Short-term breakeven inflation

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Topic Author
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Short-term breakeven inflation

Post by Kevin M »

The breakeven inflation (BEI) rate is the difference between the yield of a (nominal) Treasury and a Treasury Inflation Protected Security (TIPS) of the same maturity. These yields can be obtained for constant maturity periods of 5, 7, 10, 20 and 30 years from the Daily Treasury Yield Curve Rates and Daily Treasury Real Yield Curve Rates.

For example, for June 11, the Treasury 5-year yield is 0.76%, and the TIPS 5-year yield is -1.67%, from which we can calculate the 5-year BEI rate of 2.43% (= 0.76% - (-1.67%)). This is widely viewed as the Treasury market's expectation for the average inflation rate over the next five years.

From the treasury.gov yields, we can construct a chart of BEI rates for the maturities provided:

Image

Since the Treasury does not provide TIPS constant maturity Treasury (CMT) yields for less than 5-year maturity, other means must be used to determine BEI rates for shorter maturities. One way to do this is to obtain shorter-term Treasury and TIPS yields from a broker, and calculate the BEI rates for shorter terms. I do this using ask yields from Fidelity.

For example, for June 9, the ask yield for the 5.1-year Treasury was 0.76%, and the 5.1-year TIPS was 1.76% -1.76%, giving a 5.1-year BEI rate of 2.52%, so pretty close to the 5-year BEI from the treasury.gov site.

For shorter-term TIPS, there is a seasonality factor that results in a sawtooth pattern for the yields. This sawtooth yield curve can be smoothed by applying a seasonal adjustment factor to the yields. The chart below shows the unadjusted and seasonally-adjusted BEI rates for maturities of five years and less:

Image

I've truncated the y-axis to exclude the BEI rate (of 7.58%) for the TIPS maturing on 7/15/2021, since the real yields for these very short-term TIPS are not meaningful, as the inflation factor at maturity already is know, and these TIPS are priced to be competitive with nominal Treasuries of the same maturity.

From this chart, we can see that shorter-term inflation expectations are higher, decreasing as maturity increases.

To get some skin in the game, and to see how things work out over the next couple of years, I recently bought ten 2-year nominal Treasuries and ten 2-year TIPS. If inflation over the next two years is close to the 2-year BEI of about 2.8%, then each of these should provide about the same total return. If inflation is higher than expected, the TIPS should provide a higher total return than the nominal Treasury, and vice versa.

(Thanks to forum member Angst for catching a typo, corrected above).

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
Nate79
Posts: 9354
Joined: Thu Aug 11, 2016 6:24 pm
Location: Delaware

Re: Short-term breakeven inflation

Post by Nate79 »

Historically how well has the market predicted the future inflation rates?
User avatar
Topic Author
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Short-term breakeven inflation

Post by Kevin M »

Nate79 wrote: Sat Jun 12, 2021 8:04 pm Historically how well has the market predicted the future inflation rates?
Good question, but not the one I am answering, which is simply how to determine breakeven inflation rates for terms shorter than five years.

Answering your question requires a separate analysis, which might be fun to do, and I'm sure others have already done it. I suspect the answer is, "not very well."

Kevin
If I make a calculation error, #Cruncher probably will let me know.
Makefile
Posts: 2657
Joined: Fri Apr 22, 2016 11:03 pm

Re: Short-term breakeven inflation

Post by Makefile »

Thanks for analyzing and posting this. I remember your posts about the fluctuations in municipal bond money market yields, but obviously the practical use for that is unfortunately little until interest rates rise.

I'm not an economist so I don't fully understand all the interpretations of a 5 year TIPS bond having a yield of -1.76%. I think my closest explanation is that the yields on 5 year nominal bonds are so poor, that investors have migrated to 5 year TIPS bonds to the point of driving the price up (and the yield down to what seems like a significantly negative level), based on their guess of inflation, thus the "breakeven inflation rate." Is that right?

