When to add 30 year bonds to one's portfolio?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Tigermoose
Posts: 823
Joined: Sat Jul 21, 2012 5:02 pm

When to add 30 year bonds to one's portfolio?

Post by Tigermoose »

Eventually I would like to add some 30 year bonds to my portfolio, when the bond market returns to normal (Let's not get into why this isn't "normal"). Assuming the bond market gets back to normal and interest rates rise on longer duration bonds, at what % yield would it begin making sense to get some of these to add diversification and crash protection to one's portfolio?

5%?
Institutions matter
linguini
Posts: 262
Joined: Tue Oct 16, 2012 2:49 pm

Re: When to add 30 year bonds to one's portfolio?

Post by linguini »

Buy some ee savings bonds now and you'll likely be earning a positive real return. No need to wait.
Call_Me_Op
Posts: 9881
Joined: Mon Sep 07, 2009 2:57 pm
Location: Milky Way

Re: When to add 30 year bonds to one's portfolio?

Post by Call_Me_Op »

Tigermoose wrote:Eventually I would like to add some 30 year bonds to my portfolio, when the bond market returns to normal (Let's not get into why this isn't "normal"). Assuming the bond market gets back to normal and interest rates rise on longer duration bonds, at what % yield would it begin making sense to get some of these to add diversification and crash protection to one's portfolio?

5%?
I don't see how anyone could give you anything better than a pure guess in response to this question. Sure, 5% is as good a guess as any. We may get there next year, or it may take another 20 years or more.
Best regards, -Op | | "In the middle of difficulty lies opportunity." Einstein
User avatar
bobcat2
Posts: 6076
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: When to add 30 year bonds to one's portfolio?

Post by bobcat2 »

I wouldn't add 30 year bonds to diversify. I would add 30 year bonds when I had a liability in 30 years that I wanted to match with maturing assets.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
Call_Me_Op
Posts: 9881
Joined: Mon Sep 07, 2009 2:57 pm
Location: Milky Way

Re: When to add 30 year bonds to one's portfolio?

Post by Call_Me_Op »

bobcat2 wrote:I wouldn't add 30 year bonds to diversify. I would add 30 year bonds when I had a liability in 30 years that I wanted to match with maturing assets.

BobK
But how many people have fixed nominal liabilities 30 years into the future? A better approach is to buy a TIPS bond.
Best regards, -Op | | "In the middle of difficulty lies opportunity." Einstein
Johm221122
Posts: 6395
Joined: Fri May 13, 2011 6:27 pm

Re: When to add 30 year bonds to one's portfolio?

Post by Johm221122 »

linguini wrote:Buy some ee savings bonds now and you'll likely be earning a positive real return. No need to wait.
Or I bonds would fit OP goals now
John
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: When to add 30 year bonds to one's portfolio?

Post by dbr »

Savings bonds ok, but long term treasury or corporate bonds probably don't make sense in most portfolios most of the time any time. A ladder of TIPS with some long term components is a discussion, but that probably is not what you are talking about.

Are you trying to market time interest rates here?

Good bonds for diversification and to dilute volatility would be in the short to intermediate durations.

I agree with Bobcat above.
Topic Author
Tigermoose
Posts: 823
Joined: Sat Jul 21, 2012 5:02 pm

Re: When to add 30 year bonds to one's portfolio?

Post by Tigermoose »

Two Reasons
1. I'm thinking that 30 year bonds offer significant protection and diversification in a 2008-9 type event. Significant capital appreciation occurs for long term bonds during a major market crash. Sure, short term bonds and fixed income offers some ballast to your portfolio during these events, but I am thinking that long duration bonds offer offsetting behavior during major crashes. Seeing how these seem to come along every few years, I thought it would be a good diversifier for my portfolio. I already have Stable Value, TIPs, and short term bond funds in my portfolio to serve as ballast.

2. I also want to have these in my portfolio as a fixed income source during my retirement in case we have another period of super low interest rates during my retirement years. I am seeing how difficult it is for retirees to live off their savings at the current rates, and so I want to be proactive about securing bonds that offer a decent return during my retirement years. I'm only 37, so I can see that it might take a while for rates to go above -- say 5% -- and I want to know what a good time would be to start buying some of these bonds to minimize my risk of having low interest rates during my retirement years.
Institutions matter
Johm221122
Posts: 6395
Joined: Fri May 13, 2011 6:27 pm

Re: When to add 30 year bonds to one's portfolio?

Post by Johm221122 »

I would want a lot more than 5% for buying a 30 year bond, at least 7% probably 8%
John
Edit, this may interest you
http://www.bogleheads.org/forum/viewtop ... arket+time
John
Call_Me_Op
Posts: 9881
Joined: Mon Sep 07, 2009 2:57 pm
Location: Milky Way

Re: When to add 30 year bonds to one's portfolio?

