High Yield in Portfolio

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
Post Reply
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

High Yield in Portfolio

Post by larryswedroe » Wed Mar 28, 2007 2:59 pm

I had the following analysis done a while ago and plan on updating it shortly but thought I would share it.
The idea was to create three portfolios, One with 10% allocation to high yield via Vanguard's fund.
The other was to increase equity allocation a bit until I got the same return from portfolio, but eliminated junk bond risks---and see if got lower or higher SD (better tax efficiency not even considered)
The third was to tilt to sv instead, RAISE fixed income allocation and lower equity, getting same return and see if SD could be lowered
Here were the results my assistant came up with.

PORTFOLIO A (P-A): 60% CRSP 1-10, 30% Lehman Inter. Govt Bond Index, 10% VWEHX.
PORTFOLIO B (P-B): 61.5% CRSP 1-10, 38.5% Lehman Inter. Govt Bond Index.
PORTFOLIO C (P-C): 42.5% S&P 500 Index, 7.5% FF Small Value Index, 50% Lehman Inter. Govt Bond Index.


The annualized return for P-A from 1979-2005 was 11.74%. The annual
standard dev was 10.65. P-B also had an annualized return 11.74%.
However, its standard dev was 10.46. Some would say the difference in
std dev is small, but that's the whole point. We've completely
eliminated exposure to hi-yield, and we've BARELY added any more equity
- and we get virtually the same risk/return profile. I think the
comparison between P-A and P-B is your strongest argument against the
credit risk being rewarded - moreso than comparing P-A and P-C, which
is interesting but argues a different point.


The annualized return for P-C is 11.74% as well. However, the annual
standard dev is 8.66 - significantly less than 10.65 for P-A. Note that
this portfolio has the most fixed income - 50%. I think this supports
our stanard "take more risk w/ equities and have a higher allocation to
conservative fixed income" approach.

Draw your own conclusions.

User avatar
cfs
Posts: 4154
Joined: Fri Feb 23, 2007 1:22 am
Location: ~ Mi Propio Camino ~

Thanks

Post by cfs » Wed Mar 28, 2007 3:09 pm

Thanks Larry for the note. The more I read about fixed income the more I like my TSP G Fund !!! Oh by the way cfs is "guilty as charged." I have "only" two bond funds in my portfolio, the TSP G Fund and Vanguard Short Term Inv Grade - this is possibly "not good enuff" for some, but both have passed my pillow test !!! Thanks again.

cfs
~ Member of the Active Retired Force since 2014 ~

User avatar
steve
Posts: 135
Joined: Sat Mar 10, 2007 7:38 am

Newbie Question

Post by steve » Wed Mar 28, 2007 3:16 pm

OK, I'll take the first hit, so be kind. Our portfolio is 50 / 50, with our deferred accounts holding VFIDX and VBIIX, but that only accounts for half the bonds, the rest being munies (VMLUX, VWIUX, VWALX) in our taxable account. One would like to think that the higher credit ratings of the minicipal bond funds could depress volatility, even in the hi yield, which at Vanguard has an average credit of AA. What do you think?
"No matter where you go, there you are."

User avatar
greenspam
Posts: 578
Joined: Mon Feb 26, 2007 12:36 pm
Location: east coast

past vs. future...

Post by greenspam » Wed Mar 28, 2007 3:26 pm

"We've completely eliminated exposure to hi-yield, and we've BARELY added any more equity - and we GET virtually the same risk/return profile."

you should have said "GOT" (in this analysis of the past), not "GET", which implies that it is still true today and into the future....

anyone can crunch #s from the past, but i need to know how these 3 model portfolios will behave the next 25 years....
as always, | peace, | greenie.

User avatar
Random Musings
Posts: 5595
Joined: Thu Feb 22, 2007 4:24 pm
Location: Pennsylvania

Post by Random Musings » Wed Mar 28, 2007 3:36 pm

Larry,

My only question is - was the 0.19% difference in historical volatility over that 25 year period found to be statistically significant or not?

Regards,

RM

User avatar
hollowcave2
Posts: 1790
Joined: Thu Mar 01, 2007 3:22 pm
Location: Sacramento, CA

fascinating subject Larry

Post by hollowcave2 » Wed Mar 28, 2007 3:43 pm

Larry, I really want to thank you for all of your research and good books.