How much of the TIPS yield vs. nominal yield is based on inflation expectations, or to what extend is there oversupply or undersupply of nominal or TIPS bonds and how much might that affect the yield?
User avatar
Topic Author
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Short-term breakeven inflation

Post by Kevin M »

Makefile wrote: Sat Jun 12, 2021 8:35 pm I'm not an economist so I don't fully understand all the interpretations of a 5 year TIPS bond having a yield of -1.76%. I think my closest explanation is that the yields on 5 year nominal bonds are so poor, that investors have migrated to 5 year TIPS bonds to the point of driving the price up (and the yield down to what seems like a significantly negative level), based on their guess of inflation, thus the "breakeven inflation rate." Is that right?
I don't know that I'd say "investors have migrated" to anything. There is always a tradeoff between TIPS and nominal Treasuries.

If average inflation over a period equal to the term to maturity equals the BEI rate, then the TIPS and nominal Treasury will provide the same return, both nominal and real.

If average inflation is higher than the BEI rate, then the TIPS will have a higher nominal return, since the inflation adjustment will be higher than expected.

If average inflation is lower than the BEI rate, then the TIPS will have a lower nominal return, since the inflation adjustment will be lower than expected.

Rounding a bit, say the 5-year nominal yield is 0.75% and the TIPS yield is -1.75%, so the BEI rate is 2.25% 2.50%. The nominal Treasury will provide a certain nominal return of about 0.75%, with an uncertain real return. The TIPS will provide a certain -1.75% real return, with an uncertain nominal return. If average inflation equals 2.25% 2.50%, then both will provide a nominal return of 0.75% and a real return of -1.75%.

Makefile wrote: Sat Jun 12, 2021 8:35 pm How much of the TIPS yield vs. nominal yield is based on inflation expectations, or to what extend is there oversupply or undersupply of nominal or TIPS bonds and how much might that affect the yield?
Most of the difference between the nominal and TIPS yields represents expected inflation over a time period equal to the term of the bonds. There is a liquidity price premium for nominal Treasuries; they are more liquid than TIPS, which tends to increase the price (and decrease the yield). There is an unexpected inflation yield premium for nominal Treasuries, which tends to lower the price and increase the yield. Note that these premia work in different directions on the price and yield, so they offset each other to some extent. These premia are not directly observable, although there are various models that attempt to estimate them, so as a first-level estimate of expected inflation, I ignore them.

Kevin
Last edited by Kevin M on Mon Jun 14, 2021 2:02 pm, edited 1 time in total.
If I make a calculation error, #Cruncher probably will let me know.
Thesaints
Posts: 5087
Joined: Tue Jun 20, 2017 12:25 am

Re: Short-term breakeven inflation

Post by Thesaints »

That's very nice Kevin M !

We indeed are expecting short term inflation much higher than in the past decade+.
btw, I think the 20-year anomalous behavior of the BEI may be due to stale data.
User avatar
Topic Author
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Short-term breakeven inflation

Post by Kevin M »

Thesaints wrote: Sun Jun 13, 2021 8:19 pm That's very nice Kevin M !

We indeed are expecting short term inflation much higher than in the past decade+.
btw, I think the 20-year anomalous behavior of the BEI may be due to stale data.
Thanks.

Don't know what you mean by "stale data". The 20-year Treasury and TIPS yields are just as current as the yields for other maturities.

Here is a chart showing the BEI for all of the Fidelity quotes:

Image

Note that the BEI for the 18.7-year and 20.7-year yields are more in line with the curve, while the BEI rate for the 19.7-year yields is high relative to the curve. Whatever the Treasury is using for its constant-maturity algorithm is resulting in a yield above the curve for the 20-year constant maturity, so more in line with the 19.7-year Fidelity BEI than the two maturities on either side of it.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
Thesaints
Posts: 5087
Joined: Tue Jun 20, 2017 12:25 am

Re: Short-term breakeven inflation

Post by Thesaints »

Kevin M wrote: Sun Jun 13, 2021 10:28 pm Don't know what you mean by "stale data". The 20-year Treasury and TIPS yields are just as current as the yields for other maturities.
It is a big anomaly; maybe 15 bp. Yes, the quote is present, but it can have been derived from older trades. Admittedly, I don't know if they use bid/ask, or trades, to extract the current yields..
Even if they used bid/ask, it could be that the particular maturity only has a thinly traded issue. Again, mine are only hypotheses.
Northern Flicker
Posts: 15289
Joined: Fri Apr 10, 2015 12:29 am

