Do long-term bonds belong in one's portfolio?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
patrick013
Posts: 2820
Joined: Mon Jul 13, 2015 7:49 pm

Re: Do long-term bonds belong in one's portfolio?

Post by patrick013 » Sun Dec 09, 2018 1:41 pm

patrick013 wrote:
Sat Dec 08, 2018 6:41 pm
JackoC wrote:
Sat Dec 08, 2018 5:45 pm

It's a common observation when stock prices go up bond prices go down, and when stock prices go down bond prices go up.
I would stop you right there though if I may. It's a common *belief* that stock and US treasury bond prices are negatively correlated. It's not actually a historically strong relationship at all. The rest of what you said I confess I don't understand. But the premise is not correct so probably better to focus on that basic disagreement.
If you read L. Swedroe's bond book he displays that HY bonds have equity correlation and TRSY bonds do not. So in his eyes the negative correlation does exist tho as you said with other market factors which may limit it. Just because something doesn't correlate today as expected doesn't mean it doesn't correlate and the observation should be thrown out.

My statement is an economic principle as such that market participants are expected to sell TRSY's and buy stocks when stocks are expected to have greater returns than TRSY's. This is expected to lower TRSY prices as supply of bonds for sale is greater than buyer demand then. Further, if the market is in equilibrium where sellers equal buyers and prices are generally stable, a sudden surge or continued increase in trading volume from foreign buyers can limit bond yield as prices increase due to lower supply at lower prices.

But read Larry Swedroe's bond book if you haven't. Nobody likes old economic books anyway but I always look beyond the stats for something fundamental or economic to explain a market that's changing. :)
age in bonds, buy-and-hold, 10 year business cycle

mptfan
Posts: 5808
Joined: Mon Mar 05, 2007 9:58 am

Re: Do long-term bonds belong in one's portfolio?

Post by mptfan » Sun Dec 09, 2018 1:52 pm

HEDGEFUNDIE wrote:
Sat Dec 08, 2018 9:15 am
I agree with all three of your points and they are why my bond allocation is 100% EDV
What is EDV?

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

Re: Do long-term bonds belong in one's portfolio?

Post by vineviz » Sun Dec 09, 2018 1:55 pm

JackoC wrote:
Sun Dec 09, 2018 12:24 pm
vineviz wrote:
Sun Dec 09, 2018 2:44 am
I just wanted to circle back and note that the ACM is prone, like many econometric models, be be sensitive to the inputs used to estimate its parameters.

The confidence interval for the so-called “ term premium “ in this model is rarely reported but is very wide. Wide enough that the reported sign is often suspect.

In this case, the model likely is overestimating the future short term rates by 50 bps or so and therefore underestimating the term premium.
I think you need to provide more support for those statements. Look at the output of that model over 50 yrs. How often has the sign flipped? Very rarely. It's certainly not indicative of a confidence interval on the order of the current answer of -60 bps.
First, whoever publishes a model has a burden to provide the relevant statistical data (including confidence intervals) at the very least if not some measure of how accurately the model predicts the thing it is trying to model. As far as I can tell, the Fed doesn't do this.

When challenged on the robustness of the model, they did reply (in part):
Given the substantial model and (in some cases) parameter uncertainty involved with estimating dynamic term structure models, it is worth bearing in mind that while some features of estimated term premiums appear to be fairly robust across models (e.g. long-maturity term premium estimates from the models considered in this Note are all low by historical standards), other features (such as the sign and volatility of term premiums), are less robust.
From Cohen, Hördahl and Xia:
Estimates derived from the ACM and the HT models can differ by as much as 200 bps. The discrepancy partly reflects the assumptions embedded in these estimates....
Recent commentary from Guggenheim Investments:
Term premiums are not as low as the Fed’s models say: Because term premiums cannot be observed, investors must rely on model estimates to gauge the duration risk premiums embedded in Treasury securities. One of the most frequently cited is the Adrian, Crump, and Moench (ACM) model, which is published by the Federal Reserve Bank of New York (New York Fed). In the New York Fed’s models, term premium is defined as the observed level of 10-year Treasury yields minus the estimate of the “risk-neutral yield,” which is an estimate of the average short-term rate expected to prevail over 10 years. The ACM model currently indicates that 10-year term premiums are -35 basis points, roughly 85 basis points below where they were in mid-2005, one year before the end of the last Fed tightening cycle. This would seem to corroborate research suggesting that the Fed’s asset purchase programs cumulatively depressed 10-year term premiums by over 100 basis points (this effect would have faded somewhat in recent years).

The problem is that today the ACM model overstates expected short-term rates, falsely implying that term premiums are lower than they really are. Specifically, ACM indicates that the risk-neutral 10-year yield is currently 3.50 percent, considerably above our own forecasts of the average short-term rate that will prevail over the next decade. Other outside estimates support this view: the 10-year overnight index swap (OIS) rate, or the average fed funds rate priced in to a fixed-floating swap, is 2.90 percent. Meanwhile, the median estimates from the New York Fed’s August 2018 dealer and buy-side policy surveys are both 2.50 percent. Similarly, respondents to the Federal Reserve Bank of Philadelphia’s Survey of Professional Forecasters expect the average three-month Treasury bill return over the next 10 years to be 2.75 percent (see chart below).

Each of these independent forecasts suggests the ACM risk-neutral yield estimate is too high. This leads us to believe that the New York Fed’s model understates term premiums, which undercuts a key pillar supporting the argument that QE has depressed term premiums and made the yield curve unduly flat relative to past cycles.
As I think I said, the term premium probably IS at relatively low levels (though possibly not at historic lows: the ACM model estimate a lower term premium in 2012, since when EDV has outperformed VGSH by an annualized rate of 108 bps). And it's not IMPOSSIBLE for the term premium to be negative for short periods of time. But it is almost certain NOT negative right now.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

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

Re: Do long-term bonds belong in one's portfolio?

Post by vineviz » Sun Dec 09, 2018 1:57 pm

mptfan wrote:
Sun Dec 09, 2018 1:52 pm
HEDGEFUNDIE wrote:
Sat Dec 08, 2018 9:15 am
I agree with all three of your points and they are why my bond allocation is 100% EDV
What is EDV?
Mentioned earlier in this thread, I think, EDV is Vanguard Extended Duration Treasury ETF. It holds zero coupon Treasury securities with maturities of 20 to 30 years.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

HEDGEFUNDIE
Posts: 4038
Joined: Sun Oct 22, 2017 2:06 pm

Re: Do long-term bonds belong in one's portfolio?

Post by HEDGEFUNDIE » Sun Dec 09, 2018 2:01 pm

stlutz wrote:
Sun Dec 09, 2018 12:53 pm
2pedals wrote:
Sun Dec 09, 2018 12:05 pm
A longer-term bond carries greater risk that higher inflation could reduce the value of payments, as well as greater risk that higher overall interest rates could cause the bond's price to fall. This is something I am trying to avoid at this stage.
For the moment let me keep the inflation piece out of the equation and consider longer-term TIPS vs. shorter-term TIPS.

The shorter-term TIPS face reinvestment risk. That is, when your current bond matures you may not be able to replace it with a bond that yields as much. For example, right now 5 year TIPS yield 1.02%; five years ago they yielded -.13%. With short-term bonds, you face the risk of having to reinvest at lower real rates in the future.

The longer-term TIPS face interest-rate risk. That is, if I buy a 30 year TIPS now at 1.17%, they might yield 2% in the future and I'm stuck with my current rate for 30 years. If I sell before the 30 years are up, I have to accept a loss in principal value.