Regarding P-A vs. P-B, I think you are giving up something by eliminating HY and adding stocks, and that's income. So it would depend if that income was worth the risk.

Of course, you could just add hi quality bonds for the income, but that removes some of the growth potential.

Also, we're looking at Vanguard's fund, which is conservatively managed and takes less risk than other HY funds. That's a very important issue for me. I don't trust other HY funds, but I can take HY risk in the Vanguard fund.

But you have an excellent argument Larry and makes me think about all options. Excellent post.

Steve

Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe » Wed Mar 28, 2007 4:40 pm

Few thoughts

A) the point is to demonstrate as I have stated that there is virtually no unique risk in junk--the reason often sited to really invest in them. You can replicate the risk via equities and Treasuries or investment grade bonds as junk is nothing more than a hybrid---and an inefficient one at that. If a junk fund was lower credit rating than Vanguard you would have had to add more equity---the lower the credit rating the more equity like the risks

B) the comments were those of my assistant and not mine

c) the academic evidence from studies supports the data. And note that the data should be as favorable to junk as is about ever possible because we end the period with the lowest junk credit spreads probably ever, or at least close to it. That would be like looking at stock returns after period ending with stocks at record P/Es and expecting to be able to project the returns. The facts are that there is almost no unique risk in junk and thus you can project the data forward with confidence. And likely the future will look worse if spreads revert to historical mean.

d) I don't know if the lower SD is mathematically significant or not--but the point is still the same--I cannot think of reason to own junk--you get the same risk and expected return basically by tilting more to equities or size and value and you get return in more tax efficient way (at least for taxable investors who have choice of location).

e)Steve--yes munis, HIGH CREDIT quality, reduce volatility. But issue is still why own the high yield fund? I really cannot think of a reason to do so when there are better alternatives--especially if you have choice of locations.

BrianTH
Posts: 1340
Joined: Tue Feb 20, 2007 6:10 pm

Post by BrianTH » Wed Mar 28, 2007 4:54 pm

Well, I'm convinced. Of course, I am also the choir to Larry's preacher--I was already persuaded that there is no point in me straying outside of treasuries and perhaps high-rated munis (although I am happy to make exceptions for good stable value funds).

By the way, I don't understand why anyone would prefer more income and less capital gains, all else (like volatility) being equal. At best it is a wash, and at worst you get hit with higher taxes.

Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe » Wed Mar 28, 2007 5:13 pm

Brian

I am also a fan of stable value funds--if have very high credit rating of the provider and they stay reasonably short. TIAA has had a good product and had my mom's assets invested there. I think PIMCO runs a pretty good fund and I am sure there are others. But be careful out there.

BTW-enjoy your posts

BrianTH
Posts: 1340
Joined: Tue Feb 20, 2007 6:10 pm

Post by BrianTH » Wed Mar 28, 2007 5:27 pm

Larry.

As you may recall, you were very helpful in answering my questions about stable value funds at the old forum, including providing a preview of some of the discussion in your upcoming book. I took the checklist you provided there to heart, and fortunately the stable value fund we are using (in my wife's 401K) satisfied your requirements (it is even a bit more conservative, which is fine with me). So, thanks again.

Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe » Wed Mar 28, 2007 5:33 pm

My pleasure

SmallHi
Posts: 1718
Joined: Wed Feb 21, 2007 6:11 pm

Post by SmallHi » Wed Mar 28, 2007 6:07 pm

Larry --

good examples, I still think your bond durations are too long, however!

1979-2006

Asset Class__________P1________P2________P3_________P4
US TSM_____________60%______63%______55%_______40%
US Small Value__________________________10%_______10%
US High Yield________10%_______________________________
1yr T-notes_______30%______37%_____35%_____50%

RETURN____________11.3%_____11.3%____12.1%______11.2%
RISK______________10.2_______10.2______10.2________8.1

User avatar
drjdpowell
Posts: 882
Joined: Thu Mar 01, 2007 8:56 pm

High yield bonds

Post by drjdpowell » Wed Mar 28, 2007 7:13 pm

I don't understand why anyone would prefer more income and less capital gains, all else (like volatility) being equal. At best it is a wash, and at worst you get hit with higher taxes.