Post by Call_Me_Op »

Johm221122 wrote:I would want a lot more than 5% for buying a 30 year bond, at least 7% probably 8%
John
Edit, this may interest you
http://www.bogleheads.org/forum/viewtop ... arket+time
John
People were afraid to buy 10+% long bonds in the early 1980's. All depends upon where you think rates are going.
Best regards, -Op | | "In the middle of difficulty lies opportunity." Einstein
User avatar
kenyan
Posts: 3015
Joined: Wed Jan 12, 2011 11:16 pm

Re: When to add 30 year bonds to one's portfolio?

Post by kenyan »

Call_Me_Op wrote:
Johm221122 wrote:I would want a lot more than 5% for buying a 30 year bond, at least 7% probably 8%
John
Edit, this may interest you
http://www.bogleheads.org/forum/viewtop ... arket+time
John
People were afraid to buy 10+% long bonds in the early 1980's. All depends upon where you think rates are going.
Right. If long bonds are offering 8%, but inflation is at 7%, are you going to have the stomach to jump in, just because you previously said that would be your breakpoint? I'm sure that in that scenario, the talking heads would all be discussing how high inflation is the "new normal" and the price paid for years of monetary excess.
Retirement investing is a marathon.
Johm221122
Posts: 6395
Joined: Fri May 13, 2011 6:27 pm

Re: When to add 30 year bonds to one's portfolio?

Post by Johm221122 »

Inflation and fear of inflation is why I would not consider 5%, I woukd think about it at 8% and really think about at 10%.
John
User avatar
bobcat2
Posts: 6076
Joined: Tue Feb 20, 2007 2:27 pm
Location: just barely Outside the Beltway

Re: When to add 30 year bonds to one's portfolio?

Post by bobcat2 »

Tigermoose wrote: I'm thinking that 30 year bonds offer significant protection and diversification in a 2008-9 type event. Significant capital appreciation occurs for long term bonds during a major market crash. Sure, short term bonds and fixed income offers some ballast to your portfolio during these events, but I am thinking that long duration bonds offer offsetting behavior during major crashes. Seeing how these seem to come along every few years, I thought it would be a good diversifier for my portfolio. I already have Stable Value, TIPs, and short term bond funds in my portfolio to serve as ballast.
Unless you intend to sell your 30 year bonds during a major market crash what difference will holding them make?

I also want to have these in my portfolio as a fixed income source during my retirement in case we have another period of super low interest rates during my retirement years. I am seeing how difficult it is for retirees to live off their savings at the current rates, and so I want to be proactive about securing bonds that offer a decent return during my retirement years. I'm only 37, so I can see that it might take a while for rates to go above -- say 5% -- and I want to know what a good time would be to start buying some of these bonds to minimize my risk of having low interest rates during my retirement years.
It would appear that the best time for you to buy them would be when they go above 5%. However, this won't necessarily protect you from low interest rates when you are retired. If interest rates become very low again after you buy the bonds, you will have to reinvest the coupons at the prevailing very low interest rates. :( You could avoid this reinvestment risk problem by purchasing zero coupon bonds, but then you will have no interest payments. :|

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
Topic Author
Tigermoose
Posts: 823
Joined: Sat Jul 21, 2012 5:02 pm

Re: When to add 30 year bonds to one's portfolio?

Post by Tigermoose »

bobcat2 wrote:Unless you intend to sell your 30 year bonds during a major market crash what difference will holding them make?
I would rebalance my assett allocation -- which would mean that I sell some of the long term bonds to buy equities to get my allocations back in balance.

bobcat2 wrote:It would appear that the best time for you to buy them would be when they go above 5%. However, this won't necessarily protect you from low interest rates when you are retired. If interest rates become very low again after you buy the bonds, you will have to reinvest the coupons at the prevailing very low interest rates. :( You could avoid this reinvestment risk problem by purchasing zero coupon bonds, but then you will have no interest payments. :|
Yes, I guess the best case would be to lock in some 30 year bonds at 6%+ for each year during my 40-60 year age range. That way, these would only start getting redeemed when I was 70+ years of age. At that age, I should have no problem having to slowly eat into my capital stock if necessary if I was in a low interest rate environment.
Institutions matter
linguini
Posts: 262
Joined: Tue Oct 16, 2012 2:49 pm

Re: When to add 30 year bonds to one's portfolio?

Post by linguini »

Johm221122 wrote:Inflation and fear of inflation is why I would not consider 5%, I woukd think about it at 8% and really think about at 10%.
John
You would consider turning down a 30-year treasury bond at 10% interest in the current environment? At those rates, unless inflation rockets up within the next ten years, real return is so high that you make back your initial investment in the first ten years. If you are too risk-averse to tie money in 10% nominal yield guaranteed, why would you be willing to invest in the stock market which is far more volatile than inflation and at no higher return? Surely you're exaggerating that you wouldn't take that deal for at least part of your portfolio?
User avatar
rmelvey
Posts: 826
Joined: Sat Sep 18, 2010 5:17 pm
Contact:

Re: When to add 30 year bonds to one's portfolio?