It seems fairly common to view the former risk as not that big of the deal and the later as a big problem. I don't see why--shouldn't they balance against one another?
Quite simply: recency bias and market timing!

User avatar
Topic Author
CULater
Posts: 2421
Joined: Sun Nov 13, 2016 10:59 am

Re: Do long-term bonds belong in one's portfolio?

Post by CULater » Sun Dec 09, 2018 3:03 pm

Actually, I just plugged in US stocks and LT Treasuries into Portfolio Visualizer and the optimal portfolio (risk parity) from 1985 - present was 40% Stocks and 60% long treasuries. And from 1978 portfolios with 60% stocks and 40% stocks had an identical risk-adjusted return, with the worst drawdown of the latter being about 14% vs. 27% for the former. Looks like stocks made a nice diversifier for long term bonds!
On the internet, nobody knows you're a dog.

stlutz
Posts: 5537
Joined: Fri Jan 02, 2009 1:08 am

Re: Do long-term bonds belong in one's portfolio?

Post by stlutz » Sun Dec 09, 2018 5:08 pm

Quite simply: recency bias and market timing!
Actually, it's quite the opposite. Run any backtest over, say, the last 5 or 10 or 20 years and long-term bonds look the best, both on a standalone basis and as part of a balanced portfolio. Those who advocate for long term bonds are the ones who can be accused of recency bias (fairly or unfairly.

Having been around here for over 10 years now, the "standard" advice is that any high quality short-to-intermediate term bond fund will work just fine. Again, the standard view has been that long-term bonds are too volatile and therefore may not serve their proper purpose of reducing portfolio risk. That's a structural point, not based on recency bias.

It was poster "bobcat2" who first provoked me to start thinking about this differently in his posts on how to use TIPS as part of retirement income planning.

If anything this form has has "distance bias"--the view that any day now Treasury rates are going to shoot up to 7% like they was back in the 80s and holders of long term bonds will all get massacred.

longinvest
Posts: 4131
Joined: Sat Aug 11, 2012 8:44 am

Re: Do long-term bonds belong in one's portfolio?

Post by longinvest » Sun Dec 09, 2018 5:59 pm

stlutz wrote:
Sun Dec 09, 2018 5:08 pm
If anything this form has has "distance bias"--the view that any day now Treasury rates are going to shoot up to 7% like they was back in the 80s and holders of long term bonds will all get massacred.
Long-term bonds would severely drop in value, but would immediately start yielding 7%. More importantly, they would continue to pay their coupons and principal. They're bonds, after all.

I agree with Jack Bogle's view of bonds: "I don’t look at bond funds as a unity. You can hold a short-term bond fund if you’re willing to sacrifice income in favor or principal stability. You can hold a long-term bond fund if you want a higher interest rate and therefore a higher long-term return, but you have to be willing to tolerate the higher volatility. I tend to favor, mostly for behavioral reasons, an intermediate-term maturity, which will only have roughly half the volatility of the stock market."(Source)

Mainly, long-term bonds are volatile.

If the investor's goal is to dampen stock volatility, then a broad market bond index fund of intermediate duration is probably best. This is the choice of many Bogleheads and the default suggestion in our forums.

Note that Jack Bogle never suggests to concentrate one's bond investments into Treasuries. He likes to include corporate bonds with their higher coupons. He is OK with all maturities, short, intermediate, and long, but he has a preference for intermediate-term maturity.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

JackoC
Posts: 1009
Joined: Sun Aug 12, 2018 11:14 am

Re: Do long-term bonds belong in one's portfolio?

Post by JackoC » Sun Dec 09, 2018 6:18 pm

vineviz wrote:
Sun Dec 09, 2018 1:55 pm


1. Given the substantial model and (in some cases) parameter uncertainty involved with estimating dynamic term structure models, it is worth bearing in mind that while some features of estimated term premiums appear to be fairly robust across models (e.g. long-maturity term premium estimates from the models considered in this Note are all low by historical standards), other features (such as the sign and volatility of term premiums), are less robust.

2. Estimates derived from the ACM and the HT models can differ by as much as 200 bps. The discrepancy partly reflects the assumptions embedded in these estimates....

3. The problem is that today the ACM model overstates expected short-term rates, falsely implying that term premiums are lower than they really are.
Specifically, ACM indicates that the risk-neutral 10-year yield is currently 3.50 percent, considerably above our own forecasts of the average short-term rate that will prevail over the next decade. Other outside estimates support this view: the 10-year overnight index swap (OIS) rate, or the average fed funds rate priced in to a fixed-floating swap, is 2.90 percent. Meanwhile, the median estimates from the New York Fed’s August 2018 dealer and buy-side policy surveys are both 2.50 percent. Similarly, respondents to the Federal Reserve Bank of Philadelphia’s Survey of Professional Forecasters expect the average three-month Treasury bill return over the next 10 years to be 2.75 percent (see chart below).

4. As I think I said, the term premium probably IS at relatively low levels (though possibly not at historic lows: the ACM model estimate a lower term premium in 2012, since when EDV has outperformed VGSH by an annualized rate of 108 bps). And it's not IMPOSSIBLE for the term premium to be negative for short periods of time. But it is almost certain NOT negative right now.
First, thanks for all that but I don't agree with all of it.
1. This may be my misunderstanding of your original statement. I wouldn't find it all surprising if differing models sometimes gave a different sign for the term premium. That's not the same as saying the ACM model has an admitted error equal to a sign change when it's -60bps.

2. But not that far apart now

3. Swap rates contain a term premium just like treasury rates do. You can't tell anything about the accuracy of a model purporting to tease out the term premium just by looking at other market yield curves. It's not directly observable, like they said. That makes me wonder about whoever wrote that. It's conceivably valid to a look at a survey based measure but those are pretty fuzzy when people *can* observe the *risk neutral* short forward rates (those are the ones derived by simple calculation for the yield curve, models like ACM try to determine the *actually* expected forward rates minus the term premium) and it's common to assume the two are the same ie there's no term premium.

Also to reiterate, they claim the *Fed* is not creating a negative term premium, but the causal factors are not limited to the Fed. Other central banks' very low term rate policies feed back to the US long rates.

So point 3 per Guggenheim is pretty much of an outright fail IMO.

4. Yes and a lot of analysis of the value of long term bonds is based on periods where the term premium was clearly higher than now. Whether it's actually negative now is not the only thing. If it's zero that's still a huge difference from a 1961-current data set were it averaged 160bps in 10 yr per ACM, significantly more if a sample only goes back to the 1980's, it was enormous in the 80's/90's see above.

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

Re: Do long-term bonds belong in one's portfolio?

Post by vineviz » Sun Dec 09, 2018 7:50 pm

JackoC wrote:
Sun Dec 09, 2018 6:18 pm
1. This may be my misunderstanding of your original statement. I wouldn't find it all surprising if differing models sometimes gave a different sign for the term premium. That's not the same as saying the ACM model has an admitted error equal to a sign change when it's -60bps.
I think anyone who uses the ACM model, and especially anyone who defends it, should know what the 95% confidence interval is for that model's estimate of the the term premium. Do you know it is?

JackoC wrote:
Sun Dec 09, 2018 6:18 pm
You can't tell anything about the accuracy of a model purporting to tease out the term premium just by looking at other market yield curves. It's not directly observable, like they said.
The OIS reflects real investors putting real money into an estimate of what short-term rates will look like over a given period of time. It seems pretty evident that proponents of the ACM model should have a pretty good idea of WHY that model is generating estimates that are 60bps above what the market is pricing in. I mean, if your model of market expectations is 20% away from a direct observation of market expectations, that's an important piece of information.

