Why not 100% TMF?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
Booogle
Posts: 605
Joined: Wed Nov 09, 2016 2:57 pm

Why not 100% TMF?

Post by Booogle »

https://www.portfoliovisualizer.com/bac ... on3_2=-200

Even if you play around with the time periods, its pretty good.
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Why not 100% TMF?

Post by 000 »

Are you aware we have been in a secular declining rates environment for 39 years?
Topic Author
Booogle
Posts: 605
Joined: Wed Nov 09, 2016 2:57 pm

Re: Why not 100% TMF?

Post by Booogle »

000 wrote: Sun Nov 28, 2021 5:07 pm Are you aware we have been in a secular declining rates environment for 39 years?

That doesn't really matter:

https://portfoliocharts.com/2019/05/27/ ... convexity/
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Why not 100% TMF?

Post by 000 »

Your own link shows that at low yields even a 1% rise in rates will have deep negative returns on unlevered long term bonds:

Image
Topic Author
Booogle
Posts: 605
Joined: Wed Nov 09, 2016 2:57 pm

Re: Why not 100% TMF?

Post by Booogle »

000 wrote: Sun Nov 28, 2021 5:17 pm Your own link shows that at low yields even a 1% rise in rates will have deep negative returns on unlevered long term bonds:
If you don't want to actually read the article, don't waste everyone's time.
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Why not 100% TMF?

Post by 000 »

Booogle wrote: Sun Nov 28, 2021 5:20 pm
000 wrote: Sun Nov 28, 2021 5:17 pm Your own link shows that at low yields even a 1% rise in rates will have deep negative returns on unlevered long term bonds:
If you don't want to actually read the article, don't waste everyone's time.
I did read the article and it does not claim that leveraged long bonds are a free lunch under all market conditions.
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Why not 100% TMF?

Post by 000 »

Anyway, look at how it has done YTD: https://www.portfoliovisualizer.com/bac ... ion4_3=100

Almost 20% loss and we haven't even see a sustained multi year rising rates environment yet.
User avatar
nisiprius
Advisory Board
Posts: 52211
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: Why not 100% TMF?

Post by nisiprius »

Also, I don't want to overstate this, but it's worth noting.

Over the limited period of time TMF has been available, Booogle's "simulated TMF" has not been wildly different from the real TMF. But it hasn't been a high-precision simulation, either. It's been almost 3%/year overoptimistic compared to the real TMF. The real TMF has had both lower return and higher volatility than the simulation.

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.
Soysauceonrice
Posts: 55
Joined: Thu Aug 05, 2021 9:39 pm

Re: Why not 100% TMF?

Post by Soysauceonrice »

nisiprius wrote: Sun Nov 28, 2021 7:36 pm Also, I don't want to overstate this, but it's worth noting.

Over the limited period of time TMF has been available, Booogle's "simulated TMF" has not been wildly different from the real TMF. But it hasn't been a high-precision simulation, either. It's been almost 3%/year overoptimistic compared to the real TMF. The real TMF has had both lower return and higher volatility than the simulation.

source

Image
Interesting. I wonder if the difference in performance has anything to do with the difference in expense ratios. He simulated TMF with VUSTX, which has an expense ratio of 0.20%. TMF is 5x more expensive, at 1.06%
User avatar
typical.investor
Posts: 5263
Joined: Mon Jun 11, 2018 3:17 am

Re: Why not 100% TMF?

Post by typical.investor »

Booogle wrote: Sun Nov 28, 2021 5:02 pm https://www.portfoliovisualizer.com/bac ... on3_2=-200

Even if you play around with the time periods, its pretty good.
My $0.02 is:

1) you should consider not only performance over a particular period, but also what you know about the investment vehicle
-and-
2) how performance in the period you are looking at may change

So, I think the answer is 1) the LEFT have the potential to have returns wildly different than the multiple of the index. Just because we haven't seen in this long period of falling rates, it doesn't mean we won't see it in the future. And 2) there is always the possibility that we enter a period of rising rates. Demographics argues against that, but a decoupling of China from the global economic system would likely be inflationary.

In short, nothing about those past returns is set in future stone and due to that risk, one would be better with more reasonable diversification.
bgf
Posts: 2085
Joined: Fri Nov 10, 2017 8:35 am

Re: Why not 100% TMF?

