Faber: T-bills & Chill...Most of the Time

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

In this article, Feb outlines a bond-buying timing strategy that claims it doesn't require any guesses about the future, and improves risk-adjusted returns.

1. Hold T-bills by default.
2. When bonds yields are in the top 50%, historically, buy them

There is a lot more to it than that, and he studies the strategy first with Treasuries, but then goes to IG corporate, EM, HY, etc.

https://www.cambriainvestments.com/wp-c ... e-Time.pdf

The concluding paragraphs:
Unfortunately, and despite the current drawdown many bond investors are experiencing, some bonds categories still have even
lower yields than T-bills. The traditional yield curve is still inverted, and many, if not all of the assets examined in the paper do
not have a yield spread over T-bills that would place them in the favorable top half of history. If those bonds don’t compensate
you with a reasonable margin of safety, is that a wise decision to allocate?

We don’t think it is, but we do think there’s a better approach. With this paper, we attempted to illustrate the effectiveness
of implementing a tactical approach to fixed income investing based on the evaluation of yield spreads across a variety of fixed
income sectors. We think the power lies in its ability to guide investors over time based on objective rules that have historically
shown the ability to earn potentially higher returns than a simple buy-and-hold approach, but with lower volatility and drawdowns.
This simulation helps to answer the question, “When does it make sense to take on extra risk, and when might it not be prudent?”
Right now this systems is signaling: “T-bills and chill”. But at some point in the future, it will signal a potential green light to take
on the additional exposures once again.

With that said, we think this dynamic approach to fixed income can serve as a core approach for investors looking for a strategy
with an emphasis on the safety of Treasury bills, but systematically takes risk when opportunities and sufficient margins of
safety present themselves.

Perhaps we will need to expand the phrase in the coming years to: “Let’s just T-bills and chill…. most of the time”.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
dkturner
Posts: 1975
Joined: Sun Feb 25, 2007 6:58 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by dkturner »

How does one determine the top 50% of historical yield for a particular asset?
er999
Posts: 1466
Joined: Wed Nov 05, 2008 10:00 am

Re: Faber: T-bills & Chill...Most of the Time

Post by er999 »

Sounds reasonable, and maybe buy tips when returns good (like now). Although he seems to be advocating shifting in and out of longer term bonds which seems to increase the possibility of mistakes. With a tips ladder you only need to buy once.
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

dkturner wrote: Sat May 11, 2024 9:56 am How does one determine the top 50% of historical yield for a particular asset?
You look at the current yield and compare it to history.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Eno Deb
Posts: 798
Joined: Sun Feb 03, 2019 3:08 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by Eno Deb »

dkturner wrote: Sat May 11, 2024 9:56 am How does one determine the top 50% of historical yield for a particular asset?
I'm not sure if the author means yield or yield spread (he uses both). Yield spread would make more sense to me: only go to long durations if the reward of doing so is significant. Right now we have a historically low (negative) spread with the inverted yield curve, which would suggest to stay away from longer durations.
User avatar
Kevin M
Posts: 15881
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Faber: T-bills & Chill...Most of the Time

Post by Kevin M »

Interesting, but I'm not sold.

First problem is the way they define risk:
Generally, as an investor moves away from what everyone considers to be the risk-free asset, T-bills, they accept some
additional risk for the additional prospect of return.
Although it's true that tbills are widely considered the risk-free asset, "everyone" does not consider them to be so, myself included.

Risk in investing is the uncertainty of return for a specific holding period. Further, one must also consider the unit of account.

If the holding period is 3 months, and the unit of account is dollars, then a 3-month tbill is indeed the riskless asset. However, if the holding period is 10 years and the unit of account is real consumption, the closest thing to a riskless asset is a 10-year TIPS (the lower the coupon, the closer to riskless it is). For more on this, see Risk Basics: What is the risk-free asset? - Bogleheads.org.

Second, this reeks of data mining.

Finally, I don't think it's a good tactical approach to make decisions about buying TIPS. Below is a chart of 3-month tbill yields minus long-term real yields:

Image

Clearly we are in the bottom 50% of historical yield spreads of TIPS over 3m bills, yet I would argue that now is a good time to buy TIPS. This is especially true for building a TIPS ladder to fund residual expenses in retirement. To explore this further, let's decompose the chart above into the individual components:

Image

Eyeballing this chart, even though we are in the bottom 50% in terms of yield spreads, we are in the top 50% in terms of real yield itself, with average long-term real yield at about 2.3%. If we were to buy based on yield spreads, we would have preferred buying TIPS in early 2020 with yields of about 0% and spreads of about 0%, then now with yields about 2.3% and spreads about -3.2%.

Although a TIPS ladder serves its purpose of lowest-risk real consumption liability matching regardless of yields, it costs less to buy when real yields are higher. I'm happy I've built my ladder at yields in the 2% ballpark then when yields were 0% or lower.

Even with respect to tactical market timing, do we have enough history for TIPS (about a quarter century) to make any valid conclusions?
If I make a calculation error, #Cruncher probably will let me know.
User avatar
nisiprius
Advisory Board
Posts: 52718
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Faber: T-bills & Chill...Most of the Time

Post by nisiprius »

How well did his tactical asset allocation of stocks work?

Orange, Mebane Faber's "Global Tactical Asset Allocation" ETF, during the period he was managing it. (It continued for a few years more, with similar performance).

Blue: Vanguard LifeStrategy Conservative Growth

Image

How about the Ivy Portfolio? The Ivy Portfolio: How to Invest Like the Top Endowments and Avoid Bear Markets, by Mebane Faber and Eric W. Richardson, was published in 2011.

The Ivy Portfolio is 80/20 stocks/bonds, with the stock holdings 75/25 US/international. The red line is 60% VTI, 20% VEU, 20% AGG, thus using three of the holdings from the Ivy Portfolio and keeping the same relative proportions of those three. The yellow line is Vanguard's LifeStrategy Growth fund, which is also 80/20.

Source

Image
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

nisiprius wrote: Sat May 11, 2024 1:51 pm How well did his tactical asset allocation of stocks work?
What does this have to do with bonds?

If you read the article, the results for the bond backtests are shared.
Last edited by watchnerd on Sat May 11, 2024 3:01 pm, edited 1 time in total.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
MIretired
Posts: 1086
Joined: Fri Sep 06, 2013 12:35 pm
Location: Midwest, US

Re: Faber: T-bills & Chill...Most of the Time

Post by MIretired »

Kevin M wrote: Sat May 11, 2024 12:56 pm Interesting, but I'm not sold.

First problem is the way they define risk:
Generally, as an investor moves away from what everyone considers to be the risk-free asset, T-bills, they accept some
additional risk for the additional prospect of return.
Although it's true that tbills are widely considered the risk-free asset, "everyone" does not consider them to be so, myself included.

