logic.

To be fair, your reasoning goes further, and discusses how TIPS still could be preferred even if there is a significant unexpected inflation risk (yield) premium for nominal Treasuries, in which case the TIPS premium, as you calculate it, would be negative (ignoring any liquidity premium). But there's a difference between being preferred and being greatly preferred, and it seems that a main point of the article is that TIPS are greatly preferred (i.e., "cheap") because of the positive TIPS premium.
As to the Swap method of determining inflation, that is one measure of market's view, (though even that has counterparty risk in it) and I would argue that a consensus of 80 of the country's leading economists as financial institutions is a good measure of estimated inflation as those forecasts influence investor expectations.
That's a reasonable argument, but you've also argued strongly that no one is good at predictions. I'm left wondering if there is more than this survey that goes into institutional investors' decisions about how to place their bets on expected inflation.
As to the two premiums, liquidity and unexpected inflation, there is simply no way to parse those two risks. Which is why I have not done so. The liquidity premium is typically fairly low, accept in crisis (we are not in one), and risk premium is likely low when you have very stable inflation for long time, as we have had. And larger when it is been more volatile. IMO it is a reasonable to assume with two likely low premiums that they roughly offset.
Although I completely agree with all of this, having you say it is really confusing, because in the article you said that if these premiums offset each other:
If that is the case, the TIPS-to-nominal-bond spread is a good indicator of the market’s aggregate view of expected inflation.
Simply following the logic of these statements leads to the conclusion that the BEI rate is "a good indicator of the market’s aggregate view of expected inflation", which eliminates any TIPS premium as you calculate it in the article. It also contradicts the view that survey estimates are a better indication of expected inflation than the BEI rate (which is a good indicator of market's aggregate view of expected inflation if the two risk premiums offset each other).

If the BEI rate is a good indicator of the market's aggregate view of expected inflation, then the net effect of all risk premiums is approximately 0%, in which case I'd conclude that TIPS are fairly priced rather than cheap--at least as far as the aggregate market goes.

Of course your other points still apply, and TIPS might still be preferred by a particular individual, even greatly preferred, if TIPS are fairly priced, or even if they're expensive relative to nominal Treasuries, as it's the insurance to hedge unexpected inflation that is what might be most important to that particular individual.

Again, really appreciate you contributing to this thread. Hope you can shed some light on the apparent contradictions I've outlined here.

Thanks,

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

thx1138
Posts: 984
Joined: Fri Jul 12, 2013 2:14 pm

Re: are TIPS expensive? or are they cheap?

Post by thx1138 » Sun Jul 14, 2019 2:43 pm

I think Larry is misreading what the various posters here are saying his article claims.

Larry appears to objecting that folks are saying he is claiming expected inflation is higher than estimates. But no one is saying he wrote that.

Rather the statement made was that the article says estimates (not Larry’s estimates but various other estimates Larry cites in the article) for inflation are higher than the BEI between real and nominal. And really the article does say that... because it is true.

After that there really only seems to be a potential argument over what “cheap” means and does Larry’s article imply that or not. Larry already pointing out that it does depend on the investor but the article does certainly set the tone that things are very favorable for TIPS at the moment (which it seems reasonable to consider TIPS being “cheap” at the moment).

But it really seems like folks are talking past each other here...

IFYOUCAN
Posts: 14
Joined: Mon Dec 24, 2018 10:48 pm

Re: are TIPS expensive? or are they cheap?

Post by IFYOUCAN » Sun Jul 14, 2019 3:33 pm

I read the article only once and being under 40, Larry's suggesting of job wage keeping up with inflation sealed the deal for me, thanks Larry.

larryswedroe
Posts: 15955
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: are TIPS expensive? or are they cheap?

Post by larryswedroe » Sun Jul 14, 2019 6:27 pm

Kevin, Y
ou are calculating the TIPS premium as expected inflation minus BEI rate = 2.20% - 1.51% = 0.69% (correcting your arithmetic, but not important conceptually). For this premium to be positive, expected inflation (estimated by whatever means you choose) must be higher than the BEI rate.
that is just plain wrong as you ignore a risk premium for unexpected inflation as well as the liquidity premium. As I explained. We could easily have the difference explained by a liquidity premium for example.

Let's say TIPS yield .3 and Treasuries yield 1.8 that yields breakeven inflation of 1.5. But expected inflation is much higher than 1.5. So it is likely we have a liquidity premium at work here, people willing to pay say 0.3 premium for liquidity. Just one example. The liquidity premium could be a lot higher. It could be example that buyers of US Treasuries don't by TIPS, like perhaps foreign central banks. Could be other buyers who want liquidity and don't care about inflation hedging benefits.


Larry

User avatar
Kevin M
Posts: 10921
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Kevin M » Sun Jul 14, 2019 7:41 pm

larryswedroe wrote:
Sun Jul 14, 2019 6:27 pm
Kevin, Y
ou are calculating the TIPS premium as expected inflation minus BEI rate = 2.20% - 1.51% = 0.69% (correcting your arithmetic, but not important conceptually). For this premium to be positive, expected inflation (estimated by whatever means you choose) must be higher than the BEI rate.
that is just plain wrong as you ignore a risk premium for unexpected inflation as well as the liquidity premium. As I explained. We could easily have the difference explained by a liquidity premium for example.
I don't understand how you can say it is wrong when the calculation is straight from the article you wrote (except for the arithmetic error correction)! And the logic of what I'm saying is simple:

If A - B > 0, then A > B.

So what is wrong? The calculation in your article, or the straightforward logic of simple math?

I agree that the difference could be explained by a difference between the unexpected inflation and liquidity premiums, as I explained in great detail in an earlier reply. But, you said:
larryswedroe wrote: IMO it is a reasonable to assume with two likely low premiums that they roughly offset
You can't have it both ways. Either the premiums offset each other, in which case BEI rate = expected inflation (aggregate market estimate), or one is larger than the other, and BEI rate <> expected inflation rate.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

larryswedroe
Posts: 15955
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: are TIPS expensive? or are they cheap?

Post by larryswedroe » Mon Jul 15, 2019 8:13 am

Kevin
It is a reasonable assumption IMO in normal times to guess that the two premiums MIGHT offset. But we don't know that is the case, and it is clearly time varying---see 2008. That time varying nature is why I went through the MATH and the BEI rate (which is clearly different than expected inflation, hence labeled as BEI and not expected inflation).

So here is exactly what the article says:
The liquidity premium (which depresses yields) and the risk premium for unexpected inflation (which increases yields) work in opposite directions and may cancel each other out. If that is the case, the TIPS-to-nominal-bond spread is a good indicator of the market’s aggregate view of expected inflation.
Note the word MAY cancel. And IF THAT IS THE CASE. Doesn't say it IS THE CASE.

Looking at the current yield it appears that the liquidity premium is high and the risk premium for unexpected inflation is low (since we have had very stable and low inflation for years. That would be my operating assumption today. Hence TIPS look cheap relative to treasuries for typical investor who doesn't need liquidity (especially since likely holding fixed income in tax advantaged accounts, and may not need more than their RMD). How cheap depends on one's risk aversion to unexpected inflation (not expected inflation). The more risk averse the more the premium one should be willing to pay. Which is why I also took the pains to show the research
Again it has nothing to do with any prediction of inflation as I have never made one, only about managing risk

I will point out that I have done this type analysis for many years on my blogs and you are the first person to misunderstand what I wrote. Perhaps I didn't write it clearly enough but thousands of people have read these posts over the years and not one had ever raised the issue that you did.
Hopefully it is clear now.
Larry

User avatar
Kevin M
Posts: 10921
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Kevin M » Mon Jul 15, 2019 8:28 pm

Larry,

Thanks for clarifying your thoughts.