Post by bgf »

I put my money on rates continuing their centuries long trend downward. Rates in the 80s and 90s were the blip, not the norm.

At the same time, there's no reason that unforeseen jumps in rates can't happen, or even persist for substantial periods. This scares me from going "all in" on such a bet. Just like people should be scared of going "all in" on 3x the sp500, even though we bet that it will otherwise trend upwards over time.

It's just really hard to make long term concentrated bets. It's possible, but improbable.
“TE OCCIDERE POSSUNT SED TE EDERE NON POSSUNT NEFAS EST"
rockstar
Posts: 6326
Joined: Mon Feb 03, 2020 5:51 pm

Re: Why not 100% TMF?

Post by rockstar »

000 wrote: Sun Nov 28, 2021 5:17 pm Your own link shows that at low yields even a 1% rise in rates will have deep negative returns on unlevered long term bonds:

Image
My take away from this often used chart is that the lower the rate the higher the volatility. Usually high volatility is not good for leveraged funds.
User avatar
Beensabu
Posts: 5657
Joined: Sun Aug 14, 2016 3:22 pm

Re: Why not 100% TMF?

Post by Beensabu »

Good lord.

I wrote a full on outraged rant in reply to this and then deleted it in order to avoid a board warning, coz... I'm not exactly sure how many of those we get...

So I leave you with:

Do we understand what diversification means? Do we? Do we, really? Like what is it? What do you think it is? I'll tell you what it's not. It's not 100% anything. It's especially not 100% leveraged anything. It's just not.

Sometimes, you find interesting information. That doesn't mean go 3x all in with it. No. No, no, no, no, no, no. No. Also, no.
"The only thing that makes life possible is permanent, intolerable uncertainty; not knowing what comes next." ~Ursula LeGuin
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Why not 100% TMF?

Post by 000 »

rockstar wrote: Sun Nov 28, 2021 8:28 pm My take away from this often used chart is that the lower the rate the higher the volatility. Usually high volatility is not good for leveraged funds.
Exactly.

And it makes perfect sense.

1% of 2% is 50%

1% of 5% is 20%

Just saying "bond convexity" does not logically lead to a 100% TMF portfolio.

That said, I find it impressive how well TMF has held up over long time periods and it may very well be a good trade right now.
User avatar
Ocean77
Posts: 800
Joined: Wed Oct 23, 2019 3:20 pm

Re: Why not 100% TMF?

Post by Ocean77 »

It's actually 100% RVM Investing. (Rear View Mirror)
30% US Stocks | 30% Int Stocks | 40% Bonds
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Why not 100% TMF?

Post by 000 »

nisiprius wrote: Sun Nov 28, 2021 7:36 pm Also, I don't want to overstate this, but it's worth noting.

Over the limited period of time TMF has been available, Booogle's "simulated TMF" has not been wildly different from the real TMF. But it hasn't been a high-precision simulation, either. It's been almost 3%/year overoptimistic compared to the real TMF. The real TMF has had both lower return and higher volatility than the simulation.
Isn't it also interesting that from my YTD link above, TMF has returned 3% less in absolute value this year?

Almost like the daily reset on this fund is reducing volatility decay?? :o
DMoogle
Posts: 549
Joined: Sat Oct 31, 2020 10:24 pm

Re: Why not 100% TMF?

Post by DMoogle »

Soysauceonrice wrote: Sun Nov 28, 2021 8:03 pmInteresting. I wonder if the difference in performance has anything to do with the difference in expense ratios. He simulated TMF with VUSTX, which has an expense ratio of 0.20%. TMF is 5x more expensive, at 1.06%
That's partially it. Here are some other factors:
  • This backtest is rebalanced quarterly, whereas TMF is rebalanced daily.
  • TMF uses swaps primarily, IIRC, which will have embedded costs. CASHX is the comparison here, but it's certainly not 1:1.
  • TMF just isn't a perfect replication of 3x the underlying, even daily. Essentially, it isn't that good at doing what it aims to.
There's a much more legit TMF simulation in the Hedgefundie thread that was pulled from another thread that had a ton of smart minds collaborate on it, but even they weren't able to get a perfect simulation. OP should look at that simulation to see the long-term results of a TMF-like fund.
Topic Author
Booogle
Posts: 605
Joined: Wed Nov 09, 2016 2:57 pm

Re: Why not 100% TMF?