Risk in investing is the uncertainty of return for a specific holding period. Further, one must also consider the unit of account.

If the holding period is 3 months, and the unit of account is dollars, then a 3-month tbill is indeed the riskless asset. However, if the holding period is 10 years and the unit of account is real consumption, the closest thing to a riskless asset is a 10-year TIPS (the lower the coupon, the closer to riskless it is). For more on this, see Risk Basics: What is the risk-free asset? - Bogleheads.org.

Second, this reeks of data mining.

Finally, I don't think it's a good tactical approach to make decisions about buying TIPS. Below is a chart of 3-month tbill yields minus long-term real yields:

Image

Clearly we are in the bottom 50% of historical yield spreads of TIPS over 3m bills, yet I would argue that now is a good time to buy TIPS. This is especially true for building a TIPS ladder to fund residual expenses in retirement. To explore this further, let's decompose the chart above into the individual components:

Image

Eyeballing this chart, even though we are in the bottom 50% in terms of yield spreads, we are in the top 50% in terms of real yield itself, with average long-term real yield at about 2.3%. If we were to buy based on yield spreads, we would have preferred buying TIPS in early 2020 with yields of about 0% and spreads of about 0%, then now with yields about 2.3% and spreads about -3.2%.

Although a TIPS ladder serves its purpose of lowest-risk real consumption liability matching regardless of yields, it costs less to buy when real yields are higher. I'm happy I've built my ladder at yields in the 2% ballpark then when yields were 0% or lower.

Even with respect to tactical market timing, do we have enough history for TIPS (about a quarter century) to make any valid conclusions?
Yeah. I'd say your point, if I get it, is you're trying to hedge against inflation with TIPs, not outperform it.
Eno Deb
Posts: 798
Joined: Sun Feb 03, 2019 3:08 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by Eno Deb »

Kevin M wrote: Sat May 11, 2024 12:56 pmIf the holding period is 3 months, and the unit of account is dollars, then a 3-month tbill is indeed the riskless asset. However, if the holding period is 10 years and the unit of account is real consumption, the closest thing to a riskless asset is a 10-year TIPS (the lower the coupon, the closer to riskless it is).
That doesn't seem so obvious to me. There is historically a strong correlation between inflation and short-term interest rates after all, so holding rotating T-bills over longer time periods does offer some implicit protection against inflation.
User avatar
Kevin M
Posts: 15881
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Faber: T-bills & Chill...Most of the Time

Post by Kevin M »

Eno Deb wrote: Sat May 11, 2024 3:00 pm
Kevin M wrote: Sat May 11, 2024 12:56 pmIf the holding period is 3 months, and the unit of account is dollars, then a 3-month tbill is indeed the riskless asset. However, if the holding period is 10 years and the unit of account is real consumption, the closest thing to a riskless asset is a 10-year TIPS (the lower the coupon, the closer to riskless it is).
That doesn't seem so obvious to me. There is historically a strong correlation between inflation and short-term interest rates after all, so holding rotating T-bills over longer time periods does offer some implicit protection against inflation.
Unfortunately, the significant exceptions to tbills keeping up with inflation has been when interest rates have been artificially suppressed by the Fed. It hasn't been long since we went through one of these periods, with short term rates held close to 0% while inflation was well above 0%. The 1940s is another example of this.

Regardless of historical results, it's an economic fact that we don't know the 10-year nominal return, much less the 10-year real return, of rolling bills for 10 years (due to the reinvestment risk), while we know with much higher certainty the real 10-year return of 10-year TIPS held to maturity. There is some uncertainty in the coupon reinvestment rates, but there are TIPS with coupons as low as 0.125%, so these are pretty close to riskless if matching real consumption.

Having said that, I hold mostly short-term Treasuries for fixed income in taxable, and all TIPS for fixed income in tax-advantaged. By "short term", I don't just mean bills, but Treasuries with maturities from 0 years to 2 years, although with about 85% in maturities of one year or less. For holding periods of more than a few months, I consider my TIPS ladder as my relatively riskless liability matching portfolio (LMP), and my nominal Treasuries as part of my risk portfolio (RP) due to the reinvestment risk.

It so happens that this approach is in line with Bill Bernstein's recommendation:
To summarize, TIPS and T-bills are complementary assets. The former appeals to our System 2’s inner Spock, who first and foremost wants to secure our future consumption, while the latter assuages our System 1’s inner Daffy Duck, who wants us to bail at the worst possible time and violate Charlie Munger’s first rule of compounding, which is to never interrupt it.

The prudent retiree holds a goodly pile of both.
If I make a calculation error, #Cruncher probably will let me know.
muffins14
Posts: 5765
Joined: Wed Oct 26, 2016 4:14 am
Location: New York

Re: Faber: T-bills & Chill...Most of the Time

Post by muffins14 »

watchnerd wrote: Fri May 10, 2024 5:49 pm In this article, Feb outlines a bond-buying timing strategy that claims it doesn't require any guesses about the future, and improves risk-adjusted returns.

1. Hold T-bills by default.
2. When bonds yields are in the top 50%, historically, buy them

There is a lot more to it than that, and he studies the strategy first with Treasuries, but then goes to IG corporate, EM, HY, etc.

https://www.cambriainvestments.com/wp-c ... e-Time.pdf

The concluding paragraphs:
Unfortunately, and despite the current drawdown many bond investors are experiencing, some bonds categories still have even
lower yields than T-bills. The traditional yield curve is still inverted, and many, if not all of the assets examined in the paper do
not have a yield spread over T-bills that would place them in the favorable top half of history. If those bonds don’t compensate
you with a reasonable margin of safety, is that a wise decision to allocate?

We don’t think it is, but we do think there’s a better approach. With this paper, we attempted to illustrate the effectiveness
of implementing a tactical approach to fixed income investing based on the evaluation of yield spreads across a variety of fixed
income sectors. We think the power lies in its ability to guide investors over time based on objective rules that have historically
shown the ability to earn potentially higher returns than a simple buy-and-hold approach, but with lower volatility and drawdowns.
This simulation helps to answer the question, “When does it make sense to take on extra risk, and when might it not be prudent?”
Right now this systems is signaling: “T-bills and chill”. But at some point in the future, it will signal a potential green light to take
on the additional exposures once again.

With that said, we think this dynamic approach to fixed income can serve as a core approach for investors looking for a strategy
with an emphasis on the safety of Treasury bills, but systematically takes risk when opportunities and sufficient margins of
safety present themselves.