Re: Short-term breakeven inflation

Post by Northern Flicker »

Nate79 wrote: Sat Jun 12, 2021 8:04 pm Historically how well has the market predicted the future inflation rates?
Reasonably well since 2009. Not very well before 2009.

https://seekingalpha.com/article/270125 ... -inflation
User avatar
typical.investor
Posts: 5247
Joined: Mon Jun 11, 2018 3:17 am

Re: Short-term breakeven inflation

Post by typical.investor »

Kevin M wrote: Sun Jun 13, 2021 8:15 pm Most of the difference between the nominal and TIPS yields represents expected inflation over a time period equal to the term of the bonds. There is a liquidity price premium for nominal Treasuries; they are more liquid than TIPS, which tends to increase the price (and decrease the yield). There is an unexpected inflation yield premium for nominal Treasuries, which tends to lower the price and increase the yield. Note that these premia work in different directions on the price and yield, so they offset each other to some extent. These premia are not directly observable, although there are various models that attempt to estimate them, so as a first-level estimate of expected inflation, I ignore them.

Kevin
That is interesting and why generally BEI isn't equivalent to actual inflation expectations. Usually nominals yield more than expected inflation to compensate for risk of inflation coming in higher.

It's not a huge amount though so many simply use BEI as a quick and easy proxy for inflation expectations.

In any case, the most interesting thing is how the inflation yield premium can switch to be negative in a deflationary crash which effectively lowers the yield and raises the price and is one reason treasuries are good for rebalancing. TIPS, on the other hand, behave differently when capital is scarce, and there will be a decrease in price and increase in yield (i.e. see a good time to buy).
rossington
Posts: 1806
Joined: Fri Jun 07, 2019 2:00 am
Location: Florida

Re: Short-term breakeven inflation

Post by rossington »

:happy
Kevin M wrote: Sun Jun 13, 2021 8:15 pm
Rounding a bit, say the 5-year nominal yield is 0.75% and the TIPS yield is -1.75%, so the BEI rate is 2.25%.
Kevin
Wouldn't it be 2.5%? :happy
"Success is going from failure to failure without loss of enthusiasm." Winston Churchill.
Dude2
Posts: 1777
Joined: Fri Jun 08, 2007 3:40 pm
Location: FL

Re: Short-term breakeven inflation

Post by Dude2 »

Northern Flicker wrote: Mon Jun 14, 2021 1:35 am
Nate79 wrote: Sat Jun 12, 2021 8:04 pm Historically how well has the market predicted the future inflation rates?
Reasonably well since 2009. Not very well before 2009.

https://seekingalpha.com/article/270125 ... -inflation
During times of great uncertainty, TIPS prices have been bid up to unrealistically high levels due to the market's craving for certainty, TIPS being the only risk-free instrument that is also protected against loss of purchasing power. In other words, TIPS are a unique and safe asset that is especially prized during times of great uncertainty, and that makes them prone to underestimate future inflation.
How can something that is indexed to inflation not track inflation successfully? There is a significant watering-down of the purity of using TIPS as a reference frame for value. I think that many people will consider that a $10,000 TIPS bought today will be worth $10,000 in real dollars when it matures [for simplicity, consider held in tax-advantaged account]. However, there isn't any such thing as a $10,000 TIPS bought today because that TIPS will get auctioned off and valued based on market expectations of future interest rate and inflation expectations that may be completely wrong. The $ unit of value is floating in a soup of uncertainty that TIPS doesn't anchor. This concept of the market coming in and skewing TIPS value boggles my mind. I Bonds are probably the pure version of TIPS that we would like regular TIPS to be. In that case only can you believe that your $10,000 will be worth $10,000 real in 30 years, maybe a little more.
Then ’tis like the breath of an unfee’d lawyer.
Northern Flicker
Posts: 15289
Joined: Fri Apr 10, 2015 12:29 am

Re: Short-term breakeven inflation

Post by Northern Flicker »

By claiming that TIPS underestimate future inflation, the author is presenting an opinion that refers to how well TIPS breakeven rates predict future inflation or how well they represent market expectations for inflation. It is not a statement about the real return of holding a TIPS.
Last edited by Northern Flicker on Mon Jun 14, 2021 10:56 pm, edited 1 time in total.
Thesaints
Posts: 5087
Joined: Tue Jun 20, 2017 12:25 am