Post by Booogle »

DMoogle wrote: Sun Nov 28, 2021 9:11 pm
Soysauceonrice wrote: Sun Nov 28, 2021 8:03 pmInteresting. I wonder if the difference in performance has anything to do with the difference in expense ratios. He simulated TMF with VUSTX, which has an expense ratio of 0.20%. TMF is 5x more expensive, at 1.06%
That's partially it. Here are some other factors:
  • This backtest is rebalanced quarterly, whereas TMF is rebalanced daily.
  • TMF uses swaps primarily, IIRC, which will have embedded costs. CASHX is the comparison here, but it's certainly not 1:1.
  • TMF just isn't a perfect replication of 3x the underlying, even daily. Essentially, it isn't that good at doing what it aims to.
There's a much more legit TMF simulation in the Hedgefundie thread that was pulled from another thread that had a ton of smart minds collaborate on it, but even they weren't able to get a perfect simulation. OP should look at that simulation to see the long-term results of a TMF-like fund.
Can you link to that more precise simulation?
User avatar
firebirdparts
Posts: 4411
Joined: Thu Jun 13, 2019 4:21 pm
Location: Southern Appalachia

Re: Why not 100% TMF?

Post by firebirdparts »

If you have a time machine, this is a great strategy. Without it, risky.
This time is the same
DMoogle
Posts: 549
Joined: Sat Oct 31, 2020 10:24 pm

Re: Why not 100% TMF?

Post by DMoogle »

Booogle wrote: Mon Nov 29, 2021 6:05 amCan you link to that more precise simulation?
viewtopic.php?f=10&t=272007&start=1050#p4426310

Might need to ask some folks for the specific data file if you want it, but the results of the sim are there (scroll down, there are several sections on TMF).
User avatar
samsoes
Posts: 2802
Joined: Tue Mar 05, 2013 8:12 am
Location: Northeast Rat Race

Re: Why not 100% TMF?

Post by samsoes »

What's TMF?
"Happiness Is Not My Companion" - Gen. Gouverneur K. Warren. | (Avatar is the statue of Gen. Warren atop Little Round Top @ Gettysburg National Military Park.)
User avatar
anon_investor
Posts: 15122
Joined: Mon Jun 03, 2019 1:43 pm

Re: Why not 100% TMF?

Post by anon_investor »

samsoes wrote: Mon Nov 29, 2021 8:12 amWhat's TMF?
A triple leveraged 20 year US treasury ETF. Not very BH...
Topic Author
Booogle
Posts: 605
Joined: Wed Nov 09, 2016 2:57 pm

Re: Why not 100% TMF?

Post by Booogle »

anon_investor wrote: Mon Nov 29, 2021 8:20 am
samsoes wrote: Mon Nov 29, 2021 8:12 amWhat's TMF?
A triple leveraged 20 year US treasury ETF. Not very BH...
90% VTSAX, 10% TMF would be pretty BH

And that you can backtest.

It's gold.
KlangFool
Posts: 31525
Joined: Sat Oct 11, 2008 12:35 pm

Re: Why not 100% TMF?

Post by KlangFool »

OP,

It is very simple.

If you can predict the future, you can go with 100% of anything.

KlangFool
30% VWENX | 16% VFWAX/VTIAX | 14.5% VTSAX | 19.5% VBTLX | 10% VSIAX/VTMSX/VSMAX | 10% VSIGX| 30% Wellington 50% 3-funds 20% Mini-Larry
User avatar
anon_investor
Posts: 15122
Joined: Mon Jun 03, 2019 1:43 pm

Re: Why not 100% TMF?

Post by anon_investor »

Booogle wrote: Mon Nov 29, 2021 8:52 am
anon_investor wrote: Mon Nov 29, 2021 8:20 am
samsoes wrote: Mon Nov 29, 2021 8:12 amWhat's TMF?
A triple leveraged 20 year US treasury ETF. Not very BH...
90% VTSAX, 10% TMF would be pretty BH

And that you can backtest.