Perhaps we will need to expand the phrase in the coming years to: “Let’s just T-bills and chill…. most of the time”.
It seems like this is just dealing with a 100% bond portfolio? It would have been more interesting to see this with 50/50 stocks bonds or 60/40 or 70/30 etc
Crom laughs at your Four Winds
User avatar
HomerJ
Posts: 21433
Joined: Fri Jun 06, 2008 12:50 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by HomerJ »

I'm not trying to maximize returns with my cash/bond side of my portfolio.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

muffins14 wrote: Sat May 11, 2024 4:18 pm
watchnerd wrote: Fri May 10, 2024 5:49 pm In this article, Feb outlines a bond-buying timing strategy that claims it doesn't require any guesses about the future, and improves risk-adjusted returns.

1. Hold T-bills by default.
2. When bonds yields are in the top 50%, historically, buy them

There is a lot more to it than that, and he studies the strategy first with Treasuries, but then goes to IG corporate, EM, HY, etc.

https://www.cambriainvestments.com/wp-c ... e-Time.pdf

The concluding paragraphs:
Unfortunately, and despite the current drawdown many bond investors are experiencing, some bonds categories still have even
lower yields than T-bills. The traditional yield curve is still inverted, and many, if not all of the assets examined in the paper do
not have a yield spread over T-bills that would place them in the favorable top half of history. If those bonds don’t compensate
you with a reasonable margin of safety, is that a wise decision to allocate?

We don’t think it is, but we do think there’s a better approach. With this paper, we attempted to illustrate the effectiveness
of implementing a tactical approach to fixed income investing based on the evaluation of yield spreads across a variety of fixed
income sectors. We think the power lies in its ability to guide investors over time based on objective rules that have historically
shown the ability to earn potentially higher returns than a simple buy-and-hold approach, but with lower volatility and drawdowns.
This simulation helps to answer the question, “When does it make sense to take on extra risk, and when might it not be prudent?”
Right now this systems is signaling: “T-bills and chill”. But at some point in the future, it will signal a potential green light to take
on the additional exposures once again.

With that said, we think this dynamic approach to fixed income can serve as a core approach for investors looking for a strategy
with an emphasis on the safety of Treasury bills, but systematically takes risk when opportunities and sufficient margins of
safety present themselves.

Perhaps we will need to expand the phrase in the coming years to: “Let’s just T-bills and chill…. most of the time”.
It seems like this is just dealing with a 100% bond portfolio? It would have been more interesting to see this with 50/50 stocks bonds or 60/40 or 70/30 etc

It just comparing bonds vs bonds.

He doesn't claim the model has any bearing on stock vs bond allocation.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Eno Deb
Posts: 798
Joined: Sun Feb 03, 2019 3:08 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by Eno Deb »

HomerJ wrote: Sat May 11, 2024 4:21 pm I'm not trying to maximize returns with my cash/bond side of my portfolio.
I used to think that too, but I think it's just because we have gotten (too) used to the super-low interest rates in the 2010s. Now you can make some serious money with "super safe" treasuries and it's worth paying some attention to it. Not sure if we will return to the low-interest regime anytime soon.
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

Eno Deb wrote: Sat May 11, 2024 5:38 pm
HomerJ wrote: Sat May 11, 2024 4:21 pm I'm not trying to maximize returns with my cash/bond side of my portfolio.
I used to think that too, but I think it's just because we have gotten (too) used to the super-low interest rates in the 2010s. Now you can make some serious money with "super safe" treasuries. Not sure if we will return to the low-interest regime anytime soon.
And / or expand beyond "super safe" Treasuries.

Vanguard's active EM bond debt fund has a 30 SEC yield of 7.36% right now, and HY is yielding 6.75%.

However, average historical spread on 10 YR Treasury vs HY is 5.1%, so we're well below the 50% yield spread mark.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

muffins14 wrote: Sat May 11, 2024 4:18 pm
watchnerd wrote: Fri May 10, 2024 5:49 pm In this article, Feb outlines a bond-buying timing strategy that claims it doesn't require any guesses about the future, and improves risk-adjusted returns.

1. Hold T-bills by default.
2. When bonds yields are in the top 50%, historically, buy them

There is a lot more to it than that, and he studies the strategy first with Treasuries, but then goes to IG corporate, EM, HY, etc.

https://www.cambriainvestments.com/wp-c ... e-Time.pdf

The concluding paragraphs:
Unfortunately, and despite the current drawdown many bond investors are experiencing, some bonds categories still have even
lower yields than T-bills. The traditional yield curve is still inverted, and many, if not all of the assets examined in the paper do
not have a yield spread over T-bills that would place them in the favorable top half of history. If those bonds don’t compensate
you with a reasonable margin of safety, is that a wise decision to allocate?

We don’t think it is, but we do think there’s a better approach. With this paper, we attempted to illustrate the effectiveness
of implementing a tactical approach to fixed income investing based on the evaluation of yield spreads across a variety of fixed
income sectors. We think the power lies in its ability to guide investors over time based on objective rules that have historically
shown the ability to earn potentially higher returns than a simple buy-and-hold approach, but with lower volatility and drawdowns.
This simulation helps to answer the question, “When does it make sense to take on extra risk, and when might it not be prudent?”
Right now this systems is signaling: “T-bills and chill”. But at some point in the future, it will signal a potential green light to take
on the additional exposures once again.

With that said, we think this dynamic approach to fixed income can serve as a core approach for investors looking for a strategy
with an emphasis on the safety of Treasury bills, but systematically takes risk when opportunities and sufficient margins of
safety present themselves.

Perhaps we will need to expand the phrase in the coming years to: “Let’s just T-bills and chill…. most of the time”.
It seems like this is just dealing with a 100% bond portfolio? It would have been more interesting to see this with 50/50 stocks bonds or 60/40 or 70/30 etc
Meb didn't cover this, but this page gets into interet rates vs stock valuations:

https://www.currentmarketvaluation.com/ ... -rates.php
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Logan Roy
Posts: 2016
Joined: Sun May 29, 2022 10:15 am

Re: Faber: T-bills & Chill...Most of the Time

Post by Logan Roy »

Meb Faber's books are some of the most fun and interesting things I've read in investing. I love the open-minded approach.

But I've never found one of his strategies I'd commit capital to. I think there's a very strong back-fitting bias that he doesn't seem acutely concerned about.

e.g. In concepts like 'favourable top half of history' – that value changes all the time, as history's being written. So when you backtest a strategy from 2010, you've already factored in where valuations spent the past 20 years, and how that affected the average. Yet when you're backtesting the year 1998, you didn't know that. So you're always transmitting information from the future into the past, unless you keep recalculating history averages – but even then, you've all this scope to tweak your strategy until it works. Then there's this concept of secular shifts – which means anything you've backtested won't necessarily apply to the future at all .. Fun but a bit dangerous. I think you have to be a bit smarter. You have to understand the past in terms of things that played out, and the future in terms of how things might play out. Otherwise anything you do is really just a map of the past, and often one that's sending information backwards.
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

Logan Roy wrote: Sat May 11, 2024 6:06 pm Meb Faber's books are some of the most fun and interesting things I've read in investing. I love the open-minded approach.