Post by rmelvey »

I have a quarter of money in 30 year bonds. Jump in. The water is fine. You buying them today is no different from me already holding them. They are a great diversifier with respect to equities (or gold/commodities) and the combination of low correlation and high volatility gives you a very powerful volatility harvest over time. The risk free income payments aren't bad either.
Topic Author
Tigermoose
Posts: 823
Joined: Sat Jul 21, 2012 5:02 pm

Re: When to add 30 year bonds to one's portfolio?

Post by Tigermoose »

rmelvey wrote:I have a quarter of money in 30 year bonds. Jump in. The water is fine. You buying them today is no different from me already holding them. They are a great diversifier with respect to equities (or gold/commodities) and the combination of low correlation and high volatility gives you a very powerful volatility harvest over time. The risk free income payments aren't bad either.
I don't see how purchasing them today is better than putting money into my 401k Stable Value Fund paying 3%. Since the bond prices are at historic highs, I don't see much capital appreciation potential. And the SVF yield is greater than the 30 year bond yield.

Now, if in several years the 30 yr bond prices come down and yields go up, then I might see upside to the capital appreciation potential and thus these would make sense.

Am I missing something?
Institutions matter
User avatar
rmelvey
Posts: 826
Joined: Sat Sep 18, 2010 5:17 pm
Contact:

Re: When to add 30 year bonds to one's portfolio?

Post by rmelvey »

Tigermoose wrote:
rmelvey wrote:I have a quarter of money in 30 year bonds. Jump in. The water is fine. You buying them today is no different from me already holding them. They are a great diversifier with respect to equities (or gold/commodities) and the combination of low correlation and high volatility gives you a very powerful volatility harvest over time. The risk free income payments aren't bad either.
I don't see how purchasing them today is better than putting money into my 401k Stable Value Fund paying 3%. Since the bond prices are at historic highs, I don't see much capital appreciation potential. And the SVF yield is greater than the 30 year bond yield.

Now, if in several years the 30 yr bond prices come down and yields go up, then I might see upside to the capital appreciation potential and thus these would make sense.

Am I missing something?
The stable value fund does not have the credit quality of a Treasury bond so they are not directly comparable. They might still make sense, but only a Treasury bond is risk free in nominal terms because only the Treasury can issue an infinite amount of debt to pay back previous debts (they can always roll over their debts because the Fed fixes the price of their debts so primary dealers always bid in the auctions because they are greedy and love arbitrage). No other institution has the same dynamics for their dollar denominated debts.
Topic Author
Tigermoose
Posts: 823
Joined: Sat Jul 21, 2012 5:02 pm

Re: When to add 30 year bonds to one's portfolio?

Post by Tigermoose »

rmelvey wrote:
Tigermoose wrote:The stable value fund does not have the credit quality of a Treasury bond so they are not directly comparable. They might still make sense, but only a Treasury bond is risk free in nominal terms because only the Treasury can issue an infinite amount of debt to pay back previous debts (they can always roll over their debts because the Fed fixes the price of their debts so primary dealers always bid in the auctions because they are greedy and love arbitrage). No other institution has the same dynamics for their dollar denominated debts.
Good point. So the 30 yr bond is still optimal as protection against systematic crashes that could wipe out institutions supporting the SVF insurance.
Institutions matter
User avatar
nisiprius
Advisory Board
Posts: 52216
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: When to add 30 year bonds to one's portfolio?

Post by nisiprius »

I do not see the point of 30 year nominal bonds for an ordinary Boglehead-style investor under any circumstances. The inflation risk is just too severe.
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
rmelvey
Posts: 826
Joined: Sat Sep 18, 2010 5:17 pm
Contact:

Re: When to add 30 year bonds to one's portfolio?

Post by rmelvey »

nisiprius wrote:I do not see the point of 30 year nominal bonds for an ordinary investor under any circumstances. The inflation risk is just too severe.
From a liability matching perspective, you are entirely correct. However, when periodically rolled over (to maintain that long duration) and held in with conjunction with equities and commodities (most notably gold), you have exposure to a wide array of macroeconomic environments. Exposure to a wide degree of macroeconomic environments ensures that your portfolio is likely to be able to survive black swan events. Additionally, the high volatility of LTT can be a blessing rather than a curse because with wide rebalancing bands you can capture intermediate term momentum, long term reversion to the mean, and are able to benefit from volatility harvesting.

Volatility harvesting is something that doesn't appear when looking at measures of single period returns (such as sharpe or sortino) but for an investor who cares about compounding results it makes a big difference. I wrote a post about this here:
http://www.stableinvesting.com/2013/04/ ... demon.html

Additionally, at the end of the article I link to a paper that gives a more formal explanation. Volatility harvesting is one of the few free lunches that exist in an efficient market. Besides, profiting from falling interests rates when everyone has said "they have to go up" gives you a sense of smug satisfaction that I'm sure any Boglehead can appreciate :twisted:

Edit: Additionally, LTT have positive skew so they can mix well with equities bringing your portfolio closer to that well behaved normal distribution.
User avatar
tetractys
Posts: 6249
Joined: Sat Mar 17, 2007 3:30 pm
Location: Along the Salish Sea

Re: When to add 30 year bonds to one's portfolio?