It's gold.
Wouldn't NTSX be easier? I feel like using 90% VTSAX, 10% TMF would require a lot of attentiveness and rebalancing to be effective.
Topic Author
Booogle
Posts: 605
Joined: Wed Nov 09, 2016 2:57 pm

Re: Why not 100% TMF?

Post by Booogle »

anon_investor wrote: Mon Nov 29, 2021 9:04 am Wouldn't NTSX be easier? I feel like using 90% VTSAX, 10% TMF would require a lot of attentiveness and rebalancing to be effective.
Yes, but what about all the people who bought equities 10 years before NTSX existed?
User avatar
anon_investor
Posts: 15122
Joined: Mon Jun 03, 2019 1:43 pm

Re: Why not 100% TMF?

Post by anon_investor »

Booogle wrote: Mon Nov 29, 2021 9:38 am
anon_investor wrote: Mon Nov 29, 2021 9:04 am Wouldn't NTSX be easier? I feel like using 90% VTSAX, 10% TMF would require a lot of attentiveness and rebalancing to be effective.
Yes, but what about all the people who bought equities 10 years before NTSX existed?
TMF has too much leverage for most. Why not buy UPRO instead of VTSAX/VTI? Hedgefundie's excellent adventure.
Topic Author
Booogle
Posts: 605
Joined: Wed Nov 09, 2016 2:57 pm

Re: Why not 100% TMF?

Post by Booogle »

anon_investor wrote: Mon Nov 29, 2021 10:18 am
Booogle wrote: Mon Nov 29, 2021 9:38 am
anon_investor wrote: Mon Nov 29, 2021 9:04 am Wouldn't NTSX be easier? I feel like using 90% VTSAX, 10% TMF would require a lot of attentiveness and rebalancing to be effective.
Yes, but what about all the people who bought equities 10 years before NTSX existed?
TMF has too much leverage for most. Why not buy UPRO instead of VTSAX/VTI? Hedgefundie's excellent adventure.
You can do whatever you want.

I'm just saying 90% VTSAX, 10% TMF is pretty good.
prioritarian
Posts: 485
Joined: Tue Jul 16, 2019 6:00 pm

Re: Why not 100% TMF?

Post by prioritarian »

Booogle wrote: Mon Nov 29, 2021 9:38 am
anon_investor wrote: Mon Nov 29, 2021 9:04 am Wouldn't NTSX be easier? I feel like using 90% VTSAX, 10% TMF would require a lot of attentiveness and rebalancing to be effective.
Yes, but what about all the people who bought equities 10 years before NTSX existed?
This is definitely my dilemma. I have so much money locked into equities (e.g. >200% gains) that I will likely never rebalance. Adding some TMF would NTSXify them. Unfortunately, the high ER of TMF, volatility decay, and derivative-based counterparty risk have prevented me from using this method of diversify these holdings. If a TMF-like instrument with lower ER and less risk becomes available I will likely use it. (I don't have the spare time to roll futures so I need something easy to use.)
User avatar
anon_investor
Posts: 15122
Joined: Mon Jun 03, 2019 1:43 pm

Re: Why not 100% TMF?

Post by anon_investor »

prioritarian wrote: Mon Nov 29, 2021 12:23 pm
Booogle wrote: Mon Nov 29, 2021 9:38 am
anon_investor wrote: Mon Nov 29, 2021 9:04 am Wouldn't NTSX be easier? I feel like using 90% VTSAX, 10% TMF would require a lot of attentiveness and rebalancing to be effective.
Yes, but what about all the people who bought equities 10 years before NTSX existed?
This is definitely my dilemma. I have so much money locked into equities (e.g. >200% gains) that I will likely never rebalance. Adding some TMF would NTSXify them. Unfortunately, the high ER of TMF, volatility decay, and derivative-based counterparty risk have prevented me from using this method of diversify these holdings. If a TMF-like instrument with lower ER and less risk becomes available I will likely use it. (I don't have the spare time to roll futures so I need something easy to use.)
Why not plain old EDV?
User avatar
anon_investor
Posts: 15122
Joined: Mon Jun 03, 2019 1:43 pm

Re: Why not 100% TMF?