JackoC wrote:
Sun Dec 09, 2018 6:18 pm
Also to reiterate, they claim the *Fed* is not creating a negative term premium, but the causal factors are not limited to the Fed. Other central banks' very low term rate policies feed back to the US long rates.

So point 3 per Guggenheim is pretty much of an outright fail IMO.
I'm pretty sure the chief investment officer and head macroeconomist at Guggenheim know that the Fed is merely estimating, not setting, the term premium. Their point is that the estimate produced by the Fed, using the ACM model, is probably wildly inaccurate. For what it is worth, the Fed's response to that criticism was akin to "you might be right".
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

columbia
Posts: 2270
Joined: Tue Aug 27, 2013 5:30 am

Re: Do long-term bonds belong in one's portfolio?

Post by columbia » Sun Dec 09, 2018 8:03 pm

Negative annual returns on LT treasuries since 1977:

1978: - 0.74%
1980: - 1.29%
1987: - 2.64%
1994: - 7.19%
2009: - 10.17%
2013: - 8.57%
2018: - 1.76%

That doesn’t seem worth a large degree of handwringing. Having said:

ST: $110,537
IT: $167,172
LT: $180,211

Got more bang for buck by moving from ST to IT than from IT to LT.

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

Re: Do long-term bonds belong in one's portfolio?

Post by vineviz » Sun Dec 09, 2018 8:40 pm

longinvest wrote:
Sun Dec 09, 2018 5:59 pm
If the investor's goal is to dampen stock volatility, then a broad market bond index fund of intermediate duration is probably best. This is the choice of many Bogleheads and the default suggestion in our forums.
This is a common misperception, or at least it depends on how much the investor wants to dampen volatility.

For equity allocations over 50%, it actually turns out that long-term bonds have dampened portfolio volatility better than intermediate-term bonds. For some reason, investors have a tendency to focus on the assets to the detriment of the portfolio.

Image

https://www.portfoliovisualizer.com/eff ... y&sf3=true
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
willthrill81
Posts: 16129
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Do long-term bonds belong in one's portfolio?

Post by willthrill81 » Sun Dec 09, 2018 8:52 pm

columbia wrote:
Sun Dec 09, 2018 8:03 pm
Negative annual returns on LT treasuries since 1977:

1978: - 0.74%
1980: - 1.29%
1987: - 2.64%
1994: - 7.19%
2009: - 10.17%
2013: - 8.57%
2018: - 1.76%
Your numbers are off. According to Portfolio Visualizer, LTT had significantly worse returns in several of those years than what you cite. Further, those are nominal returns; the real returns of LTT were significantly lower, especially in 1978-1981, when inflation was 9.02%, 13.29%, 12.52%, and 8.92%, respectively.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

HEDGEFUNDIE
Posts: 4038
Joined: Sun Oct 22, 2017 2:06 pm

Re: Do long-term bonds belong in one's portfolio?

Post by HEDGEFUNDIE » Sun Dec 09, 2018 9:30 pm

columbia wrote:
Sun Dec 09, 2018 8:03 pm
Negative annual returns on LT treasuries since 1977:

1978: - 0.74%
1980: - 1.29%
1987: - 2.64%
1994: - 7.19%
2009: - 10.17%
2013: - 8.57%
2018: - 1.76%

That doesn’t seem worth a large degree of handwringing. Having said:

ST: $110,537
IT: $167,172
LT: $180,211

Got more bang for buck by moving from ST to IT than from IT to LT.
Haha willthrill doesn’t like the first part of your post, and I don’t like the second. One just can’t win on this board! :D

https://www.portfoliovisualizer.com/bac ... asury1=100

ST: $107k
IT: $165k
LT: $249k (with some serious losses in the late 70s early 80s)

garlandwhizzer
Posts: 2602
Joined: Fri Aug 06, 2010 3:42 pm

Re: Do long-term bonds belong in one's portfolio?

Post by garlandwhizzer » Mon Dec 10, 2018 12:30 pm

Historical backtesting is interesting but as a practical matter it doesn't tell you too much about the future going forward from here. The current spread between 30 yr. Treasuries and 2 yr. Treasuries is 0.42%. If you bought 30 year Treasuries in 1980 you were assured of a 14%+ yield which did not adjust downward for 30 years while 2 yr. Treasuries did frequently as rates declined. That difference distorts backtesting results of LTB versus STB and makes it an unreliable predictor of the future returns. What is true is that if you take on a massive extra 28 years of duration risk today (30 yr. versus 2 yr.) you get paid only 0.42%/yr. for doing it. Historically, 30 years is a long time for inflation to remain dormant. It is entirely possible that average inflation rate over the next 30 years will exceed the current yield of 30 year Treasuries, so at the end you wind up with a negative real yield for 30 years and your return of principal is almost meaningless due to cumulative inflation over 3 decades. I don't care what backtesting shows this isn't a great deal. I do not find the risk/reward situation for LTB attractive at this time.

Garland Whizzer

rich126
Posts: 1127
Joined: Thu Mar 01, 2018 4:56 pm

Re: Do long-term bonds belong in one's portfolio?

Post by rich126 » Mon Dec 10, 2018 2:02 pm

HEDGEFUNDIE wrote:
Sun Dec 09, 2018 9:30 pm
columbia wrote:
Sun Dec 09, 2018 8:03 pm
Negative annual returns on LT treasuries since 1977:

1978: - 0.74%
1980: - 1.29%
1987: - 2.64%
1994: - 7.19%
2009: - 10.17%
2013: - 8.57%
2018: - 1.76%

That doesn’t seem worth a large degree of handwringing. Having said:

ST: $110,537
IT: $167,172
LT: $180,211

Got more bang for buck by moving from ST to IT than from IT to LT.
Haha willthrill doesn’t like the first part of your post, and I don’t like the second. One just can’t win on this board! :D

https://www.portfoliovisualizer.com/bac ... asury1=100

ST: $107k
IT: $165k
LT: $249k (with some serious losses in the late 70s early 80s)
And some of those down years were even worse if you factor in inflation.
1978: 7.6%
1980: 13.5%
2009/2013 inflation was close to 0.

I guess it is due to how the various online calculators work but a pet peeve of mine is when people talk of nominal returns instead of real returns. I don't care if something returns 10% if its real return was -3%.

HEDGEFUNDIE
Posts: 4038
Joined: Sun Oct 22, 2017 2:06 pm

Re: Do long-term bonds belong in one's portfolio?

Post by HEDGEFUNDIE » Mon Dec 10, 2018 2:05 pm

rich126 wrote:
Mon Dec 10, 2018 2:02 pm
HEDGEFUNDIE wrote:
Sun Dec 09, 2018 9:30 pm
columbia wrote:
Sun Dec 09, 2018 8:03 pm
Negative annual returns on LT treasuries since 1977:

1978: - 0.74%
1980: - 1.29%
1987: - 2.64%
1994: - 7.19%
2009: - 10.17%
2013: - 8.57%
2018: - 1.76%

That doesn’t seem worth a large degree of handwringing. Having said:

ST: $110,537
IT: $167,172
LT: $180,211

Got more bang for buck by moving from ST to IT than from IT to LT.
Haha willthrill doesn’t like the first part of your post, and I don’t like the second. One just can’t win on this board! :D

https://www.portfoliovisualizer.com/bac ... asury1=100

ST: $107k
IT: $165k
LT: $249k (with some serious losses in the late 70s early 80s)
And some of those down years were even worse if you factor in inflation.
1978: 7.6%
1980: 13.5%
2009/2013 inflation was close to 0.

