80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
TomCat96
Posts: 790
Joined: Sun Oct 18, 2015 12:18 pm

80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Post by TomCat96 » Fri Sep 27, 2019 4:52 pm

Before I decided on my asset allocation at 100% stocks, there was one other curiousity I found in the data.
That curiousity has been noted, observed, and appears to be part of the theoretical background behind Hedgefundies excellent adventure.

I can't really speak for what others have found, but what I found personally was the 80/20 mix between stocks and long term treasuries could perform similarly to a 100% stock portfolio. In other cases it would even outperform the 100% stock portfolio. But most of all, it could do so with smaller drawdowns and lower standard deviation than the 100% stock portfolio in backtesting.

I admit, in my testing, I wasn't ready to accept the data when I ran across this in my testing. Emotions over data. It just seemed a bit radical to me.
Now here I am perusing old threads and it appears to be well documented.

So here's my question.
What have people found for the proper long term bond fund to use?

In my testing, I used TLT (ishares Barclary 20+ years treasuries).
There's also votes for VUSTX (Vanguard Long Term Treasuries), BLV (Vanguard long term bond fund)

For the stock portfolio aspect, the usual attempt is VTI. However, I saw at least one mention of S&P 600.

I just want to know what others who are trying the 80/20 (Long term bonds) or some variant thereof, what have you found?

For me I noted a few things in recent testing:

1. As Vineviz once pointed out, you need something a little more volatile than BND, that packs a little more punch to counter the stock market's movement during a severe downturn BND has low volatility and low market correlation, while in my testing long term treasuries were far more volatile and reached a market correlation of -.33

2. Nevertheless, the long term bonds don't so much outperform during a dramatic downturn as they respond to the first sign of weaknesses. In other words its not that Long Term treasuries will save you during a crash by grossly outperforming, it's that they will frequently spike modestly at the first sign of trouble, leading to gains from rebalancing.

3. The more drawn out the bull market, the more the pure market portfolio would begin to move ahead of the 80/20. Still because of point 2, the 80/20 is never far behind--slightly underperforming the market portfolio at considerably less volatility.

4. Dollar cost averaging changed the dynamics of the 80/20s advantage over the 100% portfolio considerably. In short, it weakened the benefit of the 80/20.

On point 4, dollar cost averaging results varied greatly with the relative size of the periodic purchases over the initial investment size. Generally speaking, when the size of the periodic purchases were large relative to the initial investment, it was the dollar cost averaging itself which mitigated the volatility of the overall portfolio during a downturn over the long term bond fund. The long term bond fund was more beneficial when the periodic investments were small compared to the portfolio.

5. The 80/20 has merit for the guy or gal, who wants to invest lumpsum and wants to "go in 100%" but is worried the market is going to crash soon.


What are your thoughts and experiences with the 80/20 (long term bond) portfolio?

robertmcd
Posts: 554
Joined: Tue Aug 09, 2016 9:06 am

Re: 80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Post by robertmcd » Fri Sep 27, 2019 5:11 pm

Ignore the dollar cost averaging.

EDV packs a bigger punch than TLT or VGLT. At any equity allocation 60% or above EDV is the best diversifier (historically)

60% VTI and 40% EDV is risk parity.

An 80/20 portfolio of vti/edv would be a great option

columbia
Posts: 2001
Joined: Tue Aug 27, 2013 5:30 am

Re: 80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Post by columbia » Fri Sep 27, 2019 5:37 pm

No great surprises:
https://www.portfoliovisualizer.com/bac ... 0&total3=0

It’s underperformed 100% stock, but with lower volatility.

Would that have better met your preferences? Only you can decide that.

User avatar
vineviz
Posts: 5391
Joined: Tue May 15, 2018 1:55 pm

Re: 80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Post by vineviz » Fri Sep 27, 2019 5:58 pm

TomCat96 wrote:
Fri Sep 27, 2019 4:52 pm
I can't really speak for what others have found, but what I found personally was the 80/20 mix between stocks and long term treasuries could perform similarly to a 100% stock portfolio. In other cases it would even outperform the 100% stock portfolio. But most of all, it could do so with smaller drawdowns and lower standard deviation than the 100% stock portfolio in backtesting.
Long-term Treasuries are certainly the best diversification for a77n equity portfolio, and you know I agree that the first 20% in bonds should be long-term Treasuries (or long-term TIPS if unexpectedly high inflation is a risk you face).