At best, one would get a higher return, due to the fact that many others get hit with higher taxes. Tax effects are reflected in prices. Just as taxable bonds must pay higher yields than Muni bonds, "tax-inefficient" assets should have a "tax premium", making them attrative to investors with tax-advantaged accounts like 401k's.

-- James

User avatar
tdhg566
Posts: 860
Joined: Thu Mar 08, 2007 4:37 pm
Location: Spring, TX

Post by tdhg566 » Wed Mar 28, 2007 7:15 pm

THANK YOU Larry. I feel compelled to add my gratitude for the time you take straightening out us knuckle oops Bogleheads :lol: I'm reading your latest book (...winning bond strategy...), but it's going to take some study for much of it to soak in. I'm sure I'll have some questions.

With your help (another thread) I DID dump the hi-yield in my 401, replacing it with safer bonds and better equities in taxable. And with the help of your book I'm looking MUCH more critically at the composition of the IT bond choice in our 401 (WACPX). It has 28% mortgage pass-thru, and 7% CMO, neither of which I understood at all before reading your book. It certainly does NOT pass the DFA test of having a yield that's 20bp per additional year of maturity better than ST funds. Problem is that the short term fund in our 401 (TMPXX) appears to be even worse. I'll work out something, but I wouldn't have known to look more closely without your help.

Thanks again

Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe » Wed Mar 28, 2007 7:34 pm

Small hi

First, those are just examples.

Second, I have no problem as you know staying shorter, and in fact that is what I recommend. Personally I own the DFA two year global and TIPS (just sold some today when the 14 TIPS went to 2.1. And if the real rate falls further will sell some more. Using my shifting allocation approach.

In the muni area I do go bit longer, but that is because I have the CCF hedge. Wonder what the critics of CCF will say if Iran really "blows up" and oil goes to 80,90 or 100? Wonder if they think the insurance was too expensive?

So, if you read my books you know I recommend 1-2 basically for accumulators and intermediate type ladders for retirees. Helps reduce reinvest risk. So no argument from me. Just remember that if do stay short you might face a "japan" problem with rates collapsing.

BrianTH
Posts: 1340
Joined: Tue Feb 20, 2007 6:10 pm

Post by BrianTH » Wed Mar 28, 2007 8:25 pm

James,

OK, but how many investors in bonds are actually subject to taxes on the income?

Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe » Wed Mar 28, 2007 9:34 pm

Brian, almost every high net worth person I know.

Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe » Wed Mar 28, 2007 9:35 pm

Brian, but the other side is also true--if you hold the junk in tax advantaged account you also lose the K gains treatment, you lose the tax option (valuable in volatile asset class), you lose the ability to donate appreciated shares and you lose the potential for stepped up basis. All negatives. And if high yield you would lose the FTC, like with EM bonds.

BrianTH
Posts: 1340
Joined: Tue Feb 20, 2007 6:10 pm

Post by BrianTH » Thu Mar 29, 2007 5:06 am

Larry,

Indeed, individuals are subject to income tax (although as you point out, that cuts both ways). But I wonder how much of the corporate bond market is individuals. See here:

http://www.financialservicesfacts.org/f ... vestments/

User avatar
nick22
Posts: 859
Joined: Sun Mar 04, 2007 11:00 am
Location: Ohio

Interesting Analysis

Post by nick22 » Thu Mar 29, 2007 5:22 am

Larry,
I am just finishing your bond book and your arguements are always persuasive. I currently just have my fixed income in a TIPS fund and a couple of total bond funds (TIAA-CREF Inst bond and Dodge and Cox). I guess I am too busy doing other things, but at this point considering creating my own personal bond portfolio seems painful and time consuming. At what point do you recommend someone create their own personal ladder and avoid paying mutual fund expense ratios? Is it relatively easy to do directly from the U.S. treasury? I know in the book you mentioned $500,000 was probably an entry point to this strategy. Do you recommend individual investors sell before maturity or just buy and hold? I feel like I would need my own bond advisor to execute some of these bond strategies, but maybe it is simpler than I think.
Nick22

BrianTH
Posts: 1340
Joined: Tue Feb 20, 2007 6:10 pm

Post by BrianTH » Thu Mar 29, 2007 6:07 am

Nick,

I'm sure Larry will have a more sophisticated answer, but have you looked at using treasury funds? I believe that Vanguard, for example, has treasury funds in every one of its term categories.