Re: Short-term breakeven inflation

Post by Thesaints »

Northern Flicker wrote: Mon Jun 14, 2021 1:22 pm I will add that I think the author's view is likely at odds with some published findings that TIPS breakevens slightly overestimate future expected inflation due to a liquidity risk premium incorporated into TIPS yields.
Yep. That too. Maybe at the end we can say that added liquidity risk cancels out with reduced inflation risk, more otr less.
User avatar
Topic Author
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Short-term breakeven inflation

Post by Kevin M »

rossington wrote: Mon Jun 14, 2021 4:42 am :happy
Kevin M wrote: Sun Jun 13, 2021 8:15 pm
Rounding a bit, say the 5-year nominal yield is 0.75% and the TIPS yield is -1.75%, so the BEI rate is 2.25%.
Kevin
Wouldn't it be 2.5%? :happy
Yep, thanks. Corrected.
If I make a calculation error, #Cruncher probably will let me know.
User avatar
Topic Author
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Short-term breakeven inflation

Post by Kevin M »

Thesaints wrote: Mon Jun 14, 2021 1:24 am
Kevin M wrote: Sun Jun 13, 2021 10:28 pm Don't know what you mean by "stale data". The 20-year Treasury and TIPS yields are just as current as the yields for other maturities.
It is a big anomaly; maybe 15 bp. Yes, the quote is present, but it can have been derived from older trades. Admittedly, I don't know if they use bid/ask, or trades, to extract the current yields..
Even if they used bid/ask, it could be that the particular maturity only has a thinly traded issue. Again, mine are only hypotheses.
The treasury.gov yields use bid quotes shortly before market close, which you can read about on the yields website. I use ask yields, having used ones shortly before market close for my posts above. The bid/ask spread on Treasuries typically is tiny, so it doesn't matter much whether bid or ask is used. But definitely bid or ask, not trades.

I can come back and examine trading volumes later, but Treasuries normally have high trading volume and liquidity, resulting in very low bid/ask spreads.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
Northern Flicker
Posts: 15289
Joined: Fri Apr 10, 2015 12:29 am

Re: Short-term breakeven inflation

Post by Northern Flicker »

Thesaints wrote: Mon Jun 14, 2021 1:25 pm
Northern Flicker wrote: Mon Jun 14, 2021 1:22 pm I will add that I think the author's view is likely at odds with some published findings that TIPS breakevens slightly overestimate future expected inflation due to a liquidity risk premium incorporated into TIPS yields.
Yep. That too. Maybe at the end we can say that added liquidity risk cancels out with reduced inflation risk, more otr less.
I think the opposite. TIPS will be more liquid when inflation risk is perceived to be higher and demand for TIPS is higher, and less liquid when inflation risk is deemed lower or the market is pricing in deflation or disinflation. Demand for TIPS will be lower, which will render them less liquid.

But my original comment to which you were responding was incorrect. There seems to be sn opinion that TIPS breakevens tend to overestimate expected inflation, but it woukd not be because of a liquidity risk premium in TIPS. Such a premium would increase the yield of TIPS which would lower the TIPS breakeven rate from what it otherwise would be if TIPS had the same liquidity as nominal treasuries.
Last edited by Northern Flicker on Mon Jun 14, 2021 11:02 pm, edited 2 times in total.
User avatar
Topic Author
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Short-term breakeven inflation

Post by Kevin M »

Northern Flicker wrote: Mon Jun 14, 2021 1:35 am
Nate79 wrote: Sat Jun 12, 2021 8:04 pm Historically how well has the market predicted the future inflation rates?
Reasonably well since 2009. Not very well before 2009.
https://seekingalpha.com/article/270125 ... -inflation
That article was written in 2014. It wouldn't be difficult to use the same approach to update it for more recent years. The required data series are available from FRED.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
Topic Author
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Short-term breakeven inflation

Post by Kevin M »