I didn't misunderstand what you wrote in the article at all.

Kevin
Last edited by Kevin M on Wed Jul 17, 2019 7:17 pm, edited 1 time in total.
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Kevin M
Posts: 10921
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Kevin M » Wed Jul 17, 2019 7:16 pm

Kevin M wrote:
Sat Jul 13, 2019 3:18 pm
We've seen references to models that estimate one premium or the other, but I don't recall seeing a reference to a model that estimates both premiums, including data as to what they currently are. That would be an extremely useful addition to this conversation.
I found something on the main Federal Reserve website, in the Economic Research -> FEDS notes section. Sorry if it's been cited already, but I don't recall seeing it. Here's the link to the article: Tips from TIPS: Update and Discussions. It's dated May 21, 2019, so fairly recent. From the introduction:
The spread between the yield on a nominal Treasury security and that on a Treasury inflation protected security (TIPS) of comparable maturities—usually called the "breakeven inflation rate" or "inflation compensation"—has been often used as a real-time proxy for market participants' inflation expectations. However, policymakers and market participants are also cognizant that this spread is an imperfect measure, as it contains other components that can drive a wedge between inflation compensation and market participants' true inflation expectations.

In a recently published paper (D'Amico, Kim, and Wei (DKW), 2018), D'Amico and two of us use a no-arbitrage term structure model to decompose TIPS inflation compensation into three components—inflation expectation, inflation risk premium, and TIPS liquidity premium—over the 1990-2013 period. The model is also used to decompose nominal yields or forward rates into four components—expected real short rate, expected inflation, inflation risk premium, and real term premium.
Note from the underlined part of the first paragraph that what we refer to in this thread as breakeven inflation rate is also referred to as inflation compensation. The latter term is what's mostly used in the rest of the article, as well as in some other papers I've read.

The underlined part of the second paragraph is just repeating something we've been discussing since close to the beginning of the thread.

The article is a good read, but what's really cool is this:
Beginning with the publication of this Note, we will provide monthly updates of model decompositions of the 5-year, 10-year, and 5-to-10-year nominal Treasury yields and inflation compensation up to the last business day of the previous month. In general, we will endeavor to have the updated series posted sometime after 10:00 a.m. on the fourth business day of each month. Because this is a staff research product and not an official statistical release, it is subject to delay, revision, or methodological changes without advance notice.17

The latest estimates will be posted as a comma-separated values (CSV) file at the URL https://www.federalreserve.gov/econres/notes/feds-notes/DKW-updates.csv. As of this publication, daily data are available for the period from January 3, 1983 to June 28, 2019.
From the download, the 6/28/2019 modeled 5-year values are TIPS liquidity premium of 0.23, inflation risk premium of -0.13 (so it is negative), and inflation of 1.95 (all values are in percent). However, if I'm doing the calculations correctly, the modeled TIPS yield comes to 0.14, while the 6/28 5-year TIPS yield from treasury.gov is 0.23. I can come back to the calculation of the modeled TIPS yield if anyone is interested.

The nominal raw 5-year yield in the spreadsheet is 1.77, so close to the treasury.gov value of 1.76 on 6/28, and the "fitted" nominal yield is 1.73.

The raw 5-year inflation compensation is 1.55, and the fitted value is 1.59. The raw value is close to the BEI rate of 1.53 calculated from the treasury.gov values for 6/28 (= 1.76 - 0.23).

Kevin
Last edited by Kevin M on Thu Jul 18, 2019 6:05 pm, edited 1 time in total.
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

Topic Author
grok87
Posts: 8772
Joined: Tue Feb 27, 2007 9:00 pm

Re: are TIPS expensive? or are they cheap?

Post by grok87 » Wed Jul 17, 2019 9:39 pm

Kevin M wrote:
Wed Jul 17, 2019 7:16 pm
Kevin M wrote:
Sat Jul 13, 2019 3:18 pm
We've seen references to models that estimate one premium or the other, but I don't recall seeing a reference to a model that estimates both premiums, including data as to what they currently are. That would be an extremely useful addition to this conversation.
I found something on the main Federal Reserve website, in the Economic Research -> FEDS notes section. Sorry if it's been cited already, but I don't recall seeing it. Here's the link to the article: Tips from TIPS: Update and Discussions. It's dated May 21, 2019, so fairly recent. From the introduction:
The spread between the yield on a nominal Treasury security and that on a Treasury inflation protected security (TIPS) of comparable maturities—usually called the "breakeven inflation rate" or "inflation compensation"—has been often used as a real-time proxy for market participants' inflation expectations. However, policymakers and market participants are also cognizant that this spread is an imperfect measure, as it contains other components that can drive a wedge between inflation compensation and market participants' true inflation expectations.

In a recently published paper (D'Amico, Kim, and Wei (DKW), 2018), D'Amico and two of us use a no-arbitrage term structure model to decompose TIPS inflation compensation into three components—inflation expectation, inflation risk premium, and TIPS liquidity premium—over the 1990-2013 period. The model is also used to decompose nominal yields or forward rates into four components—expected real short rate, expected inflation, inflation risk premium, and real term premium.
Note from the underlined part of the first paragraph that what we refer to in this thread as breakeven inflation rate is also referred to as inflation compensation. The latter term is what's mostly used in the rest of the article, as well as in some other papers I've read.

The underlined part of the second paragraph is just repeating something we've been discussing since close to the beginning of the thread.

The article is a good read, but what's really cool is this:
Beginning with the publication of this Note, we will provide monthly updates of model decompositions of the 5-year, 10-year, and 5-to-10-year nominal Treasury yields and inflation compensation up to the last business day of the previous month. In general, we will endeavor to have the updated series posted sometime after 10:00 a.m. on the fourth business day of each month. Because this is a staff research product and not an official statistical release, it is subject to delay, revision, or methodological changes without advance notice.17

The latest estimates will be posted as a comma-separated values (CSV) file at the URL https://www.federalreserve.gov/econres/ ... pdates.csv. As of this publication, daily data are available for the period from January 3, 1983 to June 28, 2019.
From the download, the 6/28/2019 modeled 5-year values are TIPS liquidity premium of 0.23, inflation risk premium of -0.13 (so it is negative), and inflation of 1.95 (all values are in percent). However, if I'm doing the calculations correctly, the modeled TIPS yield comes to 0.14, while the 6/28 5-year TIPS yield from treasury.gov is 0.23. I can come back to the calculation of the modeled TIPS yield if anyone is interested.

The nominal raw 5-year yield in the spreadsheet is 1.77, so close to the treasury.gov value of 1.76 on 6/28, and the "fitted" nominal yield is 1.73.

The raw 5-year inflation compensation is 1.55, and the fitted value is 1.59. The raw value is close to the BEI rate of 1.53 calculated from the treasury.gov values for 6/28 (= 1.76 - 0.23).

Kevin
thanks.
looks interesting. your link to the csv file didn't work for me though.
so if i'm understanding, at 6/28 they thought inflation would be 0.36% higher [= 0.23% - -0.13%] than
5 year BEI. or in my parlance "cheap by 36 bps"
RIP Mr. Bogle.