User avatar
steve
Posts: 135
Joined: Sat Mar 10, 2007 7:38 am

Post by steve » Thu Mar 29, 2007 6:21 am

e)Steve--yes munis, HIGH CREDIT quality, reduce volatility. But issue is still why own the high yield fund? I really cannot think of a reason to do so when there are better alternatives--especially if you have choice of locations.
Sorry to be so long in replying, Larry; can you suggest the better alternatives for the taxable account? I am unwilling to buy individual bonds, and I need some more time on the training wheels before I get into commodities.
"No matter where you go, there you are."

Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe » Thu Mar 29, 2007 7:46 am

Brian: Re the individuals holding of corporates, looks like a lot to me. Quick scan shows half-of 20 5 is households and 5 are mutual funds (meaning individuals mostly I assume) or 10 of 20.

Steve--for individuals in taxable, what you own depends on tax rate and yield curve. If value the convenience of mutual fund that buy the lowest cost and highest credit type fund that fits your tax bracket. And stay relatively short term to max intermediate. And re commodities--no one should ever invest in anything they don't fully understand the nature of the risks. Thus avoid CCF until you learn more. Appendix H of my Only Guide to a Winning Investment Strategy explains them in detail if you are interested. Also there is a good readable paper by Gorton and Rouwenhorst.

NIck: As Brian said, a good strategy is to simply own Treasuries, allocating some to TIPS and some to short to intermediate nominal bonds. And if you don't mind the incovenience, just buy them direct from Treasury. Funds offer convenience and ETFs and Vanguard funds are pretty cheap, and that makes for easier rebalancing too. And can combine the two. Using the mutual funds on the margin to rebalance and then buy and hold the individual bonds.

User avatar
jeffyscott
Posts: 8495
Joined: Tue Feb 27, 2007 9:12 am
Location: Wisconsin

How about portfolio D?

Post by jeffyscott » Thu Mar 29, 2007 10:04 am

PORTFOLIO A (P-A): 60% CRSP 1-10, 30% Lehman Inter. Govt Bond Index, 10% VWEHX.
PORTFOLIO B (P-B): 61.5% CRSP 1-10, 38.5% Lehman Inter. Govt Bond Index.
PORTFOLIO C (P-C): 42.5% S&P 500 Index, 7.5% FF Small Value Index, 50% Lehman Inter. Govt Bond Index.
Why not a portfolio including both SV and High Yield? Portfolio D should be 7.5% SV and 10% High Yield plus whatever percentage S&P 500 and bond index are needed to match the return of C.
Time is your friend; impulse is your enemy. - John C. Bogle

Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe » Thu Mar 29, 2007 11:02 am

Jeffy because there is simply no need for high yield. There are virtually no positive characteristics. You can eliminate the negative characteristics and simply add risk more efficiently by tilting to size and value or equities. Literally I cannot think of reason for individuals to own them

Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe » Thu Mar 29, 2007 11:04 am

Jeffy, just to be clear, instead of X for SV and Y for high yield. Just add bit more SV and get rid of junk and you are better off.There is simply almost nothing unique in junk that would cause you to want to add it and there are MANY negatives. Including negative skewness and excess kurtosis and tax inefficiency and call features and clawbacks, and on and on.

User avatar
jeffyscott
Posts: 8495
Joined: Tue Feb 27, 2007 9:12 am
Location: Wisconsin

Post by jeffyscott » Thu Mar 29, 2007 2:05 pm

larryswedroe wrote:Jeffy because there is simply no need for high yield. There are virtually no positive characteristics. You can eliminate the negative characteristics and simply add risk more efficiently by tilting to size and value or equities. Literally I cannot think of reason for individuals to own them
I'm well aware of your opinion of high yield. That was not the question...the question was why not do a valid comparison of a portfolio that is similar to C, with the addition of high yield...just as, despite being against using high yield, you did a valid comparison between A and B.
Time is your friend; impulse is your enemy. - John C. Bogle

Randy
Posts: 34
Joined: Thu Mar 01, 2007 12:02 pm
Location: Sandy Springs, GA

Post by Randy » Thu Mar 29, 2007 2:51 pm

The analysis of the three portfolios raises a question in my mind. Does this lead to a new way to decide how to build a portfolio? It seems to me many people for an AA that is designed to produce a specific return, then figure out the least risk way to get it. Perhaps we should start with the desired Std. Dev. then figure what portfolio providing that SD gives the best returns. So one would decide risk tolerance first, then get the "best deal available" for that much risk.
Randy

BrianTH
Posts: 1340
Joined: Tue Feb 20, 2007 6:10 pm

Post by BrianTH » Thu Mar 29, 2007 3:05 pm

Larry,

I believe you were looking at the corporate equities chart.