I guess it is due to how the various online calculators work but a pet peeve of mine is when people talk of nominal returns instead of real returns. I don't care if something returns 10% if its real return was -3%.
The problem with using real returns is that not everything you buy inflates at the same rate. Depending on your consumption patterns, your personal rate of inflation could well be negative (there was a thread on this recently). Nominal returns is at least comparable apples-to-apples.

User avatar
willthrill81
Posts: 16129
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Do long-term bonds belong in one's portfolio?

Post by willthrill81 » Mon Dec 10, 2018 2:24 pm

HEDGEFUNDIE wrote:
Mon Dec 10, 2018 2:05 pm
rich126 wrote:
Mon Dec 10, 2018 2:02 pm
HEDGEFUNDIE wrote:
Sun Dec 09, 2018 9:30 pm
columbia wrote:
Sun Dec 09, 2018 8:03 pm
Negative annual returns on LT treasuries since 1977:

1978: - 0.74%
1980: - 1.29%
1987: - 2.64%
1994: - 7.19%
2009: - 10.17%
2013: - 8.57%
2018: - 1.76%

That doesn’t seem worth a large degree of handwringing. Having said:

ST: $110,537
IT: $167,172
LT: $180,211

Got more bang for buck by moving from ST to IT than from IT to LT.
Haha willthrill doesn’t like the first part of your post, and I don’t like the second. One just can’t win on this board! :D

https://www.portfoliovisualizer.com/bac ... asury1=100

ST: $107k
IT: $165k
LT: $249k (with some serious losses in the late 70s early 80s)
And some of those down years were even worse if you factor in inflation.
1978: 7.6%
1980: 13.5%
2009/2013 inflation was close to 0.

I guess it is due to how the various online calculators work but a pet peeve of mine is when people talk of nominal returns instead of real returns. I don't care if something returns 10% if its real return was -3%.
The problem with using real returns is that not everything you buy inflates at the same rate. Depending on your consumption patterns, your personal rate of inflation could well be negative (there was a thread on this recently). Nominal returns is at least comparable apples-to-apples.
Real returns are comparable. The fact that everyone experiences inflation in a slightly different way and to a greater or lesser degree doesn't impact this. Real returns are equal to nominal returns minus inflation, no matter who you are.

Nominal rates are far more deceiving.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

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

Re: Do long-term bonds belong in one's portfolio?

Post by vineviz » Mon Dec 10, 2018 2:47 pm

willthrill81 wrote:
Mon Dec 10, 2018 2:24 pm
Real returns are comparable. The fact that everyone experiences inflation in a slightly different way and to a greater or lesser degree doesn't impact this. Real returns are equal to nominal returns minus inflation, no matter who you are.

Nominal rates are far more deceiving.
This dead horse again?

If A is a linear transformation of B, it takes a certain kind of hyperbole to assert that A is "far more deceiving" than B don't you think?
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
willthrill81
Posts: 16129
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Do long-term bonds belong in one's portfolio?

Post by willthrill81 » Mon Dec 10, 2018 2:52 pm

vineviz wrote:
Mon Dec 10, 2018 2:47 pm
willthrill81 wrote:
Mon Dec 10, 2018 2:24 pm
Real returns are comparable. The fact that everyone experiences inflation in a slightly different way and to a greater or lesser degree doesn't impact this. Real returns are equal to nominal returns minus inflation, no matter who you are.

Nominal rates are far more deceiving.
This dead horse again?

If A is a linear transformation of B, it takes a certain kind of hyperbole to assert that A is "far more deceiving" than B don't you think?
If you think that nominal rates of return are as meaningful to investors as real rates, then there's no point in discussing the issue further as we are unlikely to find any common ground on it. And if that's what you truly believe, you're definitely in the minority around here (not that that's necessarily a bad thing BTW).
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

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

Re: Do long-term bonds belong in one's portfolio?

Post by vineviz » Mon Dec 10, 2018 3:14 pm

willthrill81 wrote:
Mon Dec 10, 2018 2:52 pm

If you think that nominal rates of return are as meaningful to investors as real rates, then there's no point in discussing the issue further as we are unlikely to find any common ground on it.
I don't think I ever said that. What I think I HAVE said, and which seems obvious, is that there are times when it makes sense to use nominal returns and times when it makes sense to use real returns.

MOST of the time, however, it makes very little difference to the discussion. If we are discussing the relative performance of two investments, for instance, subtracting a constant from both returns doesn't change the relationship one iota.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

ThrustVectoring
Posts: 771
Joined: Wed Jul 12, 2017 2:51 pm

Re: Do long-term bonds belong in one's portfolio?

Post by ThrustVectoring » Mon Dec 10, 2018 3:38 pm

Given what I know of the treasury futures market, I feel like I have to be missing something for long-term bonds to make any sort of sense over levering up short-term bonds. Maybe convexity?

Let's compare investment options for $100k in the 10-year cash bond vs the 2-year treasury futures contract. As always, latest numbers come from straight from CMEgroup.

Cash bond DV01: $84.43

2-year contract DV01: $32.71

So instead of holding the cash bond, you could take out two contracts and have somewhat less exposure to interest rate risk. Now, for a very rough expected return, calculate the return if interest rates remain flat: you get the treasury futures yield on the notional amount, minus the implied repo rate for the financed portion. This is $400k * 2.74% - $300k * 2.33% = $3970. For the cash bond, just assume yield times notional: 2.77% * $100k = $2770.

Again, I feel like I have to be missing something pretty big here. Convexity maybe? Futures market pricing in expected fed activity? I really don't know. My fairly naive analysis suggests that you can earn more money with less risk through levering up short-term bonds, even though I have pretty strong efficient-market intuitions. The long-term bond buyers have to be buying something - maybe it's regulatory compliance? The yield curve makes zero sense to me otherwise.
Current portfolio: 60% VTI / 40% VXUS

Elysium
Posts: 2142
Joined: Mon Apr 02, 2007 6:22 pm

Re: Do long-term bonds belong in one's portfolio?

Post by Elysium » Mon Dec 10, 2018 3:49 pm

Yield spreads do matter, that is the bottomline.

When yield curve is indicating narrow spreads between LT and IT, it doesn't pay to go long, and in fact the spread between ST and IT makes IT at the sweetspot of the yield curve.

Back in the days Larry Swedroe used to advocate this point, and I agree with him on this. LT bond spreads aren't attractive enough to play the risk for individual investors.

You could get 90% of the benefit of LT bonds from owning IT bonds, then why go up the yield curve for much more additional risk for very little reward.

ThrustVectoring
Posts: 771
Joined: Wed Jul 12, 2017 2:51 pm

Re: Do long-term bonds belong in one's portfolio?

Post by ThrustVectoring » Mon Dec 10, 2018 4:01 pm

Elysium wrote:
Mon Dec 10, 2018 3:49 pm
Yield spreads do matter, that is the bottomline.

When yield curve is indicating narrow spreads between LT and IT, it doesn't pay to go long, and in fact the spread between ST and IT makes IT at the sweetspot of the yield curve.

Back in the days Larry Swedroe used to advocate this point, and I agree with him on this. LT bond spreads aren't attractive enough to play the risk for individual investors.

You could get 90% of the benefit of LT bonds from owning IT bonds, then why go up the yield curve for much more additional risk for very little reward.
The yield curve is currently basically flat right now from the 1-year to 7-year mark. There's little reason to go beyond short-term at the moment, unless you have known future expenses and simply want to avoid reinvestment risk.
Current portfolio: 60% VTI / 40% VXUS

hdas
Posts: 1298
Joined: Thu Jun 11, 2015 8:24 am

[Deleted]

Post by hdas » Mon Dec 10, 2018 4:26 pm

[Deleted]
Last edited by hdas on Wed Jan 29, 2020 12:24 pm, edited 1 time in total.
....

