Higher bond allocations with riskier bonds

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
ibhhvc
Posts: 94
Joined: Sat Jul 11, 2015 8:44 am

Higher bond allocations with riskier bonds

Post by ibhhvc » Mon Apr 16, 2018 11:32 am

It seems like you can reasonably compare the risk and returns of a 70/30 portfolio that invests in intermediate-term bonds with a 60/40 portfolio that invests in riskier bonds.

Here is a backtest of two portfolios starting in 1974. The first is 70% US total stock market and 30% intermediate-term treasury bonds. The second is 60% US total stock market, 20% intermediate-term treasury bonds, and 20% long-term corporate bonds:

Image

Essentially, you would have achieved similar performance from a higher bond allocation but with riskier bonds.

Does anyone choose this type of allocation? I know both portfolios ended up performing similarly so there might not be a compelling reason to go this route. But at the same time, is there a good reason not to?

User avatar
oldcomputerguy
Posts: 3556
Joined: Sun Nov 22, 2015 6:50 am
Location: In the middle of five acres of woods

Re: Higher bond allocations with riskier bonds

Post by oldcomputerguy » Mon Apr 16, 2018 11:35 am

Not sure I understand the purpose. If risk and return are similar, why go with three funds rather than two? You seem to be moving some of your equity risk to bonds with similar risk and return. Why bother?
It’s taken me a lot of years, but I’ve come around to this: If you’re dumb, surround yourself with smart people. And if you’re smart, surround yourself with smart people who disagree with you.

Valuethinker
Posts: 36722
Joined: Fri May 11, 2007 11:07 am

Re: Higher bond allocations with riskier bonds

Post by Valuethinker » Mon Apr 16, 2018 11:50 am

ibhhvc wrote:
Mon Apr 16, 2018 11:32 am
It seems like you can reasonably compare the risk and returns of a 70/30 portfolio that invests in intermediate-term bonds with a 60/40 portfolio that invests in riskier bonds.

Here is a backtest of two portfolios starting in 1974. The first is 70% US total stock market and 30% intermediate-term treasury bonds. The second is 60% US total stock market, 20% intermediate-term treasury bonds, and 20% long-term corporate bonds:

Image

Essentially, you would have achieved similar performance from a higher bond allocation but with riskier bonds.

Does anyone choose this type of allocation? I know both portfolios ended up performing similarly so there might not be a compelling reason to go this route. But at the same time, is there a good reason not to?
Historically you have not been rewarded for taking credit risk with bonds. You get higher volatility but not higher returns.

Swensen takes you through the asymmetric information problems with corporate bonds. Management seeks to maximize wealth for shareholders so choose to issue bonds when they are, on average, overpriced (creditworthiness of the company is worse than investors understand).

The exception is the "chink" between High Yield and Investment Grade bonds. Because many investors are legally restricted from holding the latter, the drop in price (rise in yield) is too large crossing below BBB credit rating relative to the increase in default risk. Don't know of a passive fund that plays that aspect.

Because many corporate bonds are callable, when interest rates fall you tend to get early redemptions by those bonds, thus a lower yield (and a smaller price rise) for the fund than with straight Treasury bonds. Unfortunately the reverse is also true-- callable bonds are priced on Yield-to-Call generally, so if interest rates rise the issuer will not call them early-- you overpaid for the bonds (in retrospect).

What you are in essence doing is increasing the correlation between risk in your bond portfolio and your equity portfolio.

Your numbers show this is merely a matter of degree. But consider that a substantial fraction of corporate bonds are issued by financial institutions. If we have another 2008-09 like scenario, those bonds will get hammered, most likely. You've increased your portfolio risk in the worst possible set of events.

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Re: Higher bond allocations with riskier bonds

Post by dbr » Mon Apr 16, 2018 11:51 am

ibhhvc wrote:
Mon Apr 16, 2018 11:32 am


Does anyone choose this type of allocation? I know both portfolios ended up performing similarly so there might not be a compelling reason to go this route. But at the same time, is there a good reason not to?

Sure people do things like this all the time. There was a recent thread with some efficient frontier charts of how duration in one's bond choice plays with allocation to stocks. The "Larry" portfolio is an extreme where a very risky stock choice in small cap value is played against a high bond allocation in very conservative bonds. You can even take the idea out of the box and consider abandoning bonds for SPIAs and see how that plays out.

The bottom line is that one concerns themselves with the portfolio and not with assets in isolation. That is supposed to be a fundamental starting point.

ibhhvc
Posts: 94
Joined: Sat Jul 11, 2015 8:44 am

Re: Higher bond allocations with riskier bonds

Post by ibhhvc » Mon Apr 16, 2018 11:56 am

oldcomputerguy wrote:
Mon Apr 16, 2018 11:35 am
Why bother?
If simplicity is the goal, why not have a portfolio that's 50% stocks and 50% long-term treasury bonds? Nice and simple, and has performed well over the past 30 years (vs. the same 70/30 portfolio with intermediate-term bonds as shown in the first post — the 50/50 portfolio is in red below):

Image

It seems like the conventional wisdom is that stocks are for high risk/high reward and bonds are the completely opposite. I'm just wondering why a more moderate approach isn't also valid: put less risk in stocks (by holding less in your portfolio), and put slightly more risk in bonds.

lack_ey
Posts: 6636
Joined: Wed Nov 19, 2014 11:55 pm

Re: Higher bond allocations with riskier bonds

Post by lack_ey » Mon Apr 16, 2018 12:20 pm

Valuethinker wrote:
Mon Apr 16, 2018 11:50 am
Historically you have not been rewarded for taking credit risk with bonds. You get higher volatility but not higher returns.
That's approximately true when looking at long-term bond returns and not duration matching. If you do a more apples-to-apples comparison, it doesn't seem to be right.