But I've never found one of his strategies I'd commit capital to. I think there's a very strong back-fitting bias that he doesn't seem acutely concerned about.

e.g. In concepts like 'favourable top half of history' – that value changes all the time, as history's being written. So when you backtest a strategy from 2010, you've already factored in where valuations spent the past 20 years, and how that affected the average. Yet when you're backtesting the year 1998, you didn't know that. So you're always transmitting information from the future into the past, unless you keep recalculating history averages – but even then, you've all this scope to tweak your strategy until it works. Then there's this concept of secular shifts – which means anything you've backtested won't necessarily apply to the future at all .. Fun but a bit dangerous. I think you have to be a bit smarter. You have to understand the past in terms of things that played out, and the future in terms of how things might play out. Otherwise anything you do is really just a map of the past, and often one that's sending information backwards.
He claims in the article that he used some sampling technique to alleviate this, but I didn't get it.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
Charles Joseph
Posts: 2737
Joined: Tue Apr 05, 2022 10:49 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by Charles Joseph »

watchnerd wrote: Sat May 11, 2024 2:55 pm
nisiprius wrote: Sat May 11, 2024 1:51 pm How well did his tactical asset allocation of stocks work?
What does this have to do with bonds?
That's like asking what Mario Mendoza's batting average has to do with his slugging percentage.
Retired June 2023
Elysium
Posts: 4218
Joined: Mon Apr 02, 2007 6:22 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by Elysium »

Deleted.

I see that the idea of staying in T-Bills most of the time is a low risk plan, unless the yield spread looks large enough to go towards longer end (or intermediate).

This is not a new idea. I recall Larry Swedroe and Bill Bernstein used to talk about it. Swedroe on the Treasury yield curve and Bernstein spoke about the HY spread.

I'd agree that corporate yield spreads are not enough atm to compensate for the risk, HY and EM isn't either. As for Long end rates, I am as confused as everyone else, because the rates are almost same between long and intermediate treasuries. It would mean there is no compensation for going longer, but then there is nothing to go from Short or T-bill to Intermediate either.

The issue is by the time we get to know this, the rates on T-Bills and Short end may fall, and then the gains will be already made in the intermediate/long end. You will end up buying after an initial surge, then what if the rates go up and you are left holding the bag??
User avatar
HomerJ
Posts: 21433
Joined: Fri Jun 06, 2008 12:50 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by HomerJ »

Eno Deb wrote: Sat May 11, 2024 5:38 pm
HomerJ wrote: Sat May 11, 2024 4:21 pm I'm not trying to maximize returns with my cash/bond side of my portfolio.
I used to think that too, but I think it's just because we have gotten (too) used to the super-low interest rates in the 2010s. Now you can make some serious money with "super safe" treasuries and it's worth paying some attention to it. Not sure if we will return to the low-interest regime anytime soon.
What i mean is, the 4%-5% I'm getting from money-markets, and the 2%+ real I'm getting from intermediate TIPs is plenty good.

I am absolutely LOVING the higher interest rates, and indeed making "serious" money even locking in longer-term current rates today.

There is zero need for me to play some "T-bill/wait to invest in longer bonds when they are above 50% historical" game.

It's a game.

I don't need to play.

The current longer-term interest rates, even if it not above the 50% historical mark, are plenty good for me. I'm set.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
Joey Jo Jo Jr
Posts: 564
Joined: Mon Jun 21, 2021 11:38 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by Joey Jo Jo Jr »

Kevin M wrote: Sat May 11, 2024 12:56 pm Interesting, but I'm not sold.

First problem is the way they define risk:
Generally, as an investor moves away from what everyone considers to be the risk-free asset, T-bills, they accept some
additional risk for the additional prospect of return.
Although it's true that tbills are widely considered the risk-free asset, "everyone" does not consider them to be so, myself included.

Risk in investing is the uncertainty of return for a specific holding period. Further, one must also consider the unit of account.

If the holding period is 3 months, and the unit of account is dollars, then a 3-month tbill is indeed the riskless asset. However, if the holding period is 10 years and the unit of account is real consumption, the closest thing to a riskless asset is a 10-year TIPS (the lower the coupon, the closer to riskless it is). For more on this, see Risk Basics: What is the risk-free asset? - Bogleheads.org.

Second, this reeks of data mining.

Finally, I don't think it's a good tactical approach to make decisions about buying TIPS. Below is a chart of 3-month tbill yields minus long-term real yields:

Image

Clearly we are in the bottom 50% of historical yield spreads of TIPS over 3m bills, yet I would argue that now is a good time to buy TIPS. This is especially true for building a TIPS ladder to fund residual expenses in retirement. To explore this further, let's decompose the chart above into the individual components:

Image

Eyeballing this chart, even though we are in the bottom 50% in terms of yield spreads, we are in the top 50% in terms of real yield itself, with average long-term real yield at about 2.3%. If we were to buy based on yield spreads, we would have preferred buying TIPS in early 2020 with yields of about 0% and spreads of about 0%, then now with yields about 2.3% and spreads about -3.2%.

Although a TIPS ladder serves its purpose of lowest-risk real consumption liability matching regardless of yields, it costs less to buy when real yields are higher. I'm happy I've built my ladder at yields in the 2% ballpark then when yields were 0% or lower.

Even with respect to tactical market timing, do we have enough history for TIPS (about a quarter century) to make any valid conclusions?
Second chart and a strong IPA has me channeling TLC: “Don’t go chasing T bills y’all, please stick to the real yields that are higher than usual. I know that you want to have some safe yield then reinvest, but I think you’ll be moving too slow. [T boz part] y’all don’t hear me!”
Eno Deb
Posts: 798
Joined: Sun Feb 03, 2019 3:08 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by Eno Deb »

Personally I don't trust TIPS. Historically there were long stretches of time where they significantly underperformed treasuries of similar duration even in real terms, then we had negative yields, and during the recent inflation spike they utterly failed since the price decline caused by the rising interest rates more than offset the principal adjustment. As far as I'm concerned TIPS haven't demonstrated yet that they are a better inflation hedge than stocks (or even T-Bills, as discussed above).

Another thing is that I don't really have a specific foreseeable "consumption liability" to match.