Kevin M wrote: Mon Jun 14, 2021 2:06 pm
Thesaints wrote: Mon Jun 14, 2021 1:24 am
Kevin M wrote: Sun Jun 13, 2021 10:28 pm Don't know what you mean by "stale data". The 20-year Treasury and TIPS yields are just as current as the yields for other maturities.
It is a big anomaly; maybe 15 bp. Yes, the quote is present, but it can have been derived from older trades. Admittedly, I don't know if they use bid/ask, or trades, to extract the current yields..
Even if they used bid/ask, it could be that the particular maturity only has a thinly traded issue. Again, mine are only hypotheses.
The treasury.gov yields use bid quotes shortly before market close, which you can read about on the yields website. I use ask yields, having used ones shortly before market close for my posts above. The bid/ask spread on Treasuries typically is tiny, so it doesn't matter much whether bid or ask is used. But definitely bid or ask, not trades.

I can come back and examine trading volumes later, but Treasuries normally have high trading volume and liquidity, resulting in very low bid/ask spreads.

Kevin
Here's what I found in researching this a bit more (we are discussing the 20-year BEI rate, which appears a bit high relative to the curve).

The maturity date I use for this is 2/15/2041, since that's the closest there is to a 20-year maturity TIPS (19.7-year). Although there is only one TIPS maturing on that date, there are two Treasuries (excluding zero coupon):
  • yield = 1.98%, coupon = 4.75%
  • yield = 2.10%, coupon = 1.875%
My spreadsheet logic happens to pick the one with the higher yield, which results in a BEI of 12 basis points higher than if the lower yield were used. Interestingly, the one my spreadsheet function (VLOOKUP) picks is closer to what Treasury would use, since it's closer to a par bond; i.e., the yield is closer to the coupon rate. This explains why the BEI derived from the treasury.gov yields has a similar bump at this point.

Note that the yield for the higher-coupon Treasury is lower because the duration is shorter, and bonds are priced more based on duration than on maturity.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
grabiner
Advisory Board
Posts: 35265
Joined: Tue Feb 20, 2007 10:58 pm
Location: Columbia, MD

Re: Short-term breakeven inflation

Post by grabiner »

Kevin M wrote: Mon Jun 14, 2021 5:55 pm Here's what I found in researching this a bit more (we are discussing the 20-year BEI rate, which appears a bit high relative to the curve).

The maturity date I use for this is 2/15/2041, since that's the closest there is to a 20-year maturity TIPS (19.7-year). Although there is only one TIPS maturing on that date, there are two Treasuries (excluding zero coupon):
  • yield = 1.98%, coupon = 4.75%
  • yield = 2.10%, coupon = 1.875%
My spreadsheet logic happens to pick the one with the higher yield, which results in a BEI of 12 basis points higher than if the lower yield were used. Interestingly, the one my spreadsheet function (VLOOKUP) picks is closer to what Treasury would use, since it's closer to a par bond; i.e., the yield is closer to the coupon rate. This explains why the BEI derived from the treasury.gov yields has a similar bump at this point.
So this is a duration issue. The bond with the 4.75% coupon has much more of its value from payments made in less than 20 years, and thus less exposure to the 20-year inflation rate and more to shorter-term inflation rates.

This is an issue even comparing new-issue bonds. If a new 30-year TIPS has a zero yield, its duration is 30 years, so the break-even inflation should be computed by comparison to a 30-year STRIP, not a nominal bond. Stripped TIPS don't trade, so you would have to compute a synthetic price for a stripped TIPS by taking the bond price and subtracting the value of all the coupons using data from shorter-term TIPS.
Wiki David Grabiner
UpperNwGuy
Posts: 9446
Joined: Sun Oct 08, 2017 7:16 pm

Re: Short-term breakeven inflation

Post by UpperNwGuy »

Kevin M wrote: Sat Jun 12, 2021 6:51 pm To get some skin in the game, and to see how things work out over the next couple of years, I recently bought ten 2-year nominal Treasuries and ten 2-year TIPS. If inflation over the next two years is close to the 2-year BEI of about 2.8%, then each of these should provide about the same total return. If inflation is higher than expected, the TIPS should provide a higher total return than the nominal Treasury, and vice versa.
I will eagerly await the results of your experiment.
User avatar
Topic Author
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Short-term breakeven inflation

Post by Kevin M »