User avatar
Kevin M
Posts: 10921
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Kevin M » Thu Jul 18, 2019 6:28 pm

grok87 wrote:
Wed Jul 17, 2019 9:39 pm
thanks.
looks interesting. your link to the csv file didn't work for me though.
I had just copied and pasted the text from the article. Took me some experimenting, but I think I got the link working. If it doesn't work, just open the article, scroll to near the end, just before the references, and click the link to the csv file there.
grok87 wrote:
Wed Jul 17, 2019 9:39 pm
so if i'm understanding, at 6/28 they thought inflation would be 0.36% higher [= 0.23% - -0.13%] than
5 year BEI. or in my parlance "cheap by 36 bps"
Yes, that would be one way of describing some of the outputs of their model.

I find the negative inflation risk premium interesting, and it's easy to find articles about it. As I mentioned in an earlier post, apparently the aggregate Treasury market is more concerned about deflation than inflation, and some of what I've read seems to be saying that this is indeed the case.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

larryswedroe
Posts: 15955
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: are TIPS expensive? or are they cheap?

Post by larryswedroe » Thu Jul 18, 2019 7:03 pm

Have not read the article yet but if there is a finding of a negative inflation risk premium, that makes no sense (to be paid to avoid a risk rather than paying, there is no place that insurance has negative cost) and thus rather than assuming there really is a negative premium my assumption would be that something is wrong with the model/methodology. Perhaps the article will explain. But without reading it seems likely they are underestimating the liquidity premium being paid. Thanks for sharing the article
Larry

User avatar
Kevin M
Posts: 10921
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Kevin M » Thu Jul 18, 2019 7:39 pm

FEDS Notes, April 4, 2016: Has the inflation risk premium fallen? Is it now negative?. From the conclusion of this paper:
To sum up, this note points out that standard consumption-based asset pricing models and the capital asset pricing model suggest that the long run inflation risk premium has trended down over time, and is likely to be negative in the current macroeconomic environment. Moreover, a nontrivial portion of the decline in far-forward inflation compensation over the past year may reflect a decline in the inflation risk premium rather than a drop in investors' expected inflation rate.

A burgeoning academic literature has also investigated this issue, providing estimates of the inflation risk premium. For example, Chernov and Mueller (2012) argue that the inflation risk premium estimates are model-dependent and can switch sign from positive to negative in a model that accounts for survey-based inflation forecasts vs. the one that does exclude the survey forecasts from the estimation. Grishchenko and Huang (2013) find that the inflation risk premium implied by the nominal yields and the TIPS-based measure of inflation compensation was positive in the early 2000s but switched signs around the Great Recession. D'Amico, Kim, and Wei (2014) also find that the inflation risk premium has been trending down. Finally, recent papers in the macro-finance literature have made strides in incorporating the fundamental insights from consumption-based asset pricing into fully-fledged models of the term structure of interest rates.
And from the already cited in a previous post:
As a result, TIPS inflation compensation (IC)—defined as the nominal yield minus the TIPS yield—can be decomposed into three components,

TIPS IC = expected inflation + inflation risk premium – TIPS liquidity premium. (4)

TIPS IC can deviate from expected inflation for two reasons: a non-zero inflation risk premium or a non-zero TIPS liquidity premium. The first term, inflation risk premium, is the extra compensation nominal bond investors demand for bearing inflation risks, and its value depends on the covariance between inflation and real economic activity. This premium is believed to have been positive and sizeable in the 1970s and 1980s, when investors were more worried about stagflation scenarios with higher inflation accompanied by lower growth, but appears to have declined in recent decades to lower or even negative levels, as investors have become more concerned about outcomes where lower inflation is associated with lower growth.
(underlines mine, for emphasis)

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
CULater
Posts: 2099
Joined: Sun Nov 13, 2016 10:59 am

Re: are TIPS expensive? or are they cheap?

Post by CULater » Thu Jul 18, 2019 11:23 pm

Maybe I'm wrong here, but wouldn't it make more sense to buy short term TIPS rather than short term nominals right now? If the Fed is drops short term rates, the interest your short nominals will earn will shrink -- that's the risk of owning short duration nominal bonds. But short TIPs will still be getting the same inflation adjustment that all TIPs get, so the interest they return will not drop as much as long as the inflation rate doesn't drop.
On the internet, nobody knows you're a dog.

stlutz
Posts: 5223
Joined: Fri Jan 02, 2009 1:08 am

Re: are TIPS expensive? or are they cheap?

Post by stlutz » Thu Jul 18, 2019 11:33 pm

CULater wrote:
Thu Jul 18, 2019 11:23 pm
Maybe I'm wrong here, but wouldn't it make more sense to buy short term TIPS rather than short term nominals right now? If the Fed is drops short term rates, the interest your short nominals will earn will shrink -- that's the risk of owning short duration nominal bonds. But short TIPs will still be getting the same inflation adjustment that all TIPs get, so the interest they return will not drop as much as long as the inflation rate doesn't drop.
An assumed rate cut has no impact on on the nominals vs. TIPS consideration. The short end of the yield curve is inverted because everybody already "knows" that short-term Fed rate cuts are coming.

The choice between nominals and TIPS is best made on the basis of which investment makes the most sense for your own situation (i.e. which risks and rewards matter most to you).

Of course, the market price can be wrong. I've made bets one or the other way when I've though the relative pricing was way off. That was the case back in 2016 when the breakeven rate fell to 1.2% for 10 year notes. It's not really the case now (with BEI ~1.8%).

Northern Flicker
Posts: 4368
Joined: Fri Apr 10, 2015 12:29 am

Re: are TIPS expensive? or are they cheap?

Post by Northern Flicker » Fri Jul 19, 2019 3:19 am

Untangling this thread would have to start with a definition of the notion of expected inflation. Is it the expected value of the inflation rate over some period when that inflation rate is modeled as a random variable, or is it the inflation rate that people project over some time period. I would prefer the term projected inflation for the rate of inflation that is predicted to materialize over some time period.

It appears that recent posters mean projected inflation when they write expected inflation. Note that the authors of the referenced article at the Fed do not define their usage of expected inflation either.

But there are many authors in general who use the term expected inflation as a modeled component of treasury yields are referring to the mean or expected value of a random variable for future inflation.

Such sloppiness does not lend itself well to generating quality results.

And I would add that some of the articles about inflation risk premia do argue that it can be negative, despite how illogical such a state of affairs would be.

Topic Author
grok87
Posts: 8772
Joined: Tue Feb 27, 2007 9:00 pm

Re: are TIPS expensive? or are they cheap?

Post by grok87 » Fri Jul 19, 2019 6:15 am

CULater wrote:
Thu Jul 18, 2019 11:23 pm
Maybe I'm wrong here, but wouldn't it make more sense to buy short term TIPS rather than short term nominals right now? If the Fed is drops short term rates, the interest your short nominals will earn will shrink -- that's the risk of owning short duration nominal bonds. But short TIPs will still be getting the same inflation adjustment that all TIPs get, so the interest they return will not drop as much as long as the inflation rate doesn't drop.
I agree. According to vanguard the real yield of their short term tips fund is like 0.44% (admiral).

I think that’s attractive
RIP Mr. Bogle.