Anyway, for me this isn't so much about TIPS vs treasuries but about durations and risk/return tradeoffs.
marcopolo
Posts: 8716
Joined: Sat Dec 03, 2016 9:22 am

Re: Faber: T-bills & Chill...Most of the Time

Post by marcopolo »

nisiprius wrote: Sat May 11, 2024 1:51 pm How well did his tactical asset allocation of stocks work?

Orange, Mebane Faber's "Global Tactical Asset Allocation" ETF, during the period he was managing it. (It continued for a few years more, with similar performance).

Blue: Vanguard LifeStrategy Conservative Growth

Image

How about the Ivy Portfolio? The Ivy Portfolio: How to Invest Like the Top Endowments and Avoid Bear Markets, by Mebane Faber and Eric W. Richardson, was published in 2011.

The Ivy Portfolio is 80/20 stocks/bonds, with the stock holdings 75/25 US/international. The red line is 60% VTI, 20% VEU, 20% AGG, thus using three of the holdings from the Ivy Portfolio and keeping the same relative proportions of those three. The yellow line is Vanguard's LifeStrategy Growth fund, which is also 80/20.

Source

Image
You beat me to it!

And why it's relevant to bonds is that it is the same person that was making the same claims about being able tactically time those investments. I am sure he was as sure about that being successful.
Once in a while you get shown the light, in the strangest of places if you look at it right.
marcopolo
Posts: 8716
Joined: Sat Dec 03, 2016 9:22 am

Re: Faber: T-bills & Chill...Most of the Time

Post by marcopolo »

watchnerd wrote: Sat May 11, 2024 2:55 pm
nisiprius wrote: Sat May 11, 2024 1:51 pm How well did his tactical asset allocation of stocks work?
What does this have to do with bonds?

If you read the article, the results for the bond backtests are shared.
Do you suppose he might have shared similarly convincing looking data when he proposed a tactical timing approach to stocks?
Once in a while you get shown the light, in the strangest of places if you look at it right.
seajay
Posts: 1762
Joined: Sat May 01, 2021 3:26 pm
Contact:

Re: Faber: T-bills & Chill...Most of the Time

Post by seajay »

From the article/paper
Term Premium – Instead of investing in short term T-bills, an investor could choose to invest in longer maturity Treasuries such as 30-year Treasury bonds. Additional compensation is expected for an uncertain future and additional volatility associated with the path of interest rates. Usually, if T-bills yield 5% then long-term bonds may yield 6% or 7%, in what would be described as an upward sloping yield curve.
5% near end is quite high in reflection of 2% target inflation rates. When at 5% the near term is suggesting above average inflation, that will later decline such that longer term bonds may be priced to lower yields, 3% or 4% perhaps.

Seems to me that in a series of rising yields that you'll switch into longer dated, see subsequent yields rise ... up to a peak point and then decline back down again before switching back to T-Bills. In bond price terms that will see earlier losses (as yields rise), gains as yields fall back down again, 0% overall average, BUT where you had a bad sequence of returns, endured losses in earlier years.

The presented results figures don't seem to fit. I tried a basic model along similar lines for UK data and observed worse results when bonds were marked-to-market, comparable to a 50/50 T-Bill/Bond barbell.
Logan Roy
Posts: 2016
Joined: Sun May 29, 2022 10:15 am

Re: Faber: T-bills & Chill...Most of the Time

Post by Logan Roy »

nisiprius wrote: Sat May 11, 2024 1:51 pm How well did his tactical asset allocation of stocks work?

Orange, Mebane Faber's "Global Tactical Asset Allocation" ETF, during the period he was managing it. (It continued for a few years more, with similar performance).

Blue: Vanguard LifeStrategy Conservative Growth

Image

How about the Ivy Portfolio? The Ivy Portfolio: How to Invest Like the Top Endowments and Avoid Bear Markets, by Mebane Faber and Eric W. Richardson, was published in 2011.

The Ivy Portfolio is 80/20 stocks/bonds, with the stock holdings 75/25 US/international. The red line is 60% VTI, 20% VEU, 20% AGG, thus using three of the holdings from the Ivy Portfolio and keeping the same relative proportions of those three. The yellow line is Vanguard's LifeStrategy Growth fund, which is also 80/20.

Source

Image
To play devil's advocate. While I think he is a rear-view-mirror investor, your backtests are a bit too short. Endowments allocate to real assets for environments that have occurred once in 50 years .. They have famously long horizons; Wall Street (where Cambria has to function), famously short – funds fizzle after 3 years of weakness .. A retail investor is probably somewhere in between, so I think there are lessons to learn.

It frustrates me no end how much pointlessly short-term data PV includes, that make any backtest completely meaningless. Any backtest that doesn't at least include each of the four basic economic environments (ideally in a few sequences) is pointless. It can't tell you anything useful about diversification .. So irritatingly, to get a decent period, I've collapsed Stocks, Int and Real Estate into US Stocks (technically, over long periods, there's no reason US and int shouldn't be expected to perform the same, and Real Estate I think has a similar record to US stocks – although I have my own caveats there), and I've had to use 10yr Treasuries, which makes 60:40 better too. I've also had to use gold instead of commodities – not by choice, but by lacking data. Again: no reason gold should continue to outperform broad commodities. 40 years of falling rates would be a headwind for commodities, so an argument commodities and intl stocks now make more sense.

Btw I'd say 40% in commodities and bonds is a 60:40 portfolio. I think the basic design is a better portfolio for long periods. I also think backtesting is 50% useless, but short-term backtesting is 90% useless.

Image
User avatar
nisiprius
Advisory Board
Posts: 52718
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Faber: T-bills & Chill...Most of the Time

Post by nisiprius »

Logan Roy wrote: Sun May 12, 2024 8:06 am...To play devil's advocate. While I think he is a rear-view-mirror investor, your backtests are a bit too short...
Good backtests prior to the publication of a strategy, or prior to the launch of a mutual fund or an ETF, are a dime a dozen. You could almost say that nobody is going to write about them or commit a fund launch to them unless the backtests are good. So they almost always have publication bias, selection bias, overfitting, p-hacking, etc.

What matters is performance after publication. Of course the time periods are "too short," but they are all that is available that is uncontaminated by hindsight.

It's not my fault that GTAA closed and ceased trading on 5/12/2017. I don't think the allocation algorithm was public. I suppose Mebane Faber is capable of calculating how GTAA would have performed since 2017 if he wanted to, but I don't think he has.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
hudson
Posts: 7336
Joined: Fri Apr 06, 2007 9:15 am

Re: Faber: T-bills & Chill...Most of the Time

Post by hudson »

Corporates and high yield bonds give me heartburn. I vote that fixed income should only be T-s or CDs.