Elysium
Posts: 2142
Joined: Mon Apr 02, 2007 6:22 pm

Re: Do long-term bonds belong in one's portfolio?

Post by Elysium » Mon Dec 10, 2018 4:35 pm

ThrustVectoring wrote:
Mon Dec 10, 2018 4:01 pm
Elysium wrote:
Mon Dec 10, 2018 3:49 pm
Yield spreads do matter, that is the bottomline.

When yield curve is indicating narrow spreads between LT and IT, it doesn't pay to go long, and in fact the spread between ST and IT makes IT at the sweetspot of the yield curve.

Back in the days Larry Swedroe used to advocate this point, and I agree with him on this. LT bond spreads aren't attractive enough to play the risk for individual investors.

You could get 90% of the benefit of LT bonds from owning IT bonds, then why go up the yield curve for much more additional risk for very little reward.
The yield curve is currently basically flat right now from the 1-year to 7-year mark. There's little reason to go beyond short-term at the moment, unless you have known future expenses and simply want to avoid reinvestment risk.
That is correct, at the moment. I hold both TBM and ST Bond Index, so my average duration is between IT and ST.

international001
Posts: 1235
Joined: Thu Feb 15, 2018 7:31 pm

Re: Do long-term bonds belong in one's portfolio?

Post by international001 » Mon Dec 10, 2018 4:41 pm

Ideally I would like to have LT because the lowest correlation with stocks and higher return. Never mind they are more volatile, they historically better with stocks.
But at the moment I have IT because the little extra benefit of LT.

How will I know to switch from IT to LT?

columbia
Posts: 2270
Joined: Tue Aug 27, 2013 5:30 am

Re: Do long-term bonds belong in one's portfolio?

Post by columbia » Mon Dec 10, 2018 4:48 pm

international001 wrote:
Mon Dec 10, 2018 4:41 pm
Ideally I would like to have LT because the lowest correlation with stocks and higher return. Never mind they are more volatile, they historically better with stocks.
But at the moment I have IT because the little extra benefit of LT.

How will I know to switch from IT to LT?
I think he advises against LT bonds, but that sounds like an article Larry Swedroe might author. 😀

HEDGEFUNDIE
Posts: 4038
Joined: Sun Oct 22, 2017 2:06 pm

Re: Do long-term bonds belong in one's portfolio?

Post by HEDGEFUNDIE » Mon Dec 10, 2018 5:01 pm

international001 wrote:
Mon Dec 10, 2018 4:41 pm
Ideally I would like to have LT because the lowest correlation with stocks and higher return. Never mind they are more volatile, they historically better with stocks.
But at the moment I have IT because the little extra benefit of LT.

How will I know to switch from IT to LT?
Apparently you will “know” it when you see it, as Elysium and other posters here are doing. Obvious market timing that would be readily dismissed in any thread about stocks.

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

Re: Do long-term bonds belong in one's portfolio?

Post by robertmcd » Mon Dec 10, 2018 5:01 pm

international001 wrote:
Mon Dec 10, 2018 4:41 pm
Ideally I would like to have LT because the lowest correlation with stocks and higher return. Never mind they are more volatile, they historically better with stocks.
But at the moment I have IT because the little extra benefit of LT.

How will I know to switch from IT to LT?
I think your equity percentage is more important when deciding what duration to target

30% or less go short term
30%-80% intermediate term
80% or higher go long term

User avatar
willthrill81
Posts: 16129
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Do long-term bonds belong in one's portfolio?

Post by willthrill81 » Mon Dec 10, 2018 5:13 pm

Even though the period from 1978-2018 (data in PV) mostly consisted of a bond bull market from 1982-2012 due to declining interest rates, the Sharpe ratio of long-term Treasuries was significantly lower than that of U.S. stocks (.37 vs. .51). So to the extent that that holds true going forward, it wouldn't make sense to include LTT in your portfolio if you're trying to maximize your absolute or risk-adjusted returns. If that was your goal, you should just increase your stock allocation.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

HEDGEFUNDIE
Posts: 4038
Joined: Sun Oct 22, 2017 2:06 pm

Re: Do long-term bonds belong in one's portfolio?

Post by HEDGEFUNDIE » Mon Dec 10, 2018 5:26 pm

willthrill81 wrote:
Mon Dec 10, 2018 5:13 pm
Even though the period from 1978-2018 (data in PV) mostly consisted of a bond bull market from 1982-2012 due to declining interest rates, the Sharpe ratio of long-term Treasuries was significantly lower than that of U.S. stocks (.37 vs. .51). So to the extent that that holds true going forward, it wouldn't make sense to include LTT in your portfolio if you're trying to maximize your absolute or risk-adjusted returns. If that was your goal, you should just increase your stock allocation.
So misleading. No one holds 100% bonds in isolation. When paired with an 60-80% stock allocation, LTT delivered a better portfolio Sharpe Ratio than ITT or STT:

https://www.portfoliovisualizer.com/bac ... easury1=40

Negative correlations for the win!

User avatar
willthrill81
Posts: 16129
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Do long-term bonds belong in one's portfolio?

Post by willthrill81 » Mon Dec 10, 2018 5:29 pm

HEDGEFUNDIE wrote:
Mon Dec 10, 2018 5:26 pm
willthrill81 wrote:
Mon Dec 10, 2018 5:13 pm
Even though the period from 1978-2018 (data in PV) mostly consisted of a bond bull market from 1982-2012 due to declining interest rates, the Sharpe ratio of long-term Treasuries was significantly lower than that of U.S. stocks (.37 vs. .51). So to the extent that that holds true going forward, it wouldn't make sense to include LTT in your portfolio if you're trying to maximize your absolute or risk-adjusted returns. If that was your goal, you should just increase your stock allocation.
So misleading. No one holds 100% bonds in isolation. When paired with an 60-80% stock allocation, LTT delivered a better portfolio Sharpe Ratio than ITT or STT:

https://www.portfoliovisualizer.com/bac ... easury1=40

Negative correlations for the win!
Your analysis compared LTT to STT and ITT. Of course LTT beat out the others over a 40 year period. But if you compared your 60% TSM / 40% LTT portfolio to 100% stocks, the former had a slightly higher Sharpe ratio but a lower absolute return (10.71% vs. 11.46%).
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

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

Re: Do long-term bonds belong in one's portfolio?

Post by vineviz » Mon Dec 10, 2018 5:35 pm

international001 wrote:
Mon Dec 10, 2018 4:41 pm
Ideally I would like to have LT because the lowest correlation with stocks and higher return. Never mind they are more volatile, they historically better with stocks.
But at the moment I have IT because the little extra benefit of LT.

How will I know to switch from IT to LT?
You won’t. This is one of the chief problems with trying to tim the market.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

HEDGEFUNDIE
Posts: 4038
Joined: Sun Oct 22, 2017 2:06 pm

Re: Do long-term bonds belong in one's portfolio?