Topic Author
grok87
Posts: 8772
Joined: Tue Feb 27, 2007 9:00 pm

Re: are TIPS expensive? or are they cheap?

Post by grok87 » Fri Jul 19, 2019 7:04 am

Kevin M wrote:
Thu Jul 18, 2019 6:28 pm
grok87 wrote:
Wed Jul 17, 2019 9:39 pm
thanks.
looks interesting. your link to the csv file didn't work for me though.
I had just copied and pasted the text from the article. Took me some experimenting, but I think I got the link working. If it doesn't work, just open the article, scroll to near the end, just before the references, and click the link to the csv file there.
grok87 wrote:
Wed Jul 17, 2019 9:39 pm
so if i'm understanding, at 6/28 they thought inflation would be 0.36% higher [= 0.23% - -0.13%] than
5 year BEI. or in my parlance "cheap by 36 bps"
Yes, that would be one way of describing some of the outputs of their model.

I find the negative inflation risk premium interesting, and it's easy to find articles about it. As I mentioned in an earlier post, apparently the aggregate Treasury market is more concerned about deflation than inflation, and some of what I've read seems to be saying that this is indeed the case.

Kevin
Thanks
RIP Mr. Bogle.

larryswedroe
Posts: 15955
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: are TIPS expensive? or are they cheap?

Post by larryswedroe » Fri Jul 19, 2019 8:16 am

Kevin
Re declining risk premium for unexpected inflation---yes that almost certainly has happened as the literature shows that expected inflation moves in line with recent years inflation (investors extrapolate the past) and that has declined over last 10 years, so logical that risk premium has fallen, but it should logically never be negative. IMO it must be that the liquidity premium is larger than the model is showing (all models are wrong, or they would be laws, doesn't mean they aren't helpful in helping us understand how world works).
FWIW, my own view is that the larger liquidity premium (making TIPS look cheap) is due to what economists would call a preferred habitat---there are participants (in this case like global central banks) that value liquidity greatly and thus willing to pay up for. It is not an inefficiency. It's the same argument that could be made about lottery stocks poor returns being a result of a preferred habitat.
Best wishes
Larry

User avatar
CULater
Posts: 2099
Joined: Sun Nov 13, 2016 10:59 am

Re: are TIPS expensive? or are they cheap?

Post by CULater » Fri Jul 19, 2019 8:37 am

grok87 wrote:
Fri Jul 19, 2019 6:15 am
CULater wrote:
Thu Jul 18, 2019 11:23 pm
Maybe I'm wrong here, but wouldn't it make more sense to buy short term TIPS rather than short term nominals right now? If the Fed is drops short term rates, the interest your short nominals will earn will shrink -- that's the risk of owning short duration nominal bonds. But short TIPs will still be getting the same inflation adjustment that all TIPs get, so the interest they return will not drop as much as long as the inflation rate doesn't drop.
I agree. According to vanguard the real yield of their short term tips fund is like 0.44% (admiral).

I think that’s attractive
So, my next question is this: the real yield for TIPS all the way out to 10 years looks about the same (it's about .25). If that's the case and I want to hold a TIPS fund, is there any reason to NOT prefer a short term TIPS fund over an intermediate term, or perhaps even a long term fund? I get the same real yield with lower term risk if real yields spike. Why the heck would I want 5 or 10-year TIPS right now over short TIPS such as 2-year?
On the internet, nobody knows you're a dog.

cheezit
Posts: 221
Joined: Sat Jul 14, 2018 7:28 pm

Re: are TIPS expensive? or are they cheap?

Post by cheezit » Fri Jul 19, 2019 8:49 am

larryswedroe wrote:
Fri Jul 19, 2019 8:16 am
[It's ]logical that risk premium has fallen, but it should logically never be negative.
Could you elaborate on this? Certainly the observed equity risk premium has sometimes been negative for long periods in a number of markets, and I don't see why circumstances (eg. extreme valuations) could not make the expected forward ERP for a particular time period negative, so I don't understand why any of the risk premia associated with TIPS could not be negative for a particular forward time period either.

JackoC
Posts: 872
Joined: Sun Aug 12, 2018 11:14 am

Re: are TIPS expensive? or are they cheap?

Post by JackoC » Fri Jul 19, 2019 10:10 am

larryswedroe wrote:
Fri Jul 19, 2019 8:16 am
Kevin
1. Re declining risk premium for unexpected inflation---yes that almost certainly has happened as the literature shows that expected inflation moves in line with recent years inflation (investors extrapolate the past) and that has declined over last 10 years, so logical that risk premium has fallen, but it should logically never be negative.

2. IMO it must be that the liquidity premium is larger than the model is showing (all models are wrong, or they would be laws, doesn't mean they aren't helpful in helping us understand how world works).
1. The links Kevin recently gave are an excellent addition to the ongoing discussion here about TIPS. The summary piece (link repeated below) specifically theorizes how the inflation risk premium could be negative, rather than it just being an output of modeling. It's not illogical necessarily. They simply apply the idea that a risk premium's sign follows the correlation between return and consumption (saying income I think makes it clearer). The sign of the equity risk premium is positive because investor income/consumption drop and stock market drops are highly correlated. It was also positive for inflation because investors focused on the outcome of high inflation choking off real growth, stagflation. In those cases nominal treasuries had real returns disastrously more negative than their initial yield minus initial inflation expectation whereas TIPS would have had (if they'd existed) real return equal to initial real yield (buy and hold). Now if investors mainly focus their risk anxiety on cases where lower inflation (or deflation) correlates with lower than expected income, nominals are the better hedge since their real return will increase more relative to initial yield than TIPS will* in such cases.
https://www.federalreserve.gov/econresd ... 60404.html

2. Although that's also possible, of course. As the same reference notes, a difference between survey and model based inflation estimates could flip the sign of the apparent inflation risk premium. For example now or recently if you take the Philly Fed survey based inflation estimate of 2.20% in 10yrs, the inflation risk premium output is obviously going to tend to be lower than if your starting point is the Cleveland Fed's model based market inflation expectation of 1.67%, v a 10 yr TIPS break even right now (none of the three numbers are exactly synchronized) around 1.75%. If the inflation expectation is really 2.20% it seems likely the inflation risk premium actually is negative (ie investor gets paid to take the risk of inferior TIPS performance in low inflation/deflation as well as being paid for the lower liquidity of TIPS). If you believe 1.67% as the inflation expectation, it's much more plausible to posit that the liquidity and inflation risk premia are modest, approximately offsetting values.
https://www.clevelandfed.org/our-resear ... tions.aspx

One other note on CD's. Best direct 5yr CD's now have much higher expected return than TIPS, higher expected pretax return up to inflation around 3.1% for 5 yr CD 3.35%. There are other miscellaneous factors to consider but they are highly unlikely to overcome a yield gap as large as now's, much larger than usual for run-of-the-mill best offerings on depositaccounts.com etc. The CD is taxed at state level but that's maybe 20bps at typical state tax rates. The put option you own on the CD is potentially worth more than that. If the issuing bank failed that's still < default probability times the difference in rates (since your rate would not necessarily be voided by a successor institution), which would amount to expected reduction of just a few bps for typical issuers. If you can tie up money for 5 yrs, and unless in such a high tax bracket that muni bond after tax yield has a compelling advantage correcting for credit risk, it's hard IMO to argue against best direct CD's right now as a safe asset.