Post by tetractys »

I've been thinking about this question from two angles. 1. When the portfolio is very risky, like with lots of small cap. Or 2. When a portfolio carries a large allocation to cash, like in DB plans where this is unavoidable. -- Tet
Jfet
Posts: 1081
Joined: Tue Dec 21, 2010 6:20 pm

Re: When to add 30 year bonds to one's portfolio?

Post by Jfet »

The EE bond is yielding almost a full point better than the 20 year treasury...and it is tax deferred and state and local tax exempt.

I think I would at least fill out my $10,000 in EE bonds each year before buying 30 year treasury.
User avatar
rmelvey
Posts: 826
Joined: Sat Sep 18, 2010 5:17 pm
Contact:

Re: When to add 30 year bonds to one's portfolio?

Post by rmelvey »

Jfet wrote:The EE bond is yielding almost a full point better than the 20 year treasury...and it is tax deferred and state and local tax exempt.

I think I would at least fill out my $10,000 in EE bonds each year before buying 30 year treasury.
If you are planning to hold until maturity that might be the way to go. However, the EE bond does not allow you to profit from the volatility and it is essentially entirely illiquid if you are treating it as a long term bond because that doubling only happens on the last year, cashing out early has a huge penalty..

In my mind an EE-bonds are best used as a cash substitute (with the thought of rolling it over if ST rates rise) but with an embedded call option on rates staying low. It is an interesting product (probably with some free lunch in there), but I don't think it is directly comparable to a 30 year Treasury held in a brokerage account.
EDN
Posts: 528
Joined: Sat Dec 29, 2012 6:55 pm
Location: Oklahoma City, OK

Re: When to add 30 year bonds to one's portfolio?

Post by EDN »

Rmelvey,

I'm not so sure things are as cut and dry as you assume. Just off the top of my head:

#1 --The utility of extended duration bonds (i.e. longer than long-term) is a recent phenomenon. I'm not sure you would have come to the same conclusion looking at the previous few decades ending in 1999 -- major stock market declines were better offset with short-term bonds than long-term bonds. The later tended to be negative when the stock premium went south, and I'm only talking about 20 year bonds. 30 year would have been highly negative on average. Even during the deflationary Great Depression, 5YR bonds made out as well/slightly better than 20YR bonds. 1973-1974? VERY ugly for long-term debt. This "new normal" for bond behavior is far from a sure thing.

#2 -- But let's say the world since 2000 is the "new normal" for ultra-long term treasuries in that they always go straight up when equities go down. That is close to -1.0 correlation with equities, and any asset class with such behavior will also carry negative returns, and the more negative the correlation, the more negative the return. Think about it, if there were an investment with negative correlation with equities but high returns, everyone would sell everything and just buy that investment! Until, of course, prices rose to the point of negative expected returns. So you have to consider the drag that negative returns will have on your portfolio. This is different than the current depressed level of short-term rates. Over time we'd expect short-term bonds (zero correlation to equities, not negative) to have a slight real return and very low risk.

#3 -- If we don't experience a "new normal", and longer bonds aren't the equity hedge they've been lately, investors are going to demand much higher returns to compensate for the additional risk. That could mean a veritable bloodbath for extended duration bonds as yields could rise back to 5-7%. If you are in retirement with a combo of stocks and extended duration bonds and both get hit hard, it's probably game over without suspending withdrawals indefinitely.

#4 -- In retirement, one of your biggest risks is spending power, and of course interest rates tend to rise sharply with increases in inflation. Extended duration bonds are a particularly poor vehicle in this scenario, as their losses during a period of sharply rising rates would be catastrophic. And if rising rates lead to a recession, ala 1981, and stocks fall as well, you are in a real bad spot. I'm not sure TIPS (lets say 0% real returns) are going to provide enough protection here either, so there isn't much of a way to diversify away this risk.

Here is what I would say:

(a) don't ignore the longer-term evidence on bonds. Only about 5YRs in maturity gets you 95% of the TERM premium beyond 1mo bills. And historically, 5YRs has been as good as longer bonds in offsetting equity declines (better from 1928-1999, worse in the 2000s)

(b) stick with something like Vanguard Int'd Treasury while you are accumulating, then consider adding (or replacing 5YR) with ST Bond Index and TIPS. You may put up with a bit more short-term volatility, but you will have significant liquidity in terms of low risk and real returning bonds which will act as a bridge to get you through stock bear markets

(c) while working, or in retirement, equities are the growth engine of your portfolio. Yes, low rates have made things more difficult, but only for those with 60%, 70%, 80% or more of their portfolio in bonds. A more balanced mix, even for a retiree, spread across small and value (which are also a benefit in inflationary times) should carry the water for your portfolio even if low risk bonds disappoint.

(d) learn lessons from big institutional investors and endowments who thought abandoning traditional equities and fixed income in favor of all sorts of alternatives would allow them to still get high returns without high risk -- there are no combos of precious metals and extended duration bonds in combo with equities that will produce the same high returns as just stocks and bonds with less risk. Either you get less risk and much less returns (not good), or same returns or slightly better with much more risk. A look at the last 5/10 years of hundred-million and billion dollar endowments and institutions finds they got less returns and more risk, which is also possible when trying to over-engineer a portfolio.