Post by HEDGEFUNDIE » Mon Dec 10, 2018 5:38 pm

willthrill81 wrote:
Mon Dec 10, 2018 5:29 pm
HEDGEFUNDIE wrote:
Mon Dec 10, 2018 5:26 pm
willthrill81 wrote:
Mon Dec 10, 2018 5:13 pm
Even though the period from 1978-2018 (data in PV) mostly consisted of a bond bull market from 1982-2012 due to declining interest rates, the Sharpe ratio of long-term Treasuries was significantly lower than that of U.S. stocks (.37 vs. .51). So to the extent that that holds true going forward, it wouldn't make sense to include LTT in your portfolio if you're trying to maximize your absolute or risk-adjusted returns. If that was your goal, you should just increase your stock allocation.
So misleading. No one holds 100% bonds in isolation. When paired with an 60-80% stock allocation, LTT delivered a better portfolio Sharpe Ratio than ITT or STT:

https://www.portfoliovisualizer.com/bac ... easury1=40

Negative correlations for the win!
Your analysis compared LTT to STT and ITT. Of course LTT beat out the others over a 40 year period. But if you compared your 60% TSM / 40% LTT portfolio to 100% stocks, the former had a slightly higher Sharpe ratio but a lower absolute return (10.71% vs. 11.46%).
I think this is the first time someone has compared a 100% stock portfolio with a 60/40 blend. Of course the 100% stock portfolio did better, at 50% higher volatility!

Long bonds are still bonds. They are less volatile than stocks and return less as well. Thus they serve the same purpose as Total Bond Market in Boglehead orthodoxy.

Elysium
Posts: 2142
Joined: Mon Apr 02, 2007 6:22 pm

Re: Do long-term bonds belong in one's portfolio?

Post by Elysium » Mon Dec 10, 2018 6:16 pm

international001 wrote:
Mon Dec 10, 2018 4:41 pm
Ideally I would like to have LT because the lowest correlation with stocks and higher return. Never mind they are more volatile, they historically better with stocks.
But at the moment I have IT because the little extra benefit of LT.

How will I know to switch from IT to LT?
I would like to own LT bonds, if there is a strong case for it. I have been following these discussions for a long time, and one thing that almost every one of the experts have agreed on is to keep bond durations shorter. Even those who disagree strongly on other contentious issues have agreed on this one aspect.

How will we know when it is attractive? I don't think we have an exact formula. Back in the days, Larry used to point to some ideas to determine the sweet spot based on the yield curve, but then almost always came down to recommending short / intermediate bonds. Perhaps someone who closely follows his work lately have an answer on his current thinking.

I guess the point is we don't have to worry about it, as we could get most of the benefits by holding IT bonds.

ThrustVectoring
Posts: 771
Joined: Wed Jul 12, 2017 2:51 pm

Re: Do long-term bonds belong in one's portfolio?

Post by ThrustVectoring » Mon Dec 10, 2018 8:22 pm

hdas wrote:
Mon Dec 10, 2018 4:26 pm
ThrustVectoring wrote:
Mon Dec 10, 2018 3:38 pm
Given what I know of the treasury futures market, I feel like I have to be missing something for long-term bonds to make any sort of sense over levering up short-term bonds. Maybe convexity?

Let's compare investment options for $100k in the 10-year cash bond vs the 2-year treasury futures contract. As always, latest numbers come from straight from CMEgroup.

Cash bond DV01: $84.43

2-year contract DV01: $32.71

So instead of holding the cash bond, you could take out two contracts and have somewhat less exposure to interest rate risk. Now, for a very rough expected return, calculate the return if interest rates remain flat: you get the treasury futures yield on the notional amount, minus the implied repo rate for the financed portion. This is $400k * 2.74% - $300k * 2.33% = $3970. For the cash bond, just assume yield times notional: 2.77% * $100k = $2770.

Again, I feel like I have to be missing something pretty big here. Convexity maybe? Futures market pricing in expected fed activity? I really don't know. My fairly naive analysis suggests that you can earn more money with less risk through levering up short-term bonds, even though I have pretty strong efficient-market intuitions. The long-term bond buyers have to be buying something - maybe it's regulatory compliance? The yield curve makes zero sense to me otherwise.
Aren't you missing that you can put say 90% in Bills and brake it in pieces as needed to fund the 2 contracts, getting you extra yield if needed?

Also, could you explain how do you get the 2.74% and 2.33% ?

Thanks
The treasury bills thing is baked in - the cost of financing is roughly equal to the 3-mo treasury bill rate, and is basically what the "implied repo rate" is. If you add in the yield from investing in 3-mo treasury bills, you have to say you're financing the full $400k notional, so it nets out to zero. Math is easier IMO if you just remove the cash backing the position from your financing cost estimate.

The 2.74% is the futures yield from the quikstrike link I put up earlier. The 2.33% was the implied repo rate, also from the same place. Generally speaking, you can use the 3-mo treasury bill yield instead, and arbitrage during market hours should basically drive the implied repo rate to within a few basis points of the 3-mo treasury bill rate. This is because the implied repo rate is the theoretical yield of buying an underlying bond, selling the futures contract, and delivering that bond into the futures contract at expiry. If you believe the central clearing party in the futures market holds zero counter-party risk (and they basically don't), then this is equivalent to buying a treasury bill that matures on expiry of the futures contract.
Current portfolio: 60% VTI / 40% VXUS

User avatar
willthrill81
Posts: 16129
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Do long-term bonds belong in one's portfolio?

Post by willthrill81 » Mon Dec 10, 2018 10:15 pm

Elysium wrote:
Mon Dec 10, 2018 6:16 pm
international001 wrote:
Mon Dec 10, 2018 4:41 pm
Ideally I would like to have LT because the lowest correlation with stocks and higher return. Never mind they are more volatile, they historically better with stocks.
But at the moment I have IT because the little extra benefit of LT.

How will I know to switch from IT to LT?
I would like to own LT bonds, if there is a strong case for it. I have been following these discussions for a long time, and one thing that almost every one of the experts have agreed on is to keep bond durations shorter. Even those who disagree strongly on other contentious issues have agreed on this one aspect.

How will we know when it is attractive? I don't think we have an exact formula. Back in the days, Larry used to point to some ideas to determine the sweet spot based on the yield curve, but then almost always came down to recommending short / intermediate bonds. Perhaps someone who closely follows his work lately have an answer on his current thinking.

I guess the point is we don't have to worry about it, as we could get most of the benefits by holding IT bonds.
Yes, the marginal benefit in terms of returns has been significantly greater with moving from short-term to intermediate-term bonds than from moving from intermediate-term to long-term bonds.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

stlutz
Posts: 5537
Joined: Fri Jan 02, 2009 1:08 am

Re: Do long-term bonds belong in one's portfolio?

Post by stlutz » Tue Dec 11, 2018 12:38 am

Since we've been talking about stock/bond correlations in this thread,I thought this piece was semi-interesting:

https://www.etf.com/sites/default/files ... TF.com.pdf

Angst
Posts: 2131
Joined: Sat Jun 09, 2007 11:31 am

Re: Do long-term bonds belong in one's portfolio?

Post by Angst » Tue Dec 11, 2018 1:47 am

stlutz wrote:
Tue Dec 11, 2018 12:38 am
Since we've been talking about stock/bond correlations in this thread,I thought this piece was semi-interesting:

https://www.etf.com/sites/default/files ... TF.com.pdf
Thank you for posting it. I agree it's half-way interesting, but I'm not really confident if I should make too much of it. Still I suppose I'll go ahead and feel even better now about the retirement LMP of TIPS I've been accumulating over much of the last decade. I wonder why a Fidelity document is being served up from etf.com?

fennewaldaj
Posts: 847
Joined: Sun Oct 22, 2017 11:30 pm

Re: Do long-term bonds belong in one's portfolio?

Post by fennewaldaj » Tue Dec 11, 2018 2:28 am

Valuethinker wrote:
Sun Dec 09, 2018 8:03 am
fennewaldaj wrote:
Sun Dec 09, 2018 2:15 am


I suppose there is one more group of us bond investors that hold things such as high yield corporate and EM bonds. Not sure how exactly to categorize this group.
Orthogonal risk

I don't have Anti Illmanen's books on asset allocation to hand but I think of them as the bible for such things.