*TIPS real return will equal initial real yield down to cumulative 0% inflation over the life of the issue, then rise in cumulative deflation because the principal redemption is floored in nominal $'s at the initial face amount. But eg. at cumulative -1% CPI inflation, a new 5yr TIPS issued today would return 1.25% real, a new 5 yr note would return 2.8% real. A seasoned TIPS issue with 5yrs to go, index factor already >1, would have lower real return than a new one in that case assuming similar real yield now.

Topic Author
grok87
Posts: 8772
Joined: Tue Feb 27, 2007 9:00 pm

Re: are TIPS expensive? or are they cheap?

Post by grok87 » Fri Jul 19, 2019 11:44 am

CULater wrote:
Fri Jul 19, 2019 8:37 am
grok87 wrote:
Fri Jul 19, 2019 6:15 am
CULater wrote:
Thu Jul 18, 2019 11:23 pm
Maybe I'm wrong here, but wouldn't it make more sense to buy short term TIPS rather than short term nominals right now? If the Fed is drops short term rates, the interest your short nominals will earn will shrink -- that's the risk of owning short duration nominal bonds. But short TIPs will still be getting the same inflation adjustment that all TIPs get, so the interest they return will not drop as much as long as the inflation rate doesn't drop.
I agree. According to vanguard the real yield of their short term tips fund is like 0.44% (admiral).

I think that’s attractive
So, my next question is this: the real yield for TIPS all the way out to 10 years looks about the same (it's about .25). If that's the case and I want to hold a TIPS fund, is there any reason to NOT prefer a short term TIPS fund over an intermediate term, or perhaps even a long term fund? I get the same real yield with lower term risk if real yields spike. Why the heck would I want 5 or 10-year TIPS right now over short TIPS such as 2-year?
On the run 5 and 10 year tips offer more deflation prtotection.
RIP Mr. Bogle.

Hector
Posts: 989
Joined: Fri Dec 24, 2010 2:21 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Hector » Fri Jul 19, 2019 12:57 pm

stlutz wrote:
Thu Jul 18, 2019 11:33 pm
CULater wrote:
Thu Jul 18, 2019 11:23 pm
Maybe I'm wrong here, but wouldn't it make more sense to buy short term TIPS rather than short term nominals right now? If the Fed is drops short term rates, the interest your short nominals will earn will shrink -- that's the risk of owning short duration nominal bonds. But short TIPs will still be getting the same inflation adjustment that all TIPs get, so the interest they return will not drop as much as long as the inflation rate doesn't drop.
An assumed rate cut has no impact on on the nominals vs. TIPS consideration. The short end of the yield curve is inverted because everybody already "knows" that short-term Fed rate cuts are coming.
If Fed rate cuts happen, do they impact only short term rate? If cut is 0.25%, what happens to today's rate?

Code: Select all

Date		1 Mo	2 Mo	3 Mo	6 Mo	1 Yr	2 Yr	3 Yr	5 Yr	7 Yr	10 Yr	20 Yr	30 Yr
07/18/19	2.11	2.12	2.05	2.01	1.90	1.77	1.74	1.78	1.89	2.04	2.34	2.5

Northern Flicker
Posts: 4368
Joined: Fri Apr 10, 2015 12:29 am

Re: are TIPS expensive? or are they cheap?

Post by Northern Flicker » Fri Jul 19, 2019 1:03 pm

The usage of the word premium is also overloaded in the above discussion and must be disambiguated. A risk premium is compensation you receive for taking a risk. An insurance premium is a premium you pay to transfer risk to someone else.

stlutz
Posts: 5223
Joined: Fri Jan 02, 2009 1:08 am

Re: are TIPS expensive? or are they cheap?

Post by stlutz » Fri Jul 19, 2019 3:24 pm

Hector wrote:
Fri Jul 19, 2019 12:57 pm

If Fed rate cuts happen, do they impact only short term rate? If cut is 0.25%, what happens to today's rate?
The 1 month Treasury was 2.51% back in March; its 2.11% now. There has not been a Fed rate cut in that time. Even at the short end, the market tends to run ahead of actual changes in the Fed Funds rate (which the rate banks charge each other to borrow overnight).

Hector
Posts: 989
Joined: Fri Dec 24, 2010 2:21 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Hector » Fri Jul 19, 2019 4:01 pm

stlutz wrote:
Fri Jul 19, 2019 3:24 pm
An assumed rate cut has no impact on on the nominals vs. TIPS consideration. The short end of the yield curve is inverted because everybody already "knows" that short-term Fed rate cuts are coming.
Hector wrote:
Fri Jul 19, 2019 12:57 pm

If Fed rate cuts happen, do they impact only short term rate? If cut is 0.25%, what happens to today's rate?
The 1 month Treasury was 2.51% back in March; its 2.11% now. There has not been a Fed rate cut in that time. Even at the short end, the market tends to run ahead of actual changes in the Fed Funds rate (which the rate banks charge each other to borrow overnight).
I meant to ask, does Fed rate change (or anticipation of it) effects only short term duration bond? Looks like change happened for all durations; from 1 month to 30 year treasury from March to today.

Code: Select all

Date		1 Mo	2 Mo	3 Mo	6 Mo	1 Yr	2 Yr	3 Yr	5 Yr	7 Yr	10 Yr	20 Yr	30 Yr
03/21/19	2.51	2.47	2.49	2.50	2.48	2.41	2.34	2.34	2.44	2.54	2.78	2.96
07/19/19	2.11	2.16	2.06	2.03	1.94	1.80	1.77	1.80	1.91	2.05	2.35	2.57
Change% 	15.94	12.55	17.27	18.80	21.77	25.31	24.36	23.08	21.72	19.29	15.47	13.18

JackoC
Posts: 872
Joined: Sun Aug 12, 2018 11:14 am

Re: are TIPS expensive? or are they cheap?

Post by JackoC » Fri Jul 19, 2019 4:31 pm

Northern Flicker wrote:
Fri Jul 19, 2019 1:03 pm
The usage of the word premium is also overloaded in the above discussion and must be disambiguated. A risk premium is compensation you receive for taking a risk. An insurance premium is a premium you pay to transfer risk to someone else.
I think the discussion so far is actually in line with academic and professional market terminology, where the term 'risk premium' does not imply a sign. Signs just need to be kept straight according to an agreed convention.

A positive inflation risk premium makes sense under the assumption that investors are generally harmed by high inflation, because it causes lower real growth and that manifests as lower income personally and/or lower real returns on risk assets like stocks: stagflation. So, investors will accept a lower expected return on an instrument that doesn't add to that problem when it arises: TIPS, which give ~constant return in higher inflation v nominals, which give reduced or negative real return in higher inflation. Just by convention, that's been defined as 'positive inflation risk premium'.

But it might be that investors now believe higher than expected inflation (especially in only the next 5 yrs say) to probably accompany a lessening of the global growth funk (extraordinary monetary and/or fiscal stimulus in many places yielding growth that still would have been deemed poor to only OK in the past), improving personal real income and returns of investments tied directly to real growth. So they don't have as much need for another instrument that does relatively better when inflation is higher. Whereas OTOH if the global growth funk intensifies it will probably be accompanied by even lower inflation, or deflation. So they put more value on an 'inflation hedge' which increases in real return when inflation decreases or goes negative. Nominal treasuries do that better than TIPS. Hence the inflation risk premium may turn negative: investors gain expected return in TIPS for giving up the superior characteristics of nominal treasuries in lower inflation/deflation, that being the most worrisome case due to expected correlation with bad economic outcomes generally.