That said, you can't outrun the fact that backtests generally benefit from starting at points when bond yields were much higher than they are today.

VUSTX launched in 1986 when long-term Treasuries were yielding over 7% and TLT launched in 2002 when Treasuries were yielding nearly 5%. Today the yield on 20-year Treasuries is under 2%, so the returns going forward are quite likely to be lower than they have been historically. If future stock returns are similar to historical returns, the 80/20 portfolio probably will return less than a 100/0 portfolio.

Now, stock valuations (e.g. Shiller CAPE ratio) are also higher now than in 1986 or 200, so maybe future stock returns will also be lower. In that case, maybe the 80/20 returns will again be similar to 100/0 returns.

I guess my point is that long-term bonds probably ARE the best choice, just maybe don't be surprised if the 80/20 portfolio fails to keep up with the 100/0 portfolio in the future.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

delamer
Posts: 9344
Joined: Tue Feb 08, 2011 6:13 pm

Re: 80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Post by delamer » Fri Sep 27, 2019 6:07 pm

This is what Vanguard found re:allocations: https://personal.vanguard.com/us/insigh ... llocations

JBeck
Posts: 147
Joined: Fri Apr 15, 2016 4:54 am

Re: 80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Post by JBeck » Mon Sep 30, 2019 10:11 am

vineviz wrote:
Fri Sep 27, 2019 5:58 pm
Long-term Treasuries are certainly the best diversification for a77n equity portfolio, and you know I agree that the first 20% in bonds should be long-term Treasuries (or long-term TIPS if unexpectedly high inflation is a risk you face).
Would this still be your recommendation even if someone had a very conservative allocation, e.g., 70% bonds (20% LTT, 50% ITT)?

WhiteMaxima
Posts: 2036
Joined: Thu May 19, 2016 5:04 pm

Re: 80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Post by WhiteMaxima » Mon Sep 30, 2019 10:44 am

80/20 will perform very similar to 100% but with less volatility. You also have 20% pool of fix income to rebalance into stock. So 80/20 would be an ideal choice.

robertmcd
Posts: 554
Joined: Tue Aug 09, 2016 9:06 am

Re: 80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Post by robertmcd » Mon Sep 30, 2019 11:00 am

JBeck wrote:
Mon Sep 30, 2019 10:11 am
vineviz wrote:
Fri Sep 27, 2019 5:58 pm
Long-term Treasuries are certainly the best diversification for a77n equity portfolio, and you know I agree that the first 20% in bonds should be long-term Treasuries (or long-term TIPS if unexpectedly high inflation is a risk you face).
Would this still be your recommendation even if someone had a very conservative allocation, e.g., 70% bonds (20% LTT, 50% ITT)?
No. At 70% bonds you would probably want to hold 7-10 yr treasuries. In my opinion you should keep your bond allocation as short as possible while adequately diversifying your stock exposure. The shorter you go on the yield curve (from 30 yrs out to 1-3 yrs) the better risk adjusted returns are. So I don't want to hold short term treasuries and long term treasuries in a barbell to get intermediate exposure. I would be better off holding all intermediate.

https://www.portfoliovisualizer.com/opt ... sisResults

User avatar
Tyler Aspect
Posts: 1568
Joined: Mon Mar 20, 2017 10:27 pm
Location: California
Contact:

Re: 80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Post by Tyler Aspect » Mon Sep 30, 2019 11:05 am

Are you a young aggressive investor?

The thing about long term Treasuries is that if the long term yield goes up, you could see a steep hit to the fund's net asset value. This is why tilting into long term Treasuries is not a short term decision. Once you commit to this idea you have to follow through as long as 40 years in order to make sure you capture that income stream. If you sell out in a panic then you would have failed to realize the long term income stream but realizing the net asset loss instead.
Past result does not predict future performance. Mentioned investments may lose money. Contents are presented "AS IS" and any implied suitability for a particular purpose are disclaimed.