Just my view.

Eric
User avatar
rmelvey
Posts: 826
Joined: Sat Sep 18, 2010 5:17 pm
Contact:

Re: When to add 30 year bonds to one's portfolio?

Post by rmelvey »

EDN wrote:Rmelvey,

I'm not so sure things are as cut and dry as you assume. Just off the top of my head:

#1 --The utility of extended duration bonds (i.e. longer than long-term) is a recent phenomenon. I'm not sure you would have come to the same conclusion looking at the previous few decades ending in 1999 -- major stock market declines were better offset with short-term bonds than long-term bonds. The later tended to be negative when the stock premium went south, and I'm only talking about 20 year bonds. 30 year would have been highly negative on average. Even during the deflationary Great Depression, 5YR bonds made out as well/slightly better than 20YR bonds. 1973-1974? VERY ugly for long-term debt. This "new normal" for bond behavior is far from a sure thing.

#2 -- But let's say the world since 2000 is the "new normal" for ultra-long term treasuries in that they always go straight up when equities go down. That is close to -1.0 correlation with equities, and any asset class with such behavior will also carry negative returns, and the more negative the correlation, the more negative the return. Think about it, if there were an investment with negative correlation with equities but high returns, everyone would sell everything and just buy that investment! Until, of course, prices rose to the point of negative expected returns. So you have to consider the drag that negative returns will have on your portfolio. This is different than the current depressed level of short-term rates. Over time we'd expect short-term bonds (zero correlation to equities, not negative) to have a slight real return and very low risk.

#3 -- If we don't experience a "new normal", and longer bonds aren't the equity hedge they've been lately, investors are going to demand much higher returns to compensate for the additional risk. That could mean a veritable bloodbath for extended duration bonds as yields could rise back to 5-7%. If you are in retirement with a combo of stocks and extended duration bonds and both get hit hard, it's probably game over without suspending withdrawals indefinitely.

#4 -- In retirement, one of your biggest risks is spending power, and of course interest rates tend to rise sharply with increases in inflation. Extended duration bonds are a particularly poor vehicle in this scenario, as their losses during a period of sharply rising rates would be catastrophic. And if rising rates lead to a recession, ala 1981, and stocks fall as well, you are in a real bad spot. I'm not sure TIPS (lets say 0% real returns) are going to provide enough protection here either, so there isn't much of a way to diversify away this risk.

Here is what I would say:

(a) don't ignore the longer-term evidence on bonds. Only about 5YRs in maturity gets you 95% of the TERM premium beyond 1mo bills. And historically, 5YRs has been as good as longer bonds in offsetting equity declines (better from 1928-1999, worse in the 2000s)

(b) stick with something like Vanguard Int'd Treasury while you are accumulating, then consider adding (or replacing 5YR) with ST Bond Index and TIPS. You may put up with a bit more short-term volatility, but you will have significant liquidity in terms of low risk and real returning bonds which will act as a bridge to get you through stock bear markets

(c) while working, or in retirement, equities are the growth engine of your portfolio. Yes, low rates have made things more difficult, but only for those with 60%, 70%, 80% or more of their portfolio in bonds. A more balanced mix, even for a retiree, spread across small and value (which are also a benefit in inflationary times) should carry the water for your portfolio even if low risk bonds disappoint.

(d) learn lessons from big institutional investors and endowments who thought abandoning traditional equities and fixed income in favor of all sorts of alternatives would allow them to still get high returns without high risk -- there are no combos of precious metals and extended duration bonds in combo with equities that will produce the same high returns as just stocks and bonds with less risk. Either you get less risk and much less returns (not good), or same returns or slightly better with much more risk. A look at the last 5/10 years of hundred-million and billion dollar endowments and institutions finds the got less returns and more risk, which is also possible when trying to over-engineer a portfolio.

Eric
Hi Eric,

1) LTT Treasuries hedge against unexpected deflationary risks, not all equity risks. I never meant to imply that they are the only hedge, which is why i was careful to always say they are great in the context of portfolio that included stocks as well as commodities (i prefer physical bullion over CCF though). As far as the historical evidence for intermediate term bonds offering higher risk adjusted returns. Yes that is true in isolation. Additionally, when solving for the highest sharpe ratio portfolio you will likely get a solution involving intermediate term bonds and stocks. However, the Sharpe ratio is only relevant if you can lever up the portfolio at the risk free rate. Retail investors can't do that. Therefore having some LTT treasury acts as a pseudo form of levered up intermediates, allowing investors to increase the equity allocation. There is a leverage aversion story for why LTT have lower sharpe ratios. I like that LTT allow me to invest more in other assets while still getting large exposure to interest rate fluctuations

2) I am not sure why this is an argument against LTT. You are essentially saying that hedge assets have low expected returns because they are so valuable. I agree. I am willing to accept lower yields on Treasuries because they are a great diversifier.