The term 'insurance premium' implies it has to be in one direction, so I think is actually more confusing than speaking of a risk premium which can change sign. And there are other cases of risk premia which can change sign, the term premium on the yield curve for example.

Topic Author
grok87
Posts: 8772
Joined: Tue Feb 27, 2007 9:00 pm

Re: are TIPS expensive? or are they cheap?

Post by grok87 » Fri Jul 19, 2019 6:22 pm

JackoC wrote:
Fri Jul 19, 2019 4:31 pm
Northern Flicker wrote:
Fri Jul 19, 2019 1:03 pm
The usage of the word premium is also overloaded in the above discussion and must be disambiguated. A risk premium is compensation you receive for taking a risk. An insurance premium is a premium you pay to transfer risk to someone else.
I think the discussion so far is actually in line with academic and professional market terminology, where the term 'risk premium' does not imply a sign. Signs just need to be kept straight according to an agreed convention.

A positive inflation risk premium makes sense under the assumption that investors are generally harmed by high inflation, because it causes lower real growth and that manifests as lower income personally and/or lower real returns on risk assets like stocks: stagflation. So, investors will accept a lower expected return on an instrument that doesn't add to that problem when it arises: TIPS, which give ~constant return in higher inflation v nominals, which give reduced or negative real return in higher inflation. Just by convention, that's been defined as 'positive inflation risk premium'.

But it might be that investors now believe higher than expected inflation (especially in only the next 5 yrs say) to probably accompany a lessening of the global growth funk (extraordinary monetary and/or fiscal stimulus in many places yielding growth that still would have been deemed poor to only OK in the past), improving personal real income and returns of investments tied directly to real growth. So they don't have as much need for another instrument that does relatively better when inflation is higher. Whereas OTOH if the global growth funk intensifies it will probably be accompanied by even lower inflation, or deflation. So they put more value on an 'inflation hedge' which increases in real return when inflation decreases or goes negative. Nominal treasuries do that better than TIPS. Hence the inflation risk premium may turn negative: investors gain expected return in TIPS for giving up the superior characteristics of nominal treasuries in lower inflation/deflation, that being the most worrisome case due to expected correlation with bad economic outcomes generally.

The term 'insurance premium' implies it has to be in one direction, so I think is actually more confusing than speaking of a risk premium which can change sign. And there are other cases of risk premia which can change sign, the term premium on the yield curve for example.
good observations. i pretty much agree.
JackoC wrote: So they put more value on an 'inflation hedge' which increases in real return when inflation decreases or goes negative. Nominal treasuries do that better than TIPS. Hence the inflation risk premium may turn negative: investors gain expected return in TIPS for giving up the superior characteristics of nominal treasuries in lower inflation/deflation, that being the most worrisome case due to expected correlation with bad economic outcomes generally.
i think another way to say this is TIPS have "beta". i.e. they are more correlated with equities than nominal treasuries for the reasons you mention. and that's one way to think about why investors might expect an excess return from tips relative to nominal treasuries- i.e. why they are cheap.
Last edited by grok87 on Sat Jul 20, 2019 8:44 am, edited 1 time in total.
RIP Mr. Bogle.

User avatar
Kevin M
Posts: 10921
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Kevin M » Fri Jul 19, 2019 7:44 pm

Happy to see the additional contributions to this discussion. Good stuff.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

Northern Flicker
Posts: 4368
Joined: Fri Apr 10, 2015 12:29 am

Re: are TIPS expensive? or are they cheap?

Post by Northern Flicker » Tue Jul 23, 2019 1:39 am

The term 'insurance premium' implies it has to be in one direction, so I think is actually more confusing than speaking of a risk premium which can change sign. And there are other cases of risk premia which can change sign, the term premium on the yield curve for example.
I agree re: insurance premium but that was how I interpreted the usage of premium as used in some of the discussion.

User avatar
Kevin M
Posts: 10921
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Kevin M » Tue Jul 23, 2019 2:20 pm

Grok's argument, made in the OP, is all about the breakeven inflation rate (BEI) being less than his estimate of expected inflation over the relevant period. This can be reduced aglebraically to saying that the liquidity risk premium embedded in TIPS yield minus the unexpected inflation risk premium embedded in nominal treasury yield is a positive number, assuming that these two risk premiums are all that are causing BEI to be different than expected inflation.

BEI = Expected inflation + nominal unexpected inflation risk premium - TIPS liquidity risk premium

BEI = EI + IRP - LRP

BEI - EI = IRP - LRP

EI - BEI = LRP - IRP

I thought it would be interesting to look at LRP - IRP using the data of the model I shared upthread. First, here's the chart of monthly LRP and IRP for 5-year term to maturity since 1/4/1999, the date when the model first provides the TIPS liquidity risk premiums:

Image

I truncated the top of the chart so there's more resolution for most of the time period--the maximum for LRP was over 4 percentage points.

Something that jumps out is the much higher volatility, and generally larger magnitude, of the LRP compared to the IRP.

Here's the chart showing LRP minus IRP for the 5-year maturity:

Image

The main conclusion I draw is that if one defines TIPS as cheap when LRP > IRP, then TIPS almost always have been cheap!

However, if one compares the recent value of LRP minus IRP to historical values, TIPS are not particularly cheap using the 6/28/2019 values from the model. The historical average of LRP minus IRP is 0.53 percentage points, while the 6/28 value is 0.36 percentage points, so the 6/28 "TIPS premium" of 36 basis points is 17 basis points below the historical average, or about 1/3 less than the historical average. So relative to historical values, TIPS are a bit expensive, and perhaps will soon be fairly valued if the recent uptrend in LRP minus IRP continues.

Note that Grok's definition of "cheap" does not consider one's personal exposure to unexpected inflation risk. Considering the latter, one might consider TIPS cheap even if expected inflation were less than BEI, in which case TIPS would have been cheap even more often than indicated by the charts above, but that's not the definition Grok chose.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

Topic Author
grok87
Posts: 8772
Joined: Tue Feb 27, 2007 9:00 pm

Re: are TIPS expensive? or are they cheap?

Post by grok87 » Wed Jul 24, 2019 10:53 am

Kevin M wrote:
Tue Jul 23, 2019 2:20 pm
Grok's argument, made in the OP, is all about the breakeven inflation rate (BEI) being less than his estimate of expected inflation over the relevant period. This can be reduced aglebraically to saying that the liquidity risk premium embedded in TIPS yield minus the unexpected inflation risk premium embedded in nominal treasury yield is a positive number, assuming that these two risk premiums are all that are causing BEI to be different than expected inflation.

BEI = Expected inflation + nominal unexpected inflation risk premium - TIPS liquidity risk premium

BEI = EI + IRP - LRP

BEI - EI = IRP - LRP

EI - BEI = LRP - IRP

I thought it would be interesting to look at LRP - IRP using the data of the model I shared upthread. First, here's the chart of monthly LRP and IRP for 5-year term to maturity since 1/4/1999, the date when the model first provides the TIPS liquidity risk premiums:

Image

I truncated the top of the chart so there's more resolution for most of the time period--the maximum for LRP was over 4 percentage points.

Something that jumps out is the much higher volatility, and generally larger magnitude, of the LRP compared to the IRP.