Post by anon_investor »

Booogle wrote: Mon Nov 29, 2021 10:21 am
anon_investor wrote: Mon Nov 29, 2021 10:18 am
Booogle wrote: Mon Nov 29, 2021 9:38 am
anon_investor wrote: Mon Nov 29, 2021 9:04 am Wouldn't NTSX be easier? I feel like using 90% VTSAX, 10% TMF would require a lot of attentiveness and rebalancing to be effective.
Yes, but what about all the people who bought equities 10 years before NTSX existed?
TMF has too much leverage for most. Why not buy UPRO instead of VTSAX/VTI? Hedgefundie's excellent adventure.
You can do whatever you want.

I'm just saying 90% VTSAX, 10% TMF is pretty good.
Is the triple leverage really necessary? EDV is pretty effective for a low ER.
prioritarian
Posts: 485
Joined: Tue Jul 16, 2019 6:00 pm

Re: Why not 100% TMF?

Post by prioritarian »

anon_investor wrote: Mon Nov 29, 2021 12:38 pm
prioritarian wrote: Mon Nov 29, 2021 12:23 pm
Booogle wrote: Mon Nov 29, 2021 9:38 am
anon_investor wrote: Mon Nov 29, 2021 9:04 am Wouldn't NTSX be easier? I feel like using 90% VTSAX, 10% TMF would require a lot of attentiveness and rebalancing to be effective.
Yes, but what about all the people who bought equities 10 years before NTSX existed?
This is definitely my dilemma. I have so much money locked into equities (e.g. >200% gains) that I will likely never rebalance. Adding some TMF would NTSXify them. Unfortunately, the high ER of TMF, volatility decay, and derivative-based counterparty risk have prevented me from using this method of diversify these holdings. If a TMF-like instrument with lower ER and less risk becomes available I will likely use it. (I don't have the spare time to roll futures so I need something easy to use.)
Why not plain old EDV?
I prefer VGLT in taxable for a variety of reasons but I should probably use EDV. (I've been considering this for some time.)
User avatar
anon_investor
Posts: 15122
Joined: Mon Jun 03, 2019 1:43 pm

Re: Why not 100% TMF?

Post by anon_investor »

prioritarian wrote: Mon Nov 29, 2021 2:58 pm
anon_investor wrote: Mon Nov 29, 2021 12:38 pm
prioritarian wrote: Mon Nov 29, 2021 12:23 pm
Booogle wrote: Mon Nov 29, 2021 9:38 am
anon_investor wrote: Mon Nov 29, 2021 9:04 am Wouldn't NTSX be easier? I feel like using 90% VTSAX, 10% TMF would require a lot of attentiveness and rebalancing to be effective.
Yes, but what about all the people who bought equities 10 years before NTSX existed?
This is definitely my dilemma. I have so much money locked into equities (e.g. >200% gains) that I will likely never rebalance. Adding some TMF would NTSXify them. Unfortunately, the high ER of TMF, volatility decay, and derivative-based counterparty risk have prevented me from using this method of diversify these holdings. If a TMF-like instrument with lower ER and less risk becomes available I will likely use it. (I don't have the spare time to roll futures so I need something easy to use.)
Why not plain old EDV?
I prefer VGLT in taxable for a variety of reasons but I should probably use EDV. (I've been considering this for some time.)
Both VGLT and EDV are likely going to have capital gain distributions, that is just the nature of longer term bond funds, they have to constantly churn to maintain the long duration.
prioritarian
Posts: 485
Joined: Tue Jul 16, 2019 6:00 pm

Re: Why not 100% TMF?

Post by prioritarian »

anon_investor wrote: Mon Nov 29, 2021 3:13 pm
prioritarian wrote: Mon Nov 29, 2021 2:58 pm
anon_investor wrote: Mon Nov 29, 2021 12:38 pm
prioritarian wrote: Mon Nov 29, 2021 12:23 pm
Booogle wrote: Mon Nov 29, 2021 9:38 am
Yes, but what about all the people who bought equities 10 years before NTSX existed?
This is definitely my dilemma. I have so much money locked into equities (e.g. >200% gains) that I will likely never rebalance. Adding some TMF would NTSXify them. Unfortunately, the high ER of TMF, volatility decay, and derivative-based counterparty risk have prevented me from using this method of diversify these holdings. If a TMF-like instrument with lower ER and less risk becomes available I will likely use it. (I don't have the spare time to roll futures so I need something easy to use.)
Why not plain old EDV?
I prefer VGLT in taxable for a variety of reasons but I should probably use EDV. (I've been considering this for some time.)
Both VGLT and EDV are likely going to have capital gain distributions, that is just the nature of longer term bond funds, they have to constantly churn to maintain the long duration.
Yes, EDV has had higher cap gain distributions than VGLT (and is more volatile with only a modest increase in CAGR). I intend to use low cap gains equities in taxable for early retirement income and high cap gains positions for charitable contributions so the "bonds in tax deferred" strategy is not optimal for me.
User avatar
anon_investor
Posts: 15122
Joined: Mon Jun 03, 2019 1:43 pm