3) Interest rates don't just magically go up. They go up when the Fed indicates they will raise short term rates, it is entirely dictated by Fed policy (even the long end of the curve is largely a function of expected fed policy in the future). Mr. Bernanke has been very clear that rates will only rise when we see decreased unemployment (this would likely be accompanied by rising growth expectations in the market and higher stock prices), or if inflation gets out of control (this would be associated with rising inflation expectations that would benefit commodites/gold). So from a portfolio perspective you can hedge a lot of the risks associated with rising rates, because rates rise based off of macroeconomic reasons. Reasons that affect other markets outside of Treasuries.

4) Same story as before. You are looking at assets in isolation. I can only speak to the benefits of long duration bonds in a portfolio that includes commodities and stocks. Besides, nothing went up in 1981 except for money in the bank.

a) Leverage aversion. I get more bang for my buck with LTT, albeit at a cost higher than the risk free rate, but I can't borrow at the rate anyways.

b) I am really happy with my allocation.

c) Equities can be a growth engine, or wealth destroyer. It depends what generation you are in. I prefer diversification.

d) All of my number crunching indicates the exact opposite. A combination of LTT, commodities, and stocks provides way higher real risk adjusted returns than a stock/intermediate portfolio.
EDN
Posts: 528
Joined: Sat Dec 29, 2012 6:55 pm
Location: Oklahoma City, OK

Re: When to add 30 year bonds to one's portfolio?

Post by EDN »

^
There is a lot there, and not a lot I agree with. But you seem pretty invested in your views.

I will just say (repeat) if long treasuries continue to counterbalance equities as they have since 2000, you will almost certainly earn large negative (not low, but negative) real returns into perpetuity. That is a major issue for a portfolio to have a component do so poorly for the price of lowering risk a bit.

I think the last 10 years has convinced a lot of people they want exposure to interest rate "fluctuations", but really that just means falling rates. If LTT go from 3 to 7, you won't be nearly as happy with -50%+ losses. We've see that level or real return loss last 40 years, by the way. Even on a slice of a portfolio, that hurts a lot.

Interest rates go up for a lot of reasons, and it doesn't happen because the Fed sends out a memo that they are raising rates in a few months. I'd refer you to a paper Eugene Fama authored in 2010. It looks critically at the role the Fed plays in interest rates, very inconclusive. If you are thinking you can fashion a portfolio that is neat and tidy to offset all assocated risks from a bunch of really risky individual holdings, you are likely to get stung as characteristics change, historical relationships evolve, and you are Harvard in 2008 without the ability to issue your own bonds to collect money for short-term funding.

I don't look at anything in isolation. LTT with negative real returns and Gold just earning inflation or less if prices are likely inflated means a good chuck of your hedged portfolio will live underwater. And a small uptick in rebalancing bonus won't begin to act as a life preserver. The only hope, then, will be that you have enough small/value and equities, and expected returns ---> actual returns on the stock side.

There is a lot of love for LTT today, and again, it's 2000-today talking. You'll just as soon sell at the wrong time when you realize a -30% loss in bonds isn't as fun as it looks on paper. For an institution hedging a long dated liability, LTT are great. For investors with real short-term liabilities, they can be catastrophically bad.

Do what you will, my guess is your attempt to "time" LTT indicates you due realize there are risks that aren't accounted for in your simulations of perfect hedged asset class combos. But don't say you weren't warned that stocks (small/value tilted) and simply 5yr bonds do 80% of what you hope to get from stocks/LTT/PM/other with much less chance of things going horribly wrong.

And by the way, TSM (big multinational stocks) are time dependent (65-81 bad, 82-99 good, 00-12 bad), but TSM+LV+SV+Int'l Value/Small is far less period dependent, and much more reliable as a return generator over a variety of market conditions (65-81 good, 82-99 good, 00-12 good). Dilute with 5YR bonds per your situation and you are much better off with a far simpler mix and aren't dancing on the head of a pin to hedge every possible future tail event.

PS -- you know a great way to reduce left tail risk? Have enough in low risk assets (5YR) to sell as the tail swings back to the middle. This is a subtle but infrequently issued bit of advice. You didn't lose in 73-74 if you sold 5YR bonds while equities came back. You didn't lose in '87 if you sold 5YR bonds while equities came back. You didn't lose in '90 if you sold 5YR bonds while equities came back. You didn't lose in '98 if you sold 5YR bonds while equities came back. You didn't lose in 00-02 if you sold 5YR bonds while equities came back. You didn't lose in 08 if you sold 5YR bonds while equities came back. You didn't lose in the Summer of '10 or '11 if you sold 5YR bonds while equities came back.

Good luck, just trying to give you stuff to ponder.

Eric

PS -- as you seem to be moving towards something like the Permanent Portfolio (some stocks, some precious metals, some LTT, etc.), I thought I'd share a stat I have with others before:

1975-2012 Real Returns (first year it was legal to own gold)
PP = +5.0%
30/70 Asset Class Mix = +6.7%

PP had 4 down years with biggest loss = -5%, 30/70 AC had 2 down years with biggest loss = -3%.