Here's the chart showing LRP minus IRP for the 5-year maturity:

Image

The main conclusion I draw is that if one defines TIPS as cheap when LRP > IRP, then TIPS almost always have been cheap!

However, if one compares the recent value of LRP minus IRP to historical values, TIPS are not particularly cheap using the 6/28/2019 values from the model. The historical average of LRP minus IRP is 0.53 percentage points, while the 6/28 value is 0.36 percentage points, so the 6/28 "TIPS premium" of 36 basis points is 17 basis points below the historical average, or about 1/3 less than the historical average. So relative to historical values, TIPS are a bit expensive, and perhaps will soon be fairly valued if the recent uptrend in LRP minus IRP continues.

Note that Grok's definition of "cheap" does not consider one's personal exposure to unexpected inflation risk. Considering the latter, one might consider TIPS cheap even if expected inflation were less than BEI, in which case TIPS would have been cheap even more often than indicated by the charts above, but that's not the definition Grok chose.

Kevin
Thanks Kevin.
Would be interesting to see your chart with stock and bond markets overlaid on it.
RIP Mr. Bogle.

Northern Flicker
Posts: 4368
Joined: Fri Apr 10, 2015 12:29 am

Re: are TIPS expensive? or are they cheap?

Post by Northern Flicker » Sun Jul 28, 2019 1:27 am

My point was just that a post such as kevinm’s and others in the thread with that much rigor really should define terms like “expected inflation”. Because I understood the point kevinm was making a priori, I was able to reverse engineer the posting to understand that he is referring to the inflation rate projected by the market in aggregate, and not expected inflation in the probabilistic sense (expected value of inflation as a random variable.

international001
Posts: 941
Joined: Thu Feb 15, 2018 7:31 pm

Re: are TIPS expensive? or are they cheap?

Post by international001 » Sun Jul 28, 2019 4:51 pm

grok87 wrote:
Fri Jul 19, 2019 6:22 pm

i think another way to say this is TIPS have "beta". i.e. they are more correlated with equities than nominal treasuries for the reasons you mention. and that's one way to think about why investors might expect an excess return from tips relative to nominal treasuries- i.e. why they are cheap.
I am an unsophisticated investor (relatively to the knowledgeable folks posting here)
But I chose treasuries for that reason, the same reason that I don't use total bond funds or high yield bond fund (that have higher sharpe). When I combine them with stocks, they offer me a better historical return. I thought price of an asset did not depend only on its returns and risk, but how well/bad combines with other assets.

After all these interesting discussions, I'd like to see some post recommending which bonds to use if you hold them along with stocks.

TIA

User avatar
Kevin M
Posts: 10921
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Kevin M » Sun Jul 28, 2019 7:40 pm

Northern Flicker wrote:
Sun Jul 28, 2019 1:27 am
My point was just that a post such as kevinm’s and others in the thread with that much rigor really should define terms like “expected inflation”. Because I understood the point kevinm was making a priori, I was able to reverse engineer the posting to understand that he is referring to the inflation rate projected by the market in aggregate, and not expected inflation in the probabilistic sense (expected value of inflation as a random variable.
Are they necessarily that different? Wouldn't we expect that the aggregate market's "projected" inflation rate is the mean of the aggregate market's estimate of the probability distribution of possible inflation rates over the relevant period? Ditto for surveys of economists projections or expectations? Ditto for the "expected inflation" output of one of the models we've discussed?

How is this any different than the widespread use of "expected return", which technically is the mean of a probability distribution of possible future returns, but in reality many estimates of expected return probably are not derived this way.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

User avatar
Kevin M
Posts: 10921
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: are TIPS expensive? or are they cheap?

Post by Kevin M » Sun Jul 28, 2019 7:46 pm

international001 wrote:
Sun Jul 28, 2019 4:51 pm
grok87 wrote:
Fri Jul 19, 2019 6:22 pm
i think another way to say this is TIPS have "beta". i.e. they are more correlated with equities than nominal treasuries for the reasons you mention. and that's one way to think about why investors might expect an excess return from tips relative to nominal treasuries- i.e. why they are cheap.
I am an unsophisticated investor (relatively to the knowledgeable folks posting here)
But I chose treasuries for that reason, the same reason that I don't use total bond funds or high yield bond fund (that have higher sharpe). When I combine them with stocks, they offer me a better historical return. I thought price of an asset did not depend only on its returns and risk, but how well/bad combines with other assets.

After all these interesting discussions, I'd like to see some post recommending which bonds to use if you hold them along with stocks.

TIA
Late 2008 is the most obvious example demonstrating that nominal Treasuries, not TIPS, are more likely to give you the negative correlation with stocks when you need it most. TIPS prices actually went down, along with stocks, corporate bonds, and just about everything else in late 2008, but nominal Treasury prices went up.

That doesn't mean that one can't also use TIPS, specifically for their inflation-protection feature, along with stocks and nominal bonds. It just means that TIPS probably aren't the best choice if you're looking for more likely negative correlation with stocks, especially during a financial crisis.

Note that Vanguard holds nominal bonds in its LifeStrategy baanced funds and all but the shortest-dated Target Retirement funds, but they add some TIPS as target retirement funds near their target date.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

Topic Author
grok87
Posts: 8772
Joined: Tue Feb 27, 2007 9:00 pm

Re: are TIPS expensive? or are they cheap?

Post by grok87 » Mon Jul 29, 2019 6:29 am

international001 wrote:
Sun Jul 28, 2019 4:51 pm
grok87 wrote:
Fri Jul 19, 2019 6:22 pm

i think another way to say this is TIPS have "beta". i.e. they are more correlated with equities than nominal treasuries for the reasons you mention. and that's one way to think about why investors might expect an excess return from tips relative to nominal treasuries- i.e. why they are cheap.
I am an unsophisticated investor (relatively to the knowledgeable folks posting here)
But I chose treasuries for that reason, the same reason that I don't use total bond funds or high yield bond fund (that have higher sharpe). When I combine them with stocks, they offer me a better historical return. I thought price of an asset did not depend only on its returns and risk, but how well/bad combines with other assets.

After all these interesting discussions, I'd like to see some post recommending which bonds to use if you hold them along with stocks.

TIA
thanks TIA. i follow David Swensen, who as you may know recommends putting half of one's bonds in treasuries and half in tips as part of a portfolio that includes stocks, real estate and bonds.

i'm not sure if David Swensen would say it this way. But my own view is that over the long run we might expect both treasuries and tips to be uncorrelated with stocks and real estate. but in specific time periods there may be some correlations. for example:

a) during the stagflation of the 70s/early 80s both stocks and nominal treasuries performed poorly.
b) during the Global Financial crisis (2008): both stocks and TIPs performed poorly (fear of deflation)

i think we are back to fear of deflation which may be why tips have undeperformed nominal treasuries recently.

cheers,
grok
RIP Mr. Bogle.

Valuethinker
Posts: 38387
Joined: Fri May 11, 2007 11:07 am

Re: are TIPS expensive? or are they cheap?