I read Anti Illmanen's book and quite enjoyed it (actually I liked it so much I read it twice). It seems he has a lukewarm positive opinion on HY (BB and Bs) and EM debt but perhaps I am reading by biases into what he wrote. It seems HY debt we have enough history we can kinda expect it to behave similar to how it did in the past. It does seem debatable how disconnected credit and equity risk are. I looking at the 3 trouble periods for HY and the 87 stock market crash. HY sailed right through the 87 debacle. In the early 90s neither HY or US stocks did all that bad (I assume the high interest rates on HY were overcoming the defaults) 2000-2002 HY sailed right through while US stocks did horrid. 2008 is discussed a lot here basically HY was about half as bad as stocks. Eyeballing these bad times they don't seem to behave in a perfectly predictable way when stocks do poorly. Perhaps when stocks tumble due to some weird technical thing (87) or mostly due to valuations (2000-2002) HY is fine while it is worse in bad economies (2008).

EM market bonds are a bit more of a wild card. Its hard to use past data to predict the behavior of the asset class as these economies have been maturing and thus some spread narrowing is to be expected. This could continue or reverse somewhat. Which one happens has a big impact on how this asset class does.

I tend to own large assets classes unless I can find a good reason not to. In the case of HY and EM bonds I have not found a compelling reason not to own them. I split my bond allocation as follows 50% Total bond 20% TIPs 15% HY 15% ex US (mostly EM). I upped my bond allocation from 20% to 22% to account for my riskier bonds. I will eventually have 50% bonds with ~15% total in these riskier segments. If I had full control of the funds I use I would likely hold treasuries and investment grade corporate in separate funds but my 403b has total bond so I am using it.

patrick
Posts: 1706
Joined: Fri Sep 04, 2009 3:39 am
Location: Mega-City One

Re: Do long-term bonds belong in one's portfolio?

Post by patrick » Tue Dec 11, 2018 4:43 am

longinvest wrote:
Sat Dec 08, 2018 6:56 pm
Of course, STRIPS aren't speculation for a pension fund with nominal liabilities! You're changing the context of my statement.

I have yet to hear a retail investor tell me of a nominal liability of his in 20 to 30 years that would be best covered by investing the money into an ETF of "equal-weighted par" STRIPS that doesn't go down in duration over time.

Are you such a retail investor? What's the amount and date of your nominal liability? How does EDV help you with with covering it?
I doubt that many retail investors would have nominal liabilities that are best covered by more ordinary long-term bond funds either. With the exception of mortgages, which would often be better prepaid instead, is there any other common nominal liability beyond the very short term? Even for those who do have a nominal liability known 10+ years in advance, the regular long-term bond fund also keeps its duration at 10+ years rather than reducing its duration as the liability becomes closer.

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

Re: Do long-term bonds belong in one's portfolio?

Post by vineviz » Tue Dec 11, 2018 7:45 am

Angst wrote:
Tue Dec 11, 2018 1:47 am
stlutz wrote:
Tue Dec 11, 2018 12:38 am
Since we've been talking about stock/bond correlations in this thread,I thought this piece was semi-interesting:

https://www.etf.com/sites/default/files ... TF.com.pdf
Thank you for posting it. I agree it's half-way interesting, but I'm not really confident if I should make too much of it. Still I suppose I'll go ahead and feel even better now about the retirement LMP of TIPS I've been accumulating over much of the last decade. I wonder why a Fidelity document is being served up from etf.com?
ETF.com is an industry site (owned by the CBOE, the ETF trading exchange) and they take advertising revenue to serve up "commentary" like this.

Doesn't make the commentary invalid (I found the Fidelity piece somewhat interesting, for instance), but it always pays to know who's paying.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
siamond
Posts: 5200
Joined: Mon May 28, 2012 5:50 am

Re: Do long-term bonds belong in one's portfolio?

Post by siamond » Tue Dec 11, 2018 7:51 am

aristotelian wrote:
Sat Dec 08, 2018 9:25 am
I have thought about this as well and in fact posted similar ideas. Ultimately I have decided against them as the core of my bond allocation.

1) Because duration is so long, it may be many years for prices to recover. Many investors do not have that long of a timeframe, especially on the bond side.

2) If you are seeking higher return, it is much easier to simply increase % of stocks.

3) Most investors are looking for stability and income on the bond side. Long bonds are just as volatile as stocks.

4) Correlation is zero but not negative. There may be a Black Swan where bond crash correlates with a stock crash.
I was going to post something along the same exact lines... +1.

I would just add that I view my position in (Intermediate) Bonds as a glorified emergency fund, mostly intended for rare times of duress and/or distress. Therefore I seek reasonable stability for it (I don't mind if it occasionally goes down 5%, but I do mind if it could go down 20% or more).

EDIT: clarified my thought.
Last edited by siamond on Tue Dec 11, 2018 11:05 am, edited 1 time in total.

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

Re: Do long-term bonds belong in one's portfolio?

Post by vineviz » Tue Dec 11, 2018 7:59 am

patrick wrote:
Tue Dec 11, 2018 4:43 am
longinvest wrote:
Sat Dec 08, 2018 6:56 pm
Of course, STRIPS aren't speculation for a pension fund with nominal liabilities! You're changing the context of my statement.

I have yet to hear a retail investor tell me of a nominal liability of his in 20 to 30 years that would be best covered by investing the money into an ETF of "equal-weighted par" STRIPS that doesn't go down in duration over time.

Are you such a retail investor? What's the amount and date of your nominal liability? How does EDV help you with with covering it?
I doubt that many retail investors would have nominal liabilities that are best covered by more ordinary long-term bond funds either. With the exception of mortgages, which would often be better prepaid instead, is there any other common nominal liability beyond the very short term? Even for those who do have a nominal liability known 10+ years in advance, the regular long-term bond fund also keeps its duration at 10+ years rather than reducing its duration as the liability becomes closer.
I'll repeat something I said earlier, which is that many investors own bonds for purposes that don't include liability matching. Liability matching is ONE approach to bond portfolio management, but not the only one.

Furthermore, I'll add that retail investors don't generally have fixed long-term liabilities of ANY sort, real or nominal, of the kind that pension funds and insurance companies have. Apart from my mortgage, for instance, I don't currently have any fixed liabilities due more than 60 days from today.

So even though some of the concepts from liability matching portfolios might be useful to retail investors, it's not self-evident that all of them are. And even though some of the expected future expenditures that most investors have are subject to ravages of general inflation, it's not self-evident that most of them are or even that future expenditures are BETTER modeled by real rates than nominal rates.

I understand why that is PROBABLY true in aggregate and MIGHT be true for some investors, but it's an empirical question for which there is (AFAIK) no credible evidence. Future levels of personal expenditure, especially decades or more from now, are so idiosyncratic for each investor that I'm not sure there is an obvious way to plan for them with enough precision to suggest that TIPS are definitively better or worse than nominal bonds.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

longinvest
Posts: 4131
Joined: Sat Aug 11, 2012 8:44 am

Re: Do long-term bonds belong in one's portfolio?

Post by longinvest » Tue Dec 11, 2018 8:37 am

patrick wrote:
Tue Dec 11, 2018 4:43 am
longinvest wrote:
Sat Dec 08, 2018 6:56 pm
Of course, STRIPS aren't speculation for a pension fund with nominal liabilities! You're changing the context of my statement.

