PF allocation: short-term vs long-term bonds

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
seeker7
Posts: 7
Joined: Sat Oct 05, 2019 4:40 am

PF allocation: short-term vs long-term bonds

Post by seeker7 » Sun Nov 17, 2019 2:43 am

Hello everyone,

When investing in only two asset classes (world equities and government bonds) for a fairly long term horizon (say 30 years), it is better for the bond part of the portfolio to be invested in short-term government bonds or long-term governments bonds? Why?

Thanks a lot for any input in the discussion!

User avatar
WoodSpinner
Posts: 921
Joined: Mon Feb 27, 2017 1:15 pm

Re: PF allocation: short-term vs long-term bonds

Post by WoodSpinner » Sun Nov 17, 2019 10:42 am

OP,

The clear answer is IT DEPENDS ....

Suggest you repost with your financial details as outlined in

viewtopic.php?t=6212

WoodSpinner

User avatar
nisiprius
Advisory Board
Posts: 39485
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: PF allocation: short-term vs long-term bonds

Post by nisiprius » Sun Nov 17, 2019 11:20 am

For bonds with high credit quality, I think there's consensus in this forum that, when viewed as an asset in isolation, as you go from short-term to intermediate-term to long-term, you see both increasing risk and increasing reward. The risk takes the form of interest rate risk, and--if they are not inflation-protected bonds--inflation risk. The amount of increased reward depends on the yield curve, large for a steep yield curve, small for a shallower one.

As to how to apply this in a portfolio, in this forum, posters seem to fall broadly into two schools of thought.

I personally am among the traditionalists, as articulated in the Bogleheads investment philosophy, in Taylor Larimore's book about the three-fund portfolio, and in the actual portfolios Vanguard uses in its all-in-one funds (Target Retirement and LifeStrategy).

According to that school of thought, "bonds are for safety," and you should "take your risk on the equity side." In this view, bonds simply act as ballast for a portfolio, and the bond allocation should not be too volatile or risky. The only reason to use intermediate-term bonds rather than short-term bonds or cash equivalents is that you expect to get higher return, and in a portfolio with stocks, the stocks will be so volatile that you will hardly notice the volatility of bonds--up to about intermediate term.

In that school of thought, bonds damp down volatility, simply through dead weight.

The second school of thought involves a theory about correlations between stock price movements and bond price movements. One idea is that they move independently. A stronger idea is that when stocks fall, there is a "flight to safety" and investors pile into Treasury securities and bid up their prices, and thus they actually move in opposition to each other. Either way, there is a mathematical benefit to a mixture of stocks and bonds. Because of the low correlation, the volatility of the portfolio will be less than the weighted average of the volatilities of the the stocks and bonds.

In this case, bonds fluctuations are beneficial volatility. Over past periods of time, such a benefit can be seen with intermediate-term bonds, but only feebly. As I said, you can hardly see the volatility of intermediate-term bonds in a portfolio with a reasonable stock allocation, and that applies even to beneficial volatility.

People belonging to the second school feel that if bond volatility is helping the portfolio, then the more volatility the better, up to about the point where the stock and bond components have equal volatility.

Therefore, people who are using bonds, not just to cushion, but to actively counteract stock volatility, arrive at portfolios that use long-term bonds--perhaps the longest-term available--or go even further by using leverage on the bonds.

The idea of trying to equalize the volatility among all of the asset classes in a portfolio was popularized by a hedge fund manager, Ray Dalio, under the name "risk parity." However, "risk parity" strategies, and "risk parity" mutual funds usually include asset classes besides stocks and bonds, and usually involve macroeconomic ideas about which asset classes to include as well as the idea of equalizing their volatility.

Now, consider strategies that combine high-volatility assets, that are expected to move in compensating directions and cancel out each others' risks. A backtest over any period in which they behaved "properly" will show them outperforming, in risk-adjusted reward. They may also outperform in return itself, but leverage may be needed to achieve that. These strategies are inherently risky in the sense that they depend on the assets continuing to behave "properly" in the future. This is especially true if the strategies are counting on actual negative correlation, not just zero correlation or low correlation. (It should go without saying that people advocating these strategies believe that they will. Just as it should go without saying that anyone who uses leverage believes that leverage is not really all that risky "as long as you know what you're doing.")
Last edited by nisiprius on Sun Nov 17, 2019 11:39 am, edited 1 time in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

AHTFY
Posts: 113
Joined: Mon Jul 23, 2018 2:33 pm

Re: PF allocation: short-term vs long-term bonds

Post by AHTFY » Sun Nov 17, 2019 11:30 am

nisiprius wrote:
Sun Nov 17, 2019 11:20 am
+1 Great post!

dbr
Posts: 30860
Joined: Sun Mar 04, 2007 9:50 am

Re: PF allocation: short-term vs long-term bonds

Post by dbr » Sun Nov 17, 2019 1:44 pm

Nisi's explanation is excellent. Note the important concept that both choices of bond strategy relate to the risk and return of the whole portfolio. Most discussions of bonds on the forum seem to focus on the bond investment in isolation rather than as a part of a whole and therefore may be a bit short sighted. In particular it is odd for someone holding half or more of their assets in stocks to worry about whether or not a bond investment can lose money (It can, especially in real dollars.). The exception is people who really are holding fixed income for a definite period of time with a need to be sure they can get that money back at that specific time.

An example of the second idea in the above discussion is that there is some credibility to the idea that high stock allocations might best be paired with risky long bonds because the risk is not much different from what it already is but there is a "diversification" benefit.

Another variation on bonds in a portfolio is the idea of the "Larry" portfolio (after Larry Swedroe). In that case one increases the risk and expected* return of the stocks and then cuts the risk by increasing the allocation to (short) bonds. The theory is that such a portfolio will have similar overall risk and return as before but less extreme downside risk. This is not a recommendation but just a discussion.

*Expected does not mean what you expect to get. I refers to the average of the imagined statistical distribution of possible returns from which year's return is imagined to be an random sample.

Topic Author
seeker7
Posts: 7
Joined: Sat Oct 05, 2019 4:40 am

Re: PF allocation: short-term vs long-term bonds

Post by seeker7 » Mon Nov 18, 2019 1:08 pm

Thanks a lot @nisiprius, for your detailed answer. This was exactly what I was looking for.

My thinking was indeed on the 2nd school of thought in your expose. Would you know of any study backtesting such an idea? Or alternatively, some proponents of the first school of thoughts showing evidence that the 2nd school of thoughts doesn't hold in practice?

Thanks a lot for your help

Post Reply