Kevin M wrote: Mon Jun 14, 2021 5:16 pm
Northern Flicker wrote: Mon Jun 14, 2021 1:35 am
Nate79 wrote: Sat Jun 12, 2021 8:04 pm Historically how well has the market predicted the future inflation rates?
Reasonably well since 2009. Not very well before 2009.
https://seekingalpha.com/article/270125 ... -inflation
That article was written in 2014. It wouldn't be difficult to use the same approach to update it for more recent years. The required data series are available from FRED.
I downloaded the 5-year BEI and CPIAUCNS series, calculated the subsequent 5-year annualized inflation from the CPI numbers, and created this chart from the data:

Image

Note that this chart is different than the one in the linked article, in that I show the subsequent 5-year inflation, whereas the article chart showed the trailing 5-year inflation. So in my chart, the BEI rate for Jan 2004 is 2.03%, and the 5-year annualized inflation from Jan 2004 to Jan 2009 is 2.66%.

Zooming in to ignore the 2008 financial crisis aberration and truncating the top of the y-axis as well, so we see more resolution for the majority of the data:

Image

Doesn't seem so bad to me. Note that subsequent 5-year inflation has been lower than the initial BEI rate more often than not for this period, although that's not the case for the last couple of years.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
Topic Author
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Short-term breakeven inflation

Post by Kevin M »

I think a better way to view these types of things is by looking at the difference between the actual and the "expectation" , as shown below:

Image

Again, let's zoom in to see more resolution for most of the data by truncating the outliers:

Image

Looks like a majority of the deltas are +/- 0.5 percentage points. The average of the deltas is 0.06 percentage points, the median is -0.16 percentage points, and the standard deviation of the deltas is 0.77 percentage points. There are 67 deltas > 0%, and 94 < 0%, so realized inflation was indeed < BEI for more periods than not.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
Thesaints
Posts: 5087
Joined: Tue Jun 20, 2017 12:25 am

Re: Short-term breakeven inflation

Post by Thesaints »

Northern Flicker wrote: Mon Jun 14, 2021 2:54 pm
Thesaints wrote: Mon Jun 14, 2021 1:25 pm
Northern Flicker wrote: Mon Jun 14, 2021 1:22 pm I will add that I think the author's view is likely at odds with some published findings that TIPS breakevens slightly overestimate future expected inflation due to a liquidity risk premium incorporated into TIPS yields.
Yep. That too. Maybe at the end we can say that added liquidity risk cancels out with reduced inflation risk, more otr less.
I think the opposite. TIPS will be more liquid when inflation risk is perceived to be higher and demand for TIPS is higher, and less liquid when inflation risk is deemed lower or the market is pricing in deflation or disinflation. Demand for TIPS will be lower, which will render them less liquid.
There are about 20 trillions of marketable fixed rate treasuries outstanding vs. 1.5 trillions of TIPS. That alone tells you that TIPS are less liquid.
In addition, fixed-rate issues can be used to match future known liabilities whereas TIPS can't. There is no doubt that fixed-rate issues pay a liquidity premium over TIPS.
A third hint is the auction calendar: how often can you buy fixed-rate new issues vs. TIPS ?
Thesaints
Posts: 5087
Joined: Tue Jun 20, 2017 12:25 am

Re: Short-term breakeven inflation

Post by Thesaints »

Kevin M wrote: Mon Jun 14, 2021 9:50 pm I think a better way to view these types of things is by looking at the difference between the actual and the "expectation" , as shown below:

Image

Again, let's zoom in to see more resolution for most of the data by truncating the outliers:

Image

Looks like a majority of the deltas are +/- 0.5 percentage points. The average of the deltas is 0.06 percentage points, the median is -0.16 percentage points, and the standard deviation of the deltas is 0.77 percentage points. There are 67 deltas > 0%, and 94 < 0%, so realized inflation was indeed < BEI for more periods than not.