Valuethinker
Posts: 36722
Joined: Fri May 11, 2007 11:07 am

Re: Higher bond allocations with riskier bonds

Post by Valuethinker » Mon Apr 16, 2018 1:02 pm

lack_ey wrote:
Mon Apr 16, 2018 12:20 pm
Valuethinker wrote:
Mon Apr 16, 2018 11:50 am
Historically you have not been rewarded for taking credit risk with bonds. You get higher volatility but not higher returns.
That's approximately true when looking at long-term bond returns and not duration matching. If you do a more apples-to-apples comparison, it doesn't seem to be right.
I would be interested in the references if you had any?

User avatar
BolderBoy
Posts: 4184
Joined: Wed Apr 07, 2010 12:16 pm
Location: Colorado

Re: Higher bond allocations with riskier bonds

Post by BolderBoy » Mon Apr 16, 2018 1:07 pm

ibhhvc wrote:
Mon Apr 16, 2018 11:32 am
But at the same time, is there a good reason not to?
Yes. KISS. Take your risks on the stock side. The more variables one weaves into one's investing strategies, the tougher it is to understand why it doesn't work out, when it doesn't.
"Never underestimate one's capacity to overestimate one's abilities" - The Dunning-Kruger Effect

cherijoh
Posts: 4994
Joined: Tue Feb 20, 2007 4:49 pm
Location: Charlotte NC

Re: Higher bond allocations with riskier bonds

Post by cherijoh » Mon Apr 16, 2018 1:16 pm

ibhhvc wrote:
Mon Apr 16, 2018 11:56 am
oldcomputerguy wrote:
Mon Apr 16, 2018 11:35 am
Why bother?
If simplicity is the goal, why not have a portfolio that's 50% stocks and 50% long-term treasury bonds? Nice and simple, and has performed well over the past 30 years (vs. the same 70/30 portfolio with intermediate-term bonds as shown in the first post — the 50/50 portfolio is in red below):

Image

It seems like the conventional wisdom is that stocks are for high risk/high reward and bonds are the completely opposite. I'm just wondering why a more moderate approach isn't also valid: put less risk in stocks (by holding less in your portfolio), and put slightly more risk in bonds.
I think you are overlooking the fact that capital gains in stocks are taxed at a preferential rate whereas bond income is taxed at ordinary tax rates.

lack_ey
Posts: 6636
Joined: Wed Nov 19, 2014 11:55 pm

Re: Higher bond allocations with riskier bonds

Post by lack_ey » Mon Apr 16, 2018 2:12 pm

Valuethinker wrote:
Mon Apr 16, 2018 1:02 pm
lack_ey wrote:
Mon Apr 16, 2018 12:20 pm
Valuethinker wrote:
Mon Apr 16, 2018 11:50 am
Historically you have not been rewarded for taking credit risk with bonds. You get higher volatility but not higher returns.
That's approximately true when looking at long-term bond returns and not duration matching. If you do a more apples-to-apples comparison, it doesn't seem to be right.
I would be interested in the references if you had any?
The main and most up-to-date reference would be "The Credit Risk Premium" by Asvanunt and Richardson in The Journal of Fixed Income (2017). An earlier version is on SSRN:
https://papers.ssrn.com/sol3/papers.cfm ... id=2563482

The authors are associated with AQR, so to the extent you think they have horses in this race, there's that to consider. But it also means the full text is available for the latest version on their website. Click the "view" link on this page to get to the download:
https://www.aqr.com/Insights/Research/J ... sk-Premium

Some of their derived data they make available:
https://www.aqr.com/Insights/Datasets/C ... Paper-Data

I took a look at some of that in this prior thread:
viewtopic.php?t=209748

One of the comparisons made there (comparing bond allocations using 1-3 yr Treasuries to 1-3 yr credit bonds, where the durations are actually similar, over the past 10+ years):
https://www.portfoliovisualizer.com/bac ... tion3_3=24
Most past research computes credit excess returns as the simple difference between long-term corporate bond returns and long-term government bond returns... Fama and French [1993] suffer from this shortfall when they note that the average default return is only 2 basis points per month
A related and older paper is "Ibbotson’s Default Premium: Risky Data" by Hallerbach and Houweling (2013), but I don't know if the text is accessible anywhere (for free). Abstract is on SSRN:
https://papers.ssrn.com/sol3/papers.cfm ... id=1898178

DesertDiva
Posts: 189
Joined: Thu Mar 01, 2018 12:49 pm

Re: Higher bond allocations with riskier bonds

Post by DesertDiva » Mon Apr 16, 2018 2:21 pm

This strategy seems very un-Bogleheadish to me. I would re-read concepts related to Indexing and Asset Allocation - then revisit the reason for including bonds in your portfolio before considering this strategy.
♫ Stocks go up ♫ Stocks go down ♫ Stocks go up ♫ Stocks go down ♫

Post Reply