User avatar
nisiprius
Advisory Board
Posts: 39472
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: 80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Post by nisiprius » Mon Sep 30, 2019 11:17 am

TomCat96 wrote:
Fri Sep 27, 2019 4:52 pm
...I can't really speak for what others have found, but what I found personally was the 80/20 mix between stocks and long term treasuries could perform similarly to a 100% stock portfolio...
1) Over what time period did you find this?

2) If the time period wasn't "all available data," how did you decide which data matter, which should count, and which are the better predictors of the future?

I don't think there's any mystery here. I think most of us, including HEDGEFUNDIE and I, agree that the behavior of long-term bonds 1926-present has been different from the behavior of long-term bonds 2000-present. The question is whether you believe which period of time is the best guide to what to expect in future. The advocates of long-term bonds and/or extended duration and/or leveraged bonds believe that the behavior changed and was different before and after some date, possibly Volcker's tenure... but that it won't change again.

But if you use the SBBI data from 1926 through 2018, what you see is that an 80/20 has had superior risk-adjusted return than 100% stocks. (And 60/40 was better, and 45/55 was better yet). However, yes, just as you'd expect, 80/20 had lower return than 100% stocks.

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
nisiprius
Advisory Board
Posts: 39472
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: 80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Post by nisiprius » Mon Sep 30, 2019 11:19 am

TomCat96 wrote:
Fri Sep 27, 2019 4:52 pm
...I can't really speak for what others have found, but what I found personally was the 80/20 mix between stocks and long term treasuries could perform similarly to a 100% stock portfolio...
1) Over what time period did you find this?

2) If the time period wasn't "all available data," how did you decide which data should count and which should be ignored?

I don't think there's any mystery here. I think most of us, including HEDGEFUNDIE and I, agree that the behavior of long-term bonds 2000-present has been meaningfully different from the behavior of long-term bonds 1926-present. The question is which period of time you believe is the most reliable guide to the future. The advocates of long-term bonds (and EDV and leveraged bonds) are confident that a permanent change occurred at some date--possibly Volcker's tenure although I don't think the years match up--and thus, the time period from 2000 on is a better predictor than "all available data."

If you choose to use the SBBI data from 1926 through 2018, what you see is that, just as you'd naïvely expect, over that time period 80/20 had lower return than 100% stocks. Certainly, 80/20 had superior risk-adjusted return than 100% stocks (and 60/40 was better, and 45/55 was better yet) but the actual return was lower.

Specifically, 80/20 had about 86% of the excess return of 100% stocks. There was a beneficial effect of diversification, because cutting stock allocation 20% only cut excess return by 14%, not 20%--but still, it was a cut.

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
vineviz
Posts: 5391
Joined: Tue May 15, 2018 1:55 pm

Re: 80/20 vs 100% stocks, Long Term Treasuries, Can I get a primer?

Post by vineviz » Mon Sep 30, 2019 12:41 pm

JBeck wrote:
Mon Sep 30, 2019 10:11 am
vineviz wrote:
Fri Sep 27, 2019 5:58 pm
Long-term Treasuries are certainly the best diversification for a77n equity portfolio, and you know I agree that the first 20% in bonds should be long-term Treasuries (or long-term TIPS if unexpectedly high inflation is a risk you face).
Would this still be your recommendation even if someone had a very conservative allocation, e.g., 70% bonds (20% LTT, 50% ITT)?
Basically yes, as a general rule 20% of that portfolio could very reasonably be long-term treasuries.

But a portfolio with 70% bonds raises two questions:

1) how sensitive is the investor to unexpected inflation;
2) what is the investment horizon of the investor.

The overall weighted average duration of the bond portion of the portfolio should roughly match the investment horizon. Usually, this can be managed even with 20% of the portfolio in long-term bonds but an investor with very short time horizons (e.g. a retired couple in their late 80s) might find this difficult to manage and therefore might need all of their bonds to be intermediate-term.

Beyond that, sensitivity to inflation can play a role. A retired investor for whom Social Security and other inflation-indexed annuities do not cover at least their non-discretionary expenses should almost certainly have a significant portion (and maybe all) of their bonds in TIPS. These could be long-term TIPS (or a TIPS ladder), but a significant allocation to nominal bonds would magnify their inflation risk.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

Post Reply