There isn't anything you can do with these more exotic allocations that a good 'ole mix of 5YR bonds and stocks (small/value tilted) don't do better.
User avatar
rmelvey
Posts: 826
Joined: Sat Sep 18, 2010 5:17 pm
Contact:

Re: When to add 30 year bonds to one's portfolio?

Post by rmelvey »

Eric,

I appreciate the dialogue and I am always open to new ideas. I certainly don't mean to come across as a stick in the mud.

I find the value studies interesting, but I have hard time making capital allocation decisions based on it.

You have a combo of slices of the stock market that mixed well in the past, but what is the story behind it? I care tremendously about the narrative of causation that is driving the diversification. Interactions between stocks, bonds, and gold/commodities have huge powerful macroeconomic reasons that drive those relationships. Interactions between intersections of the stock market (that were not neatly bundled up and packaged to investors in the past like they are now) makes me extremely skeptical about extrapolating past diversification relationships between different cross sections of the stock market. Furthermore, when is the last time you heard of someone explicitly tilting towards growth stocks. Everyone and their grandma (literally) tilts towards value in a way that they didn't decades ago and rebalances between value and the market. That could drive up correlations as well as lowers returns on value stocks.

I think you misread something about "timing." I have no plans to time the LTT market. I simply have wide bands for rebalancing to lower transaction costs and capture any momentum affect that might exist in the markets, but the wide bands are largely due to the former. I also am a big believer in volatility harvesting. That is market timing independent, but is simply a function of low correlation and high volatility. It can be illustrated with random walk coin flips as an extreme example.

PS: Right on the money. I have been using the PP for about 3 years (instead of T-Bills though I have I-bonds). Also, I want to be a value believer. It just isn't something I can hang my hat on yet. I understand how well it has done in the past, but I haven't seen a convincing story that leads me to believe it something I can count on for the future :happy
linguini
Posts: 262
Joined: Tue Oct 16, 2012 2:49 pm

Re: When to add 30 year bonds to one's portfolio?

Post by linguini »

rmelvey wrote:
Jfet wrote:The EE bond is yielding almost a full point better than the 20 year treasury...and it is tax deferred and state and local tax exempt.

I think I would at least fill out my $10,000 in EE bonds each year before buying 30 year treasury.
If you are planning to hold until maturity that might be the way to go. However, the EE bond does not allow you to profit from the volatility and it is essentially entirely illiquid if you are treating it as a long term bond because that doubling only happens on the last year, cashing out early has a huge penalty..

In my mind an EE-bonds are best used as a cash substitute (with the thought of rolling it over if ST rates rise) but with an embedded call option on rates staying low. It is an interesting product (probably with some free lunch in there), but I don't think it is directly comparable to a 30 year Treasury held in a brokerage account.
True that EE bonds are essentially illiquid if you are planning on getting the interest through waiting out the doubling period and not directly comparable to a 30-year treasury held in a brokerage account. If you have any plans to redeem them within the next 20 years, then they are more similar to holding short duration treasury bonds, and I bonds would be more appropriate. But I'm not quite sure if profiting off the volatility of 30-year treasuries is really appropriate for the purpose the OP has in mind. The OP apparently wants to hold long term bonds as diversification and crash protection. For that, strategy of buy-and-hold for 20-year EE bonds works just fine. On the other hand, engaging in frequent buying and selling of 30 year treasuries based on which direction you expect the interest rate to move so that you can profit off volatility can be pretty risky and does not provide dependable, predictable returns the way an EE bond does.
User avatar
nisiprius
Advisory Board
Posts: 52216
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: When to add 30 year bonds to one's portfolio?

Post by nisiprius »

rmelvey wrote:
nisiprius wrote:I do not see the point of 30 year nominal bonds for an ordinary investor under any circumstances. The inflation risk is just too severe.
From a liability matching perspective, you are entirely correct. However, when periodically rolled over (to maintain that long duration) and held in with conjunction with equities and commodities (most notably gold), you have exposure to a wide array of macroeconomic environments. Exposure to a wide degree of macroeconomic environments ensures that your portfolio is likely to be able to survive black swan events. Additionally, the high volatility of LTT can be a blessing rather than a curse because with wide rebalancing bands you can capture intermediate term momentum, long term reversion to the mean, and are able to benefit from volatility harvesting.

Volatility harvesting is something that doesn't appear when looking at measures of single period returns (such as sharpe or sortino) but for an investor who cares about compounding results it makes a big difference. I wrote a post about this here:
http://www.stableinvesting.com/2013/04/ ... demon.html.... Volatility harvesting is one of the few free lunches that exist in an efficient market....
They are a specialty item that makes sense from the point of view of a certain specific investing theory and a certain specific way of using them. I will continue to say that it's not a Bogleheadish theory. Whenever people say that they like longer-term bonds because they want the volatility, or indeed when people say in general that they like volatility because it's how they make money, that may be OK for them but it tells me that it is a different way of using bonds than I use them.

I use bonds to damp down volatility, not play off against it. And, yes, I am trying to "defease" my holdings and do rough liability matching by holding individual TIPS which I plan to hold to maturity.