Re: Why not 100% TMF?

Post by anon_investor »

prioritarian wrote: Mon Nov 29, 2021 3:25 pm
anon_investor wrote: Mon Nov 29, 2021 3:13 pm
prioritarian wrote: Mon Nov 29, 2021 2:58 pm
anon_investor wrote: Mon Nov 29, 2021 12:38 pm
prioritarian wrote: Mon Nov 29, 2021 12:23 pm

This is definitely my dilemma. I have so much money locked into equities (e.g. >200% gains) that I will likely never rebalance. Adding some TMF would NTSXify them. Unfortunately, the high ER of TMF, volatility decay, and derivative-based counterparty risk have prevented me from using this method of diversify these holdings. If a TMF-like instrument with lower ER and less risk becomes available I will likely use it. (I don't have the spare time to roll futures so I need something easy to use.)
Why not plain old EDV?
I prefer VGLT in taxable for a variety of reasons but I should probably use EDV. (I've been considering this for some time.)
Both VGLT and EDV are likely going to have capital gain distributions, that is just the nature of longer term bond funds, they have to constantly churn to maintain the long duration.
Yes, EDV has had higher cap gain distributions than VGLT (and is more volatile with only a modest increase in CAGR). I intend to use low cap gains equities in taxable for early retirement income and high cap gains positions for charitable contributions so the "bonds in tax deferred" strategy is not optimal for me.
I had some EDV in spring 2020, used it more as equity crash insurance, and it worked. I sold all of it and bought a bunch of equities in March 2020. But I wouldn't buy it now, since the coupon is only around 2% now. Instead I have been buying I Bonds last year and this year and will buy more in 2022.
User avatar
zaboomafoozarg
Posts: 2431
Joined: Sun Jun 12, 2011 12:34 pm

Re: Why not 100% TMF?

Post by zaboomafoozarg »

Booogle wrote: Sun Nov 28, 2021 5:02 pm Even if you play around with the time periods, its pretty good.
So I was searching for TMF and happened upon this post.

I was also intrigued by TMF in the past, but now in the past 7 months since this was posted it has dropped 58%.

This makes BND's 12% drop in that time period feel like a walk in the park.
Northern Flicker
Posts: 15363
Joined: Fri Apr 10, 2015 12:29 am

Re: Why not 100% TMF?

Post by Northern Flicker »

Booogle wrote: Sun Nov 28, 2021 5:02 pm https://www.portfoliovisualizer.com/bac ... on3_2=-200

Even if you play around with the time periods, its pretty good.
Why did you end the time period at 2021 in your posting? Could this be the reason?

https://www.portfoliovisualizer.com/bac ... on3_2=-200

Using the max number of years in PV, the S&P 500 had a higher return with much lower volatility and a significantly lower max drawdown:

https://www.portfoliovisualizer.com/bac ... on3_2=-200

:oops:
Corvidae
Posts: 183
Joined: Tue May 10, 2022 1:11 am

Re: Why not 100% TMF?

Post by Corvidae »

Northern Flicker wrote: Mon Jun 27, 2022 9:30 pm
Why did you end the time period at 2021 in your posting?
Well, I'm not sure about this, but hear me out...




The original post is from 2021. :)

I'm glad someone dug up this thread though because I missed this madness.
Northern Flicker
Posts: 15363
Joined: Fri Apr 10, 2015 12:29 am

Re: Why not 100% TMF?