Almost anything that ties decisions to past history gives me heartburn. When I have money to invest, I like to shop around and get the best available deal. If T-Bills are paying 0%, give me TIPS.

I do like T-bills & Chill because that sounds like W. Bernstein. As long as you throw in TIPS, and intermediate T's and CDs.

What would Larry Swedroe or Bill Bernstein say about the article?
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

hudson wrote: Sun May 12, 2024 10:30 am Corporates and high yield bonds give me heartburn. I vote that fixed income should only be T-s or CDs.

Almost anything that ties decisions to past history gives me heartburn. When I have money to invest, I like to shop around and get the best available deal. If T-Bills are paying 0%, give me TIPS.

I do like T-bills & Chill because that sounds like W. Bernstein. As long as you throw in TIPS, and intermediate T's and CDs.

What would Larry Swedroe or Bill Bernstein say about the article?
Directionally, Bernstein confesses to "soft market timing" when it comes to agreeing there are better and worse times to buy bonds.

So I don't think he'd probably with issue with the concept, but might not agree with the specific model.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
rule of law guy
Posts: 673
Joined: Mon Jun 07, 2021 9:35 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by rule of law guy »

Eno Deb wrote: Sat May 11, 2024 12:45 pm
dkturner wrote: Sat May 11, 2024 9:56 am How does one determine the top 50% of historical yield for a particular asset?
I'm not sure if the author means yield or yield spread (he uses both). Yield spread would make more sense to me: only go to long durations if the reward of doing so is significant. Right now we have a historically low (negative) spread with the inverted yield curve, which would suggest to stay away from longer durations.
I agree with this.

while I look at historical values to assess equity market valuation, I am more likely to look at yields comparatively at a single moment in time. for example, with 70% of my portfolio currently in VUSXX, I am assessing when to lengthen maturates at the Fed looks to eventually cut rates. I am in no hurry. but I dont think I will do it until there is a positive yield curve, irrespective of nominal values.
Never wrong, unless my wife tells me that I am.
hudson
Posts: 7336
Joined: Fri Apr 06, 2007 9:15 am

Re: Faber: T-bills & Chill...Most of the Time

Post by hudson »

rule of law guy wrote: Sun May 12, 2024 11:39 am
Eno Deb wrote: Sat May 11, 2024 12:45 pm
dkturner wrote: Sat May 11, 2024 9:56 am How does one determine the top 50% of historical yield for a particular asset?
I'm not sure if the author means yield or yield spread (he uses both). Yield spread would make more sense to me: only go to long durations if the reward of doing so is significant. Right now we have a historically low (negative) spread with the inverted yield curve, which would suggest to stay away from longer durations.
I agree with this.

while I look at historical values to assess equity market valuation, I am more likely to look at yields comparatively at a single moment in time. for example, with 70% of my portfolio currently in VUSXX, I am assessing when to lengthen maturates at the Fed looks to eventually cut rates. I am in no hurry. but I dont think I will do it until there is a positive yield curve, irrespective of nominal values.
VUSXX...Vanguard Treasury Money Market Fund...5.3% dividend on May 1.
Nice! ...at least for now.
rule of law guy
Posts: 673
Joined: Mon Jun 07, 2021 9:35 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by rule of law guy »

"VUSXX...Vanguard Treasury Money Market Fund...5.3% dividend on May 1.
Nice! ...at least for now.

with both 2 yr and 10 yr yields below where the Fed takes fed funds to with its expected initial 25 bps cut, whenever that is, I foresee "now" for at least 9-12 months
Never wrong, unless my wife tells me that I am.
Logan Roy
Posts: 2016
Joined: Sun May 29, 2022 10:15 am

Re: Faber: T-bills & Chill...Most of the Time

Post by Logan Roy »

nisiprius wrote: Sun May 12, 2024 10:03 am
Logan Roy wrote: Sun May 12, 2024 8:06 am...To play devil's advocate. While I think he is a rear-view-mirror investor, your backtests are a bit too short...
Good backtests prior to the publication of a strategy, or prior to the launch of a mutual fund or an ETF, are a dime a dozen. You could almost say that nobody is going to write about them or commit a fund launch to them unless the backtests are good. So they almost always have publication bias, selection bias, overfitting, p-hacking, etc.

What matters is performance after publication. Of course the time periods are "too short," but they are all that is available that is uncontaminated by hindsight.

It's not my fault that GTAA closed and ceased trading on 5/12/2017. I don't think the allocation algorithm was public. I suppose Mebane Faber is capable of calculating how GTAA would have performed since 2017 if he wanted to, but I don't think he has.
After publication's definitely a different hurdle, but it's also a total crapshoot over such a short period .. Post-publication performance could be completely in line with expectations, yet still look like it's losing against a market riding on luck and momentum.

I'd put more weight on a good 50 year backtest than a great 10 year forward-test.
User avatar
Charles Joseph
Posts: 2737
Joined: Tue Apr 05, 2022 10:49 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by Charles Joseph »

watchnerd wrote: Fri May 10, 2024 5:49 pm In this article, Feb outlines a bond-buying timing strategy that claims it doesn't require any guesses about the future, and improves risk-adjusted returns.

1. Hold T-bills by default.
2. When bonds yields are in the top 50%, historically, buy them
An unnecessarily complicated scheme from a guy who is, well, unnecessarily complicated in general.

Forget it.
Retired June 2023
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

Charles Joseph wrote: Sun May 12, 2024 1:05 pm
watchnerd wrote: Fri May 10, 2024 5:49 pm In this article, Feb outlines a bond-buying timing strategy that claims it doesn't require any guesses about the future, and improves risk-adjusted returns.

1. Hold T-bills by default.
2. When bonds yields are in the top 50%, historically, buy them
An unnecessarily complicated scheme from a guy who is, well, unnecessarily complicated in general.

Forget it.
That's complicated?

Hold T-bills, otherwise buy bonds when they have good yields better than T-bills?
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
Charles Joseph
Posts: 2737
Joined: Tue Apr 05, 2022 10:49 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by Charles Joseph »

watchnerd wrote: Sun May 12, 2024 1:09 pm
Charles Joseph wrote: Sun May 12, 2024 1:05 pm
watchnerd wrote: Fri May 10, 2024 5:49 pm In this article, Feb outlines a bond-buying timing strategy that claims it doesn't require any guesses about the future, and improves risk-adjusted returns.

1. Hold T-bills by default.
2. When bonds yields are in the top 50%, historically, buy them
An unnecessarily complicated scheme from a guy who is, well, unnecessarily complicated in general.

Forget it.
That's complicated?

Hold T-bills, otherwise buy bonds when they have good yields better than T-bills?
Yes. That is unnecessarily complicated.