Kevin
But there are big deltas every now and then. If TIPS forecast a 2% inflation and CPI turns out to increase by 2.5%, one could say it is only a 0.5% error. Others may say the forecast was off by 20%.
Your plots I believe kind of confirm that TIPS are priced based on the expected inflation and unexpected inflation makes them yield less or more than fixed-rate.
Northern Flicker
Posts: 15289
Joined: Fri Apr 10, 2015 12:29 am

Re: Short-term breakeven inflation

Post by Northern Flicker »

There always will be deltas because the market cannot incorporate future unpredictable events into the rates. In 2018, TIPS breakevens established by the market did not incorporate the effect of the Covid pandemic because nobody would have predicted it. While we still don't know what its effect will be on inflation over say even a medium term, I think it is safe to say that it increased the variance in any prediction.
User avatar
Topic Author
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Short-term breakeven inflation

Post by Kevin M »

Thesaints wrote: Mon Jun 14, 2021 11:02 pm In addition, fixed-rate issues can be used to match future known liabilities whereas TIPS can't.
Not sure what you mean by this.

TIPS can be used to match future known real liabilities. They are the primary component of a liability matching portfolio, as popularized, for example, by Bill Bernstein. A number of Bogleheads build TIPS ladders for this purpose.

Nominal Treasuries can be used to match future nominal liabilities.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
rockstar
Posts: 6308
Joined: Mon Feb 03, 2020 5:51 pm

Re: Short-term breakeven inflation

Post by rockstar »

Kevin M wrote: Tue Jun 15, 2021 7:31 pm
Thesaints wrote: Mon Jun 14, 2021 11:02 pm In addition, fixed-rate issues can be used to match future known liabilities whereas TIPS can't.
Not sure what you mean by this.

TIPS can be used to match future known real liabilities. They are the primary component of a liability matching portfolio, as popularized, for example, by Bill Bernstein. A number of Bogleheads build TIPS ladders for this purpose.

Nominal Treasuries can be used to match future nominal liabilities.

Kevin
TIPS can't keep up with inflation right now, and you're even worse after taxes. What's the argument for buying them?
Thesaints
Posts: 5087
Joined: Tue Jun 20, 2017 12:25 am

Re: Short-term breakeven inflation

Post by Thesaints »

Kevin M wrote: Tue Jun 15, 2021 7:31 pm
Thesaints wrote: Mon Jun 14, 2021 11:02 pm In addition, fixed-rate issues can be used to match future known liabilities whereas TIPS can't.
Not sure what you mean by this.

TIPS can be used to match future known real liabilities. They are the primary component of a liability matching portfolio, as popularized, for example, by Bill Bernstein. A number of Bogleheads build TIPS ladders for this purpose.

Nominal Treasuries can be used to match future nominal liabilities.

Kevin
Sure, but few institutions have future "real" liabilities and institutional trade forms prices and yield, not the individuals'.
User avatar
Svensk Anga
Posts: 1596
Joined: Sun Dec 23, 2012 4:16 pm

Re: Short-term breakeven inflation

Post by Svensk Anga »

rockstar wrote: Tue Jun 15, 2021 7:38 pm
TIPS can't keep up with inflation right now, and you're even worse after taxes. What's the argument for buying them?
Nominal Treasuries' return might turn out much worse than even negative real yield TIPS.
Thesaints
Posts: 5087
Joined: Tue Jun 20, 2017 12:25 am

Re: Short-term breakeven inflation

Post by Thesaints »

One thing to consider is also the maximum upside.
If actual inflation turns out to be exactly equal to the implied inflation, owning fixed-rate and TIPS will be equivalent (ok, minus some tax difference to be accurate).
If actual inflation turns out to be higher, then the TIPS owner will enjoy a higher return, but his real return can never be positive, however high inflation might end up being.. In fact, his real return will be exactly -1.67% regardless of inflation (5 year TIPS).
On the other hand, the fixed-rate treasuries holder has a positive real return upside: if inflation turns out to be lower than the implied value his real return would be positive. Best possible outcome being zero inflation, corresponding to a +0.79% real return; I'm neglecting the very small possibility of having net negative inflation over a 5-year span.
rockstar
Posts: 6308
Joined: Mon Feb 03, 2020 5:51 pm

Re: Short-term breakeven inflation

Post by rockstar »

Svensk Anga wrote: Wed Jun 16, 2021 1:19 pm
rockstar wrote: Tue Jun 15, 2021 7:38 pm
TIPS can't keep up with inflation right now, and you're even worse after taxes. What's the argument for buying them?
Nominal Treasuries' return might turn out much worse than even negative real yield TIPS.
They're both awful. TIPS are slightly less awful. But I wouldn't own either.
Post Reply