Post by Valuethinker » Mon Jul 29, 2019 7:30 am

grok87 wrote:
Mon Jul 29, 2019 6:29 am
international001 wrote:
Sun Jul 28, 2019 4:51 pm
grok87 wrote:
Fri Jul 19, 2019 6:22 pm

i think another way to say this is TIPS have "beta". i.e. they are more correlated with equities than nominal treasuries for the reasons you mention. and that's one way to think about why investors might expect an excess return from tips relative to nominal treasuries- i.e. why they are cheap.
I am an unsophisticated investor (relatively to the knowledgeable folks posting here)
But I chose treasuries for that reason, the same reason that I don't use total bond funds or high yield bond fund (that have higher sharpe). When I combine them with stocks, they offer me a better historical return. I thought price of an asset did not depend only on its returns and risk, but how well/bad combines with other assets.

After all these interesting discussions, I'd like to see some post recommending which bonds to use if you hold them along with stocks.

TIA
thanks TIA. i follow David Swensen, who as you may know recommends putting half of one's bonds in treasuries and half in tips as part of a portfolio that includes stocks, real estate and bonds.

i'm not sure if David Swensen would say it this way. But my own view is that over the long run we might expect both treasuries and tips to be uncorrelated with stocks and real estate. but in specific time periods there may be some correlations. for example:

a) during the stagflation of the 70s/early 80s both stocks and nominal treasuries performed poorly.
b) during the Global Financial crisis (2008): both stocks and TIPs performed poorly (fear of deflation)

i think we are back to fear of deflation which may be why tips have undeperformed nominal treasuries recently.

cheers,
grok
I don't think the TIPS moves in Q4 2008 were about fear of deflation. At least I think there is at least a partial explanation which is not about market expectations.

I think rather it was technical. TIPS were used as collateral for a number of REPO transactions (Repurchase Obligation - short term borrowing between financial institutions, collateralized by securities). When Lehman Brothers went down, a lot of counterparties all over the market had to quickly sell collateral to cover. TIPS being thinly traded, the prices fell and the implied inflation fell with it (real interest rates rose).

It's unclear whether this would be repeated in another financial crisis. TIPS bonds might be much more likely to move in line with US Treasury bonds, which is what they normally do?

Topic Author
grok87
Posts: 8772
Joined: Tue Feb 27, 2007 9:00 pm

Re: are TIPS expensive? or are they cheap?

Post by grok87 » Mon Jul 29, 2019 9:51 am

Valuethinker wrote:
Mon Jul 29, 2019 7:30 am
grok87 wrote:
Mon Jul 29, 2019 6:29 am
international001 wrote:
Sun Jul 28, 2019 4:51 pm
grok87 wrote:
Fri Jul 19, 2019 6:22 pm

i think another way to say this is TIPS have "beta". i.e. they are more correlated with equities than nominal treasuries for the reasons you mention. and that's one way to think about why investors might expect an excess return from tips relative to nominal treasuries- i.e. why they are cheap.
I am an unsophisticated investor (relatively to the knowledgeable folks posting here)
But I chose treasuries for that reason, the same reason that I don't use total bond funds or high yield bond fund (that have higher sharpe). When I combine them with stocks, they offer me a better historical return. I thought price of an asset did not depend only on its returns and risk, but how well/bad combines with other assets.

After all these interesting discussions, I'd like to see some post recommending which bonds to use if you hold them along with stocks.

TIA
thanks TIA. i follow David Swensen, who as you may know recommends putting half of one's bonds in treasuries and half in tips as part of a portfolio that includes stocks, real estate and bonds.

i'm not sure if David Swensen would say it this way. But my own view is that over the long run we might expect both treasuries and tips to be uncorrelated with stocks and real estate. but in specific time periods there may be some correlations. for example:

a) during the stagflation of the 70s/early 80s both stocks and nominal treasuries performed poorly.
b) during the Global Financial crisis (2008): both stocks and TIPs performed poorly (fear of deflation)

i think we are back to fear of deflation which may be why tips have undeperformed nominal treasuries recently.

cheers,
grok
I don't think the TIPS moves in Q4 2008 were about fear of deflation. At least I think there is at least a partial explanation which is not about market expectations.

I think rather it was technical. TIPS were used as collateral for a number of REPO transactions (Repurchase Obligation - short term borrowing between financial institutions, collateralized by securities). When Lehman Brothers went down, a lot of counterparties all over the market had to quickly sell collateral to cover. TIPS being thinly traded, the prices fell and the implied inflation fell with it (real interest rates rose).

It's unclear whether this would be repeated in another financial crisis. TIPS bonds might be much more likely to move in line with US Treasury bonds, which is what they normally do?
Thanks Valuethinker,
i'm sure your explanation is more standard and is part of the story as well. Personally i have never found the liquidity argument all that convincing. it would be interesting to dig into further. for what it's worth here is the history of 10 year BEI through the period
https://fred.stlouisfed.org/series/T10YIE
cheers,
grok
RIP Mr. Bogle.

Northern Flicker
Posts: 4368
Joined: Fri Apr 10, 2015 12:29 am

Re: are TIPS expensive? or are they cheap?

Post by Northern Flicker » Tue Jul 30, 2019 8:14 pm

Are they necessarily that different? Wouldn't we expect that the aggregate market's "projected" inflation rate is the mean of the aggregate market's estimate of the probability distribution of possible inflation rates over the relevant period?
I won’t try to claim to know what market participants use for their estimates but there is no a priori reason to assume it is the expected value of the probability distribution of future inflation. I think the estimates are more likely to be a projection of what they believe is the most likely outcome, which is different from a weighted average of all possible outcomes, weighted by their probabilities of occurrence.

bagle
Posts: 61
Joined: Tue Feb 22, 2011 5:59 am

Re: are TIPS expensive? or are they cheap?

Post by bagle » Wed Jul 31, 2019 9:51 am

Rather than viewing TIPS as cheap or expensive relative to nominal US Treasuries (e.g., BEI vs. consensus inflation estimates) or relative to financial crisis levels (e.g. Lehman bankruptcy illiquidity), let's simply look at the constant maturity real 10-year TIPS yield since, say, 2014. This looks be around 50 bp, somewhat higher than the current 29 bp real yield.

On this basis, assuming reversion to the mean and ongoing "secular stagnation", TIPS look to be a bit expensive.

https://fred.stlouisfed.org/series/DFII10

Topic Author
grok87
Posts: 8772
Joined: Tue Feb 27, 2007 9:00 pm

Re: are TIPS expensive? or are they cheap?

Post by grok87 » Wed Jul 31, 2019 1:57 pm

bagle wrote:
Wed Jul 31, 2019 9:51 am
Rather than viewing TIPS as cheap or expensive relative to nominal US Treasuries (e.g., BEI vs. consensus inflation estimates) or relative to financial crisis levels (e.g. Lehman bankruptcy illiquidity), let's simply look at the constant maturity real 10-year TIPS yield since, say, 2014. This looks be around 50 bp, somewhat higher than the current 29 bp real yield.

On this basis, assuming reversion to the mean and ongoing "secular stagnation", TIPS look to be a bit expensive.

https://fred.stlouisfed.org/series/DFII10
it's an interesting way to look at it. the 10 year CMT tips yield has been in a 0%-1% range since 2014. Before that period it ranged from 3.1% to -0.81%.
I think the theory as to why inflation might be mean-reverting is that the fed has a mandate to target stable prices which they have interpreted to mean 2% inflation.

I'm not aware of any real reason why real interest rates should be mean-reverting.
1) I think Bill Bernstein argues in one of his books that there has been a long term trend toward lower real interest rates.
2) I've also seen some arguments that the level of real interest rates should be in some relationship to the level of real GDP.
RIP Mr. Bogle.

Post Reply