Yes.
Retired June 2023
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

Charles Joseph wrote: Sun May 12, 2024 1:12 pm
watchnerd wrote: Sun May 12, 2024 1:09 pm
Charles Joseph wrote: Sun May 12, 2024 1:05 pm
watchnerd wrote: Fri May 10, 2024 5:49 pm In this article, Feb outlines a bond-buying timing strategy that claims it doesn't require any guesses about the future, and improves risk-adjusted returns.

1. Hold T-bills by default.
2. When bonds yields are in the top 50%, historically, buy them
An unnecessarily complicated scheme from a guy who is, well, unnecessarily complicated in general.

Forget it.
That's complicated?

Hold T-bills, otherwise buy bonds when they have good yields better than T-bills?
Yes. That is unnecessarily complicated.

Yes.
Complicated compared to what alternative?

The only thing that makes it more complicated than buying and holding T-bills + bond fund is you're being a little picky about when you buy the bonds.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Logan Roy
Posts: 2016
Joined: Sun May 29, 2022 10:15 am

Re: Faber: T-bills & Chill...Most of the Time

Post by Logan Roy »

I do think the past 15 years have demonstrated why you don't want to buy bonds 'at any cost' .. Tying capital up for 10 (20, 30) years in things that will return nothing doesn't make sense just because we're pursuing a 'passive' approach.

At worst, by having your capital tied up (or suffering large losses), while inflation destroys the value of those assets, you're actually footing the bill for all that stimulus (they're inflating their debt away, and in the process inflating your investments away). We talk about 'greater fool' in crypto and gold, but in this case you're buying truly worthless bonds off traders who've been front-running fed policy.

So you need a way of assessing value, and then you certainly don't want to be long duration in assets that offer terrible value.
User avatar
Charles Joseph
Posts: 2737
Joined: Tue Apr 05, 2022 10:49 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by Charles Joseph »

watchnerd wrote: Sun May 12, 2024 1:14 pm
Charles Joseph wrote: Sun May 12, 2024 1:12 pm
watchnerd wrote: Sun May 12, 2024 1:09 pm
Charles Joseph wrote: Sun May 12, 2024 1:05 pm
watchnerd wrote: Fri May 10, 2024 5:49 pm In this article, Feb outlines a bond-buying timing strategy that claims it doesn't require any guesses about the future, and improves risk-adjusted returns.

1. Hold T-bills by default.
2. When bonds yields are in the top 50%, historically, buy them
An unnecessarily complicated scheme from a guy who is, well, unnecessarily complicated in general.

Forget it.
That's complicated?

Hold T-bills, otherwise buy bonds when they have good yields better than T-bills?
Yes. That is unnecessarily complicated.

Yes.
Complicated compared to what alternative?

The only thing that makes it more complicated than buying and holding T-bills + bond fund is you're being a little picky about when you buy the bonds.
Complicated, unnecessary and, I should add, almost certainly suboptimal.

Suboptimal because longer-maturity bonds have under most circumstances outperformed T-Bills. I could provide documentation for that claim but I'm going out to dinner with my family. And besides, I just don't feel like it.

But I don't have to provide the data because, well, I'm right.
Retired June 2023
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

Charles Joseph wrote: Sun May 12, 2024 1:28 pm Complicated, unnecessary and, I should add, almost certainly suboptimal.

Suboptimal because longer-maturity bonds have under most circumstances outperformed T-Bills. I could provide documentation for that claim but I'm going out to dinner with my family. And besides, I just don't feel like it.

But I don't have to provide the data because, well, I'm right.
Complicated is subjective, but please point to the data that shows it is suboptimal?

The data in the article refutes your take on suboptimal. The modeled ports mostly had:

--Higher returns
--Higher Sharpe ratios
--Lower Std Dev
--Smaller draw downs

The backtests in the article would challenge your assertion that you are right.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
Kevin M
Posts: 15881
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Faber: T-bills & Chill...Most of the Time

Post by Kevin M »

Eno Deb wrote: Sat May 11, 2024 9:07 pm Personally I don't trust TIPS.
This seems to me analogous to saying, "I don't trust bonds". TIPS are simply bonds with a twist (inflation indexing), and they perform exactly according to bond math as they are expected to. Not trusting TIPS, or bonds, seems to be a function of not understanding them.
Historically there were long stretches of time where they significantly underperformed treasuries of similar duration even in real terms, <snip>
In my first reply, I discussed three criticisms of the article that is the subject of this thread, none of which addressed the relative historical performance of TIPS and nominal bonds. The one criticism that perhaps is relevant to this comment is that risk is related to uncertainty of realized return relative to expected return for a specified holding period and for the relevant unit of account.

A common understanding in investing is that expected return and risk are related. Most of us believe that stocks have a higher expected return than bonds, but the risk is that the realized return is likely to diverge more from the expected return for the asset with the higher expected return. Most of us who buy individual TIPS are looking for the highest possible certainty for a term equal to the maturity of the TIPS, with real consumption as the unit of account. Nominal Treasuries simply don't give us that, and especially for longer holding periods when it comes to Tbills, which the article was promoting when yields spreads are historically low, as they are now.
<snip> then we had negative yields, <snip>

Right, and one of my points was that at times when we had negative yields, the yield spread of TIPS over 3m bills was smaller than it is now, when we have historically attractive yields. So perhaps the tactical timing model proposed by the article is not so good for TIPS.
<snip>and during the recent inflation spike they utterly failed since the price decline caused by the rising interest rates more than offset the principal adjustment.<snip>
I have an example that disproves this, at least in comparing nominals to TIPS, and I have shared this multiple times in the forum. In mid-June 2021, I bought some 2-year TIPS and 2-year nominal Treasuries (both maturing 7/15/2023). When they matured in July 2023, the total cumulative return of the TIPS was about 8.6%, while that for the nominal Treasuries was about 0.3%; that's quite an advantage for the TIPS.

Now, when I bought them, the TIPS real yield was -2.89% and the nominal yield was 0.16%. So of course the realized real return was negative, and close to the original real yield, and the annualized return of the nominal was close to the original nominal yield. The large difference in nominal returns was because of the TIPS inflation return, which was a cumulative 14.4%.
As far as I'm concerned TIPS haven't demonstrated yet that they are a better inflation hedge than stocks (or even T-Bills, as discussed above).
The article we are discussing has nothing to do with stocks, and the example above demonstrates a case where TIPS clearly were a better inflation hedge than Tbills.