Post by Northern Flicker »

Corvidae wrote: Mon Jun 27, 2022 10:13 pm
Northern Flicker wrote: Mon Jun 27, 2022 9:30 pm
Why did you end the time period at 2021 in your posting?
Well, I'm not sure about this, but hear me out...

The original post is from 2021. :)

I'm glad someone dug up this thread though because I missed this madness.
Well, gosh, without a crystal ball I guess it just doesn’t work!
User avatar
nisiprius
Advisory Board
Posts: 52211
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: Why not 100% TMF?

Post by nisiprius »

Image

Green: long-term Treasury bonds.
Yellow: intermediate-term Treasury... notes.
Red: Treasury bills, which can also be taken as a proxy for "cash" (in an interest-earning account)

Years and years and years ago, I stared at the inflation-adjusted cumulative real growth charts for the SBBI data for intermediate-term and long-term government bonds, and concluded that any big commitment to long-term bonds was simply unsuitable unless you had known future dollar liabilities you could match with individual long-term bonds.

From 1913 to the present, there have been four episodes of high inflation (my own judgement, I don't know if there's any NBER-like institution that scores "inflation"), so that's an average of one every 27 years. That tells us that it is prudent to assume there's a good chance of encountering one in one's investing-and-retirement lifetime.

There are, indeed, two completely different philosophies on the way bonds can function in a portfolio.

The boring, but robust way is to treat them as not all that different from cash, but with a combination meaningfully higher expected return, and a hardly noticeable higher risk if it is in a portfolio together with stocks. To just buy and hold them, get the coupon interest and money back at maturity. With low correlation just meaning the small amount added risk is a bit smaller than you might expect. That calls for intermediate-term bonds.

The sophisticated, but fragile way is to hold them, not for their basic features of interest and return of principal, but as a fluctuating asset that you hold because you want the volatility... because you have some kind of prediction or expectation of negative correlation with stocks, or negative correlation under market conditions you expect to persist, or negative correlation provided the Fed does what you expect them to do. That calls for long-term bonds, maybe zero coupon, maybe leveraged.

But in any case it calls for holding them in a financially-engineered combination together with stocks. Not 100% bonds.
Last edited by nisiprius on Tue Jun 28, 2022 8:40 am, edited 1 time in total.
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
Cheez-It Guy
Posts: 4007
Joined: Sun Mar 03, 2019 3:20 pm

Re: Why not 100% TMF?

Post by Cheez-It Guy »

Why not 100% TSLA in 2019?

Dang, I'm a genius!
Topic Author
Booogle
Posts: 605
Joined: Wed Nov 09, 2016 2:57 pm

Re: Why not 100% TMF?

Post by Booogle »

I meant 100% TMV
caklim00
Posts: 2420
Joined: Mon May 26, 2008 10:09 am

Re: Why not 100% TMF?

Post by caklim00 »

prioritarian wrote: Mon Nov 29, 2021 12:23 pm
Booogle wrote: Mon Nov 29, 2021 9:38 am
anon_investor wrote: Mon Nov 29, 2021 9:04 am Wouldn't NTSX be easier? I feel like using 90% VTSAX, 10% TMF would require a lot of attentiveness and rebalancing to be effective.
Yes, but what about all the people who bought equities 10 years before NTSX existed?
This is definitely my dilemma. I have so much money locked into equities (e.g. >200% gains) that I will likely never rebalance. Adding some TMF would NTSXify them. Unfortunately, the high ER of TMF, volatility decay, and derivative-based counterparty risk have prevented me from using this method of diversify these holdings. If a TMF-like instrument with lower ER and less risk becomes available I will likely use it. (I don't have the spare time to roll futures so I need something easy to use.)
Probably use TYA https://www.morningstar.com/etfs/bats/tya/quote to simulate the futures. I used to roll futures but got out of that a few years ago, but haven't started using EDV or TYA or anything like that yet since I have a fair amount in cash due to requiring it for some other activities.
User avatar
firebirdparts
Posts: 4411
Joined: Thu Jun 13, 2019 4:21 pm
Location: Southern Appalachia

Re: Why not 100% TMF?

Post by firebirdparts »

Northern Flicker wrote: Mon Jun 27, 2022 11:31 pm
Corvidae wrote: Mon Jun 27, 2022 10:13 pm
Northern Flicker wrote: Mon Jun 27, 2022 9:30 pm
Why did you end the time period at 2021 in your posting?
Well, I'm not sure about this, but hear me out...