I have yet to hear a retail investor tell me of a nominal liability of his in 20 to 30 years that would be best covered by investing the money into an ETF of "equal-weighted par" STRIPS that doesn't go down in duration over time.

Are you such a retail investor? What's the amount and date of your nominal liability? How does EDV help you with with covering it?
I doubt that many retail investors would have nominal liabilities that are best covered by more ordinary long-term bond funds either. With the exception of mortgages, which would often be better prepaid instead, is there any other common nominal liability beyond the very short term? Even for those who do have a nominal liability known 10+ years in advance, the regular long-term bond fund also keeps its duration at 10+ years rather than reducing its duration as the liability becomes closer.
If markets dried up, tomorrow, an index fund of 2,141 long-term bonds like VBLTX wouldn't need to make any trade to receive coupons and use them to continue making monthly payments to its fund investors.

A STRIPS is a derivative product with embedded hidden fees. The stripping process has costs and those who do the stripping want to make a profit. Despite these hidden fees, a single STRIPS could still be attractive for retails investors with a nominal liability in the future. They might be willing to pay these undisclosed hidden fees in exchange of certainty of a precise nominal payment on a specific future date.

EDV is different. It's a portfolio of only 81 STRIPS kept within a maturity range of 20 to 30 years. It isn't capitalization-weighted; it aims to weight its holdings on an equal par value basis. If markets dried up, it would take 20 years before EDV would receive its first payment. In normal times, EDV's objective is to sell STRIPS when they fall below 20 years to maturity and reinvest the proceeds within the target maturity range. EDV never receives payments; the only way it can pay for expenses (Expense Ratio) and make some payments to its investors is as a result of trading or through the ETF creation and redemption process.

Just for comparison, the average maturity of VBLTX's bonds is 23.9 years. At 2 coupons per year, it's similar to a collection of ((23.9 X 2) coupons + 1 principal) X 2,141 = 104,481 STRIPS* with an average 14.6 years duration and without embedded hidden fees (VBLTX's low expense ratio isn't hidden). VBLTX is a diversified bond investment; EDV with its 81 STRIPS isn't.

* Not exactly, because a STRIPS is a stripped Treasury. But, I'm just trying to make a quick comparison, here.

Let me repeat Vanguard's warning on EDV's web page: "The fund is primarily intended for institutional investors with extremely long-term liabilities—20 years or more. Prospective individual investors are urged to consult with their own advisors to determine if the fund is suitable for their overall investment programs and financial positions."

Unsurprisingly, Vanguard's long-term bond index fund (VBLTX) has no such warning.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW

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

Re: Do long-term bonds belong in one's portfolio?

Post by vineviz » Tue Dec 11, 2018 10:17 am

longinvest wrote:
Tue Dec 11, 2018 8:37 am
A STRIPS is a derivative product with embedded hidden fees. The stripping process has costs and those who do the stripping want to make a profit.
I don't understand the paranoia here. STRIPS are incredibly simple to understand, and much easier to model than an unstripped bond is. The cost of separating the coupons from the principal payment is not "hidden" any more than the cost of trading any other bond is hidden: the yield-to-maturity at purchase fully reflects all the costs. When you enter an order to buy a zero-coupon bond you know EXACTLY what your yield is, just like when you buy a coupon-bond is. And there are no ongoing expenses with zero-coupon bonds, just like there are none with coupon bonds.
longinvest wrote:
Tue Dec 11, 2018 8:37 am
EDV is different. It's a portfolio of only 81 STRIPS kept within a maturity range of 20 to 30 years. It isn't capitalization-weighted; it aims to weight its holdings on an equal par value basis. If markets dried up, it would take 20 years before EDV would receive its first payment.
It's not different: the payouts are the same no matter whether you hold the zeros in a bond fund or individually in a brokerage account.

And any investor who thinks that not receiving a coupon payment is a downside is NOT an investor who should be buying zero-coupon bonds to begin with. In other words, zero-coupon bond buyers are choosing these bonds precisely BECAUSE they don't make regular coupon payments
longinvest wrote:
Tue Dec 11, 2018 8:37 am
VBLTX is a diversified bond investment; EDV with its 81 STRIPS isn't.
If VBLTX is more diversified than EDV it has nothing to do with the parts bolded here. The number of assets is absolutely NOT a reasonable measure of diversification.

It's true that VBLTX owns a wider variety of bonds than EDV, but it is the benefit to AVOIDING that variety which makes EDV such a potentially useful investment.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

international001
Posts: 1235
Joined: Thu Feb 15, 2018 7:31 pm

Re: Do long-term bonds belong in one's portfolio?

Post by international001 » Tue Dec 11, 2018 11:40 am

HEDGEFUNDIE wrote:
Mon Dec 10, 2018 5:01 pm
international001 wrote:
Mon Dec 10, 2018 4:41 pm
Ideally I would like to have LT because the lowest correlation with stocks and higher return. Never mind they are more volatile, they historically better with stocks.
But at the moment I have IT because the little extra benefit of LT.

How will I know to switch from IT to LT?
Apparently you will “know” it when you see it, as Elysium and other posters here are doing. Obvious market timing that would be readily dismissed in any thread about stocks.
Well.. Know when you see is how many people pick stocks. I'd like something more rational

For those who say to just hold more stocks, it's not good enough. Just looking at PV's EF, last 40 years using LT in a 60/40 porfolio improves IT for 1% (same risk)

You can argue that next 40 years won't be like the past 40, but then it becomes a question of market decade-timing, what I don't really know how to do. We can start market decade-timing to know how many international stocks you can hold, etc.

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

Re: Do long-term bonds belong in one's portfolio?

Post by vineviz » Tue Dec 11, 2018 12:07 pm

international001 wrote:
Tue Dec 11, 2018 11:40 am
HEDGEFUNDIE wrote:
Mon Dec 10, 2018 5:01 pm
international001 wrote:
Mon Dec 10, 2018 4:41 pm
Ideally I would like to have LT because the lowest correlation with stocks and higher return. Never mind they are more volatile, they historically better with stocks.
But at the moment I have IT because the little extra benefit of LT.

How will I know to switch from IT to LT?
Apparently you will “know” it when you see it, as Elysium and other posters here are doing. Obvious market timing that would be readily dismissed in any thread about stocks.
Well.. Know when you see is how many people pick stocks. I'd like something more rational
But trying to time the bond market isn't entirely rational to begin with, right? All the evidence tells us that it is nearly impossible.

Over the past month, Vanguard Long-Term Treasury Index ETF (VGLT) has outperformed Vanguard Total Bond Market Index ETF (BND) by something like 360 bps.

What was the signal on November 1st that said "now it's safe to buy long-term bonds"? And how many "experts" saw it?
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

ThrustVectoring
Posts: 771
Joined: Wed Jul 12, 2017 2:51 pm

Re: Do long-term bonds belong in one's portfolio?

Post by ThrustVectoring » Tue Dec 11, 2018 1:55 pm

international001 wrote:
Tue Dec 11, 2018 11:40 am
For those who say to just hold more stocks, it's not good enough. Just looking at PV's EF, last 40 years using LT in a 60/40 porfolio improves IT for 1% (same risk)
What about holding the same amount of stocks and more bonds of a shorter duration? Leverage is extraordinarily cheap in the highly liquid treasury futures market, so it's pretty straightforward to grab the DV01 of the cash bond you're considering buying and the DV01 of a shorter-term futures contract and buy those up until you've gotten your desired interest rate sensitivity. It's no more risky (by the measure of dollar change per basis point), and you get to use whatever part of the yield curve is most attractive.
Current portfolio: 60% VTI / 40% VXUS

Post Reply