Although stocks may have at times been a decent inflation hedge, at other times they have not. With individual TIPS, held to maturity, we have high certainty about how they will perform relative to inflation, and that's what most of us are looking for with our TIPS.
If I make a calculation error, #Cruncher probably will let me know.
Eno Deb
Posts: 798
Joined: Sun Feb 03, 2019 3:08 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by Eno Deb »

Kevin M wrote: Sun May 12, 2024 3:16 pmThis seems to me analogous to saying, "I don't trust bonds". TIPS are simply bonds with a twist (inflation indexing), and they perform exactly according to bond math as they are expected to. Not trusting TIPS, or bonds, seems to be a function of not understanding them.
Or maybe a recognition that they underperform in most scenarios. If you want to use them as insurance to meet a specific need at a specific time that's fine, but that is not my goal and not what the author of the paper investigated. I simply want reasonable performance out of my bond allocation.
The article we are discussing has nothing to do with stocks
It also has nothing to do with inflation hedging. That was brought up by commenters in this thread.
User avatar
Kevin M
Posts: 15881
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Faber: T-bills & Chill...Most of the Time

Post by Kevin M »

Eno Deb wrote: Sun May 12, 2024 7:14 pm
Kevin M wrote: Sun May 12, 2024 3:16 pmThis seems to me analogous to saying, "I don't trust bonds". TIPS are simply bonds with a twist (inflation indexing), and they perform exactly according to bond math as they are expected to. Not trusting TIPS, or bonds, seems to be a function of not understanding them.
Or maybe a recognition that they underperform in most scenarios.
Really?

Image

I honestly didn't know what this chart would show when I created it, as I'm not too interested in bond funds at this point, so I haven't been doing these kinds of comparisons in quite awhile.
Eno Deb wrote: Sun May 12, 2024 7:14 pm
The article we are discussing has nothing to do with stocks
It also has nothing to do with inflation hedging. That was brought up by commenters in this thread.
Ah, but it did include TIPS in its analysis!
If I make a calculation error, #Cruncher probably will let me know.
Eno Deb
Posts: 798
Joined: Sun Feb 03, 2019 3:08 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by Eno Deb »

Kevin M wrote: Sun May 12, 2024 8:35 pmReally?
Yes, really. I checked a bunch of scenarios a few years back. Of course you can always cherry pick a scenario where TIPS come out on top.
Ah, but it did include TIPS in its analysis!
Only in a portfolio mixed together with treasuries, REITs, junk bonds, mortgage backed and emerging market bonds. The word inflation does not appear in the report.
User avatar
Kevin M
Posts: 15881
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Faber: T-bills & Chill...Most of the Time

Post by Kevin M »

Eno Deb wrote: Sun May 12, 2024 8:57 pm
Kevin M wrote: Sun May 12, 2024 8:35 pmReally?
Yes, really. I checked a bunch of scenarios a few years back. Of course you can always cherry pick a scenario where TIPS come out on top.
The PV chart was not based on cherry picking, but on the longest history available for TIPS in PV. This highlights another point we should keep in mind, which is that we don't have enough history to make valid statistical conclusions for many asset classes, with TIPS having perhaps the least history of all of them. This is relevant to one of my points about the article, which is that it reeked of data mining.

One thing I might do when I have more time than I have now is to create a telltale chart based on the TIPS vs. Intermediate-term Treasuries. I consider the telltale chart, which John Bogle introduced many of us to, to be more informative than simple historical comparison charts. We still don't have much history, but at least this would show us the periods when either one was outperforming the other.
Eno Deb wrote: Sun May 12, 2024 8:57 pm
Ah, but it did include TIPS in its analysis!
Only in a portfolio mixed together with treasuries, REITs, junk bonds, mortgage backed and emerging market bonds. The word inflation does not appear in the report.
The "I" in TIPS stands for "Inflation". I really don't see your point here.

Stepping back, several of us have suggested that "Tbills and Chill most of the time" may not be a good tactical timing approach for buying or holding TIPS. You're correct that none of the analyses addressed just Tbills and TIPS, but only TIPS along with other bond categories. This could be considered further evidence that the article doesn't provide any evidence that its tactical timing approach is valid for TIPS.

Remember that this thread is not about TIPS, but about an article that suggested a specific tactical timing approach for when to hold only Tbills or when to hold other bond categories (including TIPS).
If I make a calculation error, #Cruncher probably will let me know.
comeinvest
Posts: 2778
Joined: Mon Mar 12, 2012 6:57 pm

Re: Faber: T-bills & Chill...Most of the Time

Post by comeinvest »

Logan Roy wrote: Sun May 12, 2024 1:17 pm I do think the past 15 years have demonstrated why you don't want to buy bonds 'at any cost' .. Tying capital up for 10 (20, 30) years in things that will return nothing doesn't make sense just because we're pursuing a 'passive' approach.

At worst, by having your capital tied up (or suffering large losses), while inflation destroys the value of those assets, you're actually footing the bill for all that stimulus (they're inflating their debt away, and in the process inflating your investments away). We talk about 'greater fool' in crypto and gold, but in this case you're buying truly worthless bonds off traders who've been front-running fed policy.

So you need a way of assessing value, and then you certainly don't want to be long duration in assets that offer terrible value.
Nobody knew end of 2021 that treasury yields wouldn't keep sinking to zero or below, following international yields. It was an accepted theory not only by media but also by rigorous academic studies, with very reasonable arguments and macroeconomic models.
Perhaps in 2026 long term treasury yields will be 1.5% again. You ignore them based on your backtesting theory. Between 2025 and 2035, the stock market crashes and treasuries go to -1% (similar to Europe, Japan between the GFC and 2021). By 2045, the stock market recovered, and treasury yields go back to 3%. You lost out on diversification benefits along with juicy rebalancing returns and yield curve rolldown returns for the rest of your life, by ignoring bonds based on some mean-reversion assumption with backtested ranges.
Having that said, I too might sell all my treasuries next time yields are below 1.5% ;)
Last edited by comeinvest on Tue May 14, 2024 1:00 am, edited 1 time in total.
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

comeinvest wrote: Tue May 14, 2024 12:56 am
Nobody knew end of 2021 that treasury yields wouldn't keep sinking to zero or below, following international yields.
Eh, I dunno...

Seemed pretty clear by November 2021 that the Fed was going to pivot based on their public statements and their change in tone regarding inflation not being just 'transitory'.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
Topic Author
watchnerd
Posts: 14292
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Faber: T-bills & Chill...Most of the Time

Post by watchnerd »

comeinvest wrote: Tue May 14, 2024 12:56 am Between 2025 and 2035, the stock market crashes and treasuries go to -1% (similar to Europe, Japan between the GFC and 2021).
I certainly won't be buying any new Treasuries if yields go to -1%.

Especially not when I already have TIPS at positive real yields covering 2025 to 2034.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Post Reply