The original post is from 2021. :)

I'm glad someone dug up this thread though because I missed this madness.
Well, gosh, without a crystal ball I guess it just doesn’t work!
Time machine is better in this case. I had forgotten this thread occurred. It seems it was beautifully timed.
This time is the same
garlandwhizzer
Posts: 3565
Joined: Fri Aug 06, 2010 3:42 pm

Re: Why not 100% TMF?

Post by garlandwhizzer »

The fate of TMF points out the flaws of using backtesting as a reliable guide to the future. Backtesting leveraged LTTs from 1982 to 2020 looked like a sure bet for massively outsized risk adjusted returns. LTTs had incredible returns with no risk for 38 years, why not leverage it up? Instead the last 2 years produced a total disaster for TMF with losses of 75%+ in principal value. That is enough to answer the question: Why not 100% TMF?

Looking at where LTT yields were in 2020, the disaster that followed should not have been a surprise. We were in a Treasury bubble in 2020, historically lowest nominal yields for 13 years. That is precisely the worst time to buy, let alone leverage, Treasuries especially LTT. LTT had done so well for so long that, for those who put their faith in backtesting, 3X leverage looked like the optimal route to maximize those easy risk-free gains. In general get rich quicker schemes carry exceptionally high risk, like ARKK, TMF, and BTC. They can be big winners short term but only rarely does massive outperformance last long term.

Backtesting works until it doesn't. Both market sentiment and the underlying macroeconomy can change drastically and quickly from what it has been like for prior decades in which case backtesting can be a joke. That's what happened this time with TMF.

Garland Whizzer
ChinchillaWhiplash
Posts: 1574
Joined: Sat Jan 20, 2018 4:40 pm
Location: Land of Hypoxia

Re: Why not 100% TMF?

Post by ChinchillaWhiplash »

Booogle wrote: Tue Jun 28, 2022 11:13 am I meant 100% TMV
No way I would go 100% but I did buy some after selling off TMF last year. Has worked out well so far :mrgreen:
moontower
Posts: 46
Joined: Sun Jan 03, 2021 10:07 am

Re: Why not 100% TMF?

Post by moontower »

"In general get rich quicker schemes carry exceptionally high risk, like ARKK, TMF, and BTC. They can be big winners short term but only rarely does massive outperformance last long term."

WELL SAID! There really isn't a easy route to beating the SPY or VTI long term....although the ONE instance where "massive outperformance" did last was in TMF/TYD/EDV for past decade plus based on the 40 year bull in LTT.

QUESTION: Is there a point where TMF goes so low that the risk reward proposition returns? I'd say perhaps at the end of this year. Also, anytime the 10 year treasury hits about 3.5% or higher, I'd say good chance that wont last.

WHY? Because US govt cannot pay its debt service at that high of an interest rate and not erode the value of the USD. Thoughts?
toddthebod
Posts: 5735
Joined: Wed May 18, 2022 12:42 pm

Re: Why not 100% TMF?

Post by toddthebod »

moontower wrote: Wed Jun 29, 2022 12:13 pm QUESTION: Is there a point where TMF goes so low that the risk reward proposition returns? I'd say perhaps at the end of this year. Also, anytime the 10 year treasury hits about 3.5% or higher, I'd say good chance that wont last.

WHY? Because US govt cannot pay its debt service at that high of an interest rate and not erode the value of the USD. Thoughts?
I always point to this graph whenever people bring up how big the US debt has gotten (usually they are complaining that it's 120% of GDP, but this is the same question really.)
Image
Northern Flicker
Posts: 15363
Joined: Fri Apr 10, 2015 12:29 am

Re: Why not 100% TMF?

Post by Northern Flicker »

Booogle wrote: Tue Jun 28, 2022 11:13 am I meant 100% TMV
TMV had nearly a 99% drawdown through the end of 2021.

https://www.portfoliovisualizer.com/bac ... ion1_1=100
Diego_Quant
Posts: 38
Joined: Fri Jul 27, 2018 11:27 pm

Re: Why not 100% TMF?

Post by Diego_Quant »

Now is the time to go all in on $TMF $TLT?
Post Reply