To me, bonds are there to stabilize a portfolio--about the same way cash would--while, at least in Ye Olde Days, earning a worthwhile extra return over cash. That extra return is in exchange for some volatility. For intermediate-term bonds, that volatility is too high to serve as the household checking account, but is practically hidden in any portfolio that includes any meaningful amount of stocks. Ideally, one would use individual bonds and do liability matching; as a practical matter, bond funds are close enough if you follow Vanguard's suggested holding times (4-10 years for Total Bond, for example).

There certainly are investing theories under which you can use bonds in other ways. Anyone who is planning to hold long-term nominal bonds, either individually or in a bond fund, had better be sure that they understand and are in accord with those theories. To me, the fact that Vanguard does not use long-term bonds (other than those within Total Bond) in their all-in-one products is evidence--weak evidence, but evidence--that these theories are not cut-and-dried or universally accepted. If I thought it all worked as nicely as you say, I would be very agreeable to paying someone a SMALL expense ratio to have them hold the bonds and do the "volatility harvesting" for me.

The important thing is that anyone who's going to hold long-term bonds had better be a sophisticate who understand the theory and knows why they are doing it, rather than just blindly looking at the table of returns and taking on more risk because the return of shorter-term bonds is just ridiculous these days.
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
abuss368
Posts: 27850
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!
Contact:

Re: When to add 30 year bonds to one's portfolio?

Post by abuss368 »

Invest in intermediate term bonds instead. No need to take unnecessary risks with bonds.
John C. Bogle: “Simplicity is the master key to financial success."
Johm221122
Posts: 6395
Joined: Fri May 13, 2011 6:27 pm

Re: When to add 30 year bonds to one's portfolio?

Post by Johm221122 »

linguini wrote:
Johm221122 wrote:Inflation and fear of inflation is why I would not consider 5%, I woukd think about it at 8% and really think about at 10%.
John
You would consider turning down a 30-year treasury bond at 10% interest in the current environment? At those rates, unless inflation rockets up within the next ten years, real return is so high that you make back your initial investment in the first ten years. If you are too risk-averse to tie money in 10% nominal yield guaranteed, why would you be willing to invest in the stock market which is far more volatile than inflation and at no higher return? Surely you're exaggerating that you wouldn't take that deal for at least part of your portfolio?
OP is asking when to jump into 30 year treasury bond, 10% is not going to happen in this environment. I don't recommend it for his goal unless it hits say 8-10%.He wants to protection against lower interest rates in his retirement, but the fear of buying long-term treasuries is inflation. I would definitely consider long-term bonds at 8-10% it's others on this forum that are not if you read carefully the replies
John
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: When to add 30 year bonds to one's portfolio?

Post by nedsaid »

For me, never.

There are people that do a barbell strategy with long bonds and short bonds. Or you could have the 30 year bonds as a deflation hedge.

There was a once in a lifetime opportunity in the early 1980's, just after Mr. Volker at the Federal Reserve jacked up rates to kill inflation. That was the time to back up the truck.

I suppose the historical interest rate for 30 year bonds is probably about five percent. We got used to higher rates in the 70's and 80's. If rates got to more "normal" rates, these would be okay if you could take the volatility. To me, this defeats the purpose of bonds which is to decrease volatility in a portfolio while still providing a real return. I am a fan of the medium term bonds, 5-7 years or so in maturity.

So there is a place for these in a portfolio if it fits your investment philosophy. Always a good deflation hedge.
A fool and his money are good for business.
User avatar
rmelvey
Posts: 826
Joined: Sat Sep 18, 2010 5:17 pm
Contact:

Re: When to add 30 year bonds to one's portfolio?

Post by rmelvey »

linguini wrote:
rmelvey wrote:
Jfet wrote:On the other hand, engaging in frequent buying and selling of 30 year treasuries based on which direction you expect the interest rate to move so that you can profit off volatility can be pretty risky and does not provide dependable, predictable returns the way an EE bond does.
I am not a believer in market timing (especially with a market as intelligent as the bond market). Nothing about my posts were meant to be interpreted as favoring a timing based approach. I was trying to say that if you want exposure to the 30 year bonds, obviously you have to sell it and buy a fresh one every once and while. Within 5 years your 30 year bonds is a 25 year bond... This is not about timing but sticking to your investment plan (if your plan calls for exposure to a long bond).

Additionally, I think the term volatility harvesting is throwing off some of the readers of this thread. Volatility harvesting has nothing to do with market timing nor reversion to mean. Volatility harvesting can exist even with a purely random coin flip game. It can totally exist within the confines of a perfectly efficient market. I provided a link to a page that explains the concept in an earlier post.
WHEEE
Posts: 6
Joined: Fri Dec 21, 2012 4:56 pm

Re: When to add 30 year bonds to one's portfolio?

Post by WHEEE »

The only 30 year bonds I would ever buy is 30 year TIPS and hold to maturity... I feel 30 years is much too long of a time to make a coin-flip gamble about inflation.
Post Reply