The "US corporate and foreign bonds" chart (although I admit I am not sure what foreign means in this context--I guess foreign bonds trading on US exchanges?) is the one I am looking at. There the percentages are much lower for households and mutual funds.

Edit: By the way, the next chart down is munis, and there (as one would expect) households and mutual funds really dominate.

Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe » Thu Mar 29, 2007 3:14 pm

Jeffy
There is NO REASON to do that, you simply replicate the junk holding by shifting equity and fixed income allocations--get same exposures but more efficiently. That is why I am against junk. No reason to own it.

Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe » Thu Mar 29, 2007 3:19 pm

Brian

Something weird about that table, how can the total of US and foreign be less than US alone.

At any rate, the US data shows high percentage

Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe » Thu Mar 29, 2007 3:46 pm

Jeffy
Just to be clear, most of the returns of junk are explained by the five factor model plus the tax differential. Which is why junk is a HYBRID security, exhibiting both equity and fixed income risks--which you can recreate in different mixes as I did. And you can do so while eliminating the negatives of junk

1.As a hybrid security, they provide exposure to equity risks in tax inefficient manner (interest income instead of capital gains).
2.Small amount of unique risk, resulting in only a small diversification benefit.
3.Returns exhibit negative skewness and excess kurtosis (increased risk of large losses).
4.Negative correlation to inflation.
5.Need to pay a mutual fund to obtain sufficient diversification, the costs of which reduce the expected return.
6.Asymmetric risk created by presence of calls and clawbacks. Exacerbating the problem is that low-grade bonds generally have weaker call protection than high-grade bonds. In addition, they are more likely to be called because the credit quality of low-grade bonds is more likely to rise.
7.Does not fit any of the main roles fixed-income investments should play in a portfolio—provide liquidity for emergencies, create a stable cash flow or reduce the overall risk of the portfolio.

BrianTH
Posts: 1340
Joined: Tue Feb 20, 2007 6:10 pm

Post by BrianTH » Thu Mar 29, 2007 4:50 pm

Larry,

The first table is stocks, not bonds, so the second table isn't related to the first.

Edit: By the way, you are right about mutual funds--they are 70% held by households according to the table at the bottom, so you can more or less attribute at least 70% of mutual funds to individuals as well. Again, thought, I think that means individuals hold a lot of the stock market, and even more of the munis market, but much less of the corporate bond market.

User avatar
Chas
Posts: 846
Joined: Sat Mar 24, 2007 8:41 pm
Location: America

Re: High Yield in Portfolio

Post by Chas » Thu Mar 29, 2007 5:11 pm

larryswedroe wrote:I had the following analysis done a while ago and plan on updating it shortly but thought I would share it.
The idea was to create three portfolios, One with 10% allocation to high yield via Vanguard's fund.
The other was to increase equity allocation a bit until I got the same return from portfolio, but eliminated junk bond risks---and see if got lower or higher SD (better tax efficiency not even considered)
The third was to tilt to sv instead, RAISE fixed income allocation and lower equity, getting same return and see if SD could be lowered
Here were the results my assistant came up with.

PORTFOLIO A (P-A): 60% CRSP 1-10, 30% Lehman Inter. Govt Bond Index, 10% VWEHX.
PORTFOLIO B (P-B): 61.5% CRSP 1-10, 38.5% Lehman Inter. Govt Bond Index.
PORTFOLIO C (P-C): 42.5% S&P 500 Index, 7.5% FF Small Value Index, 50% Lehman Inter. Govt Bond Index.


The annualized return for P-A from 1979-2005 was 11.74%. The annual
standard dev was 10.65. P-B also had an annualized return 11.74%.
However, its standard dev was 10.46. Some would say the difference in
std dev is small, but that's the whole point. We've completely
eliminated exposure to hi-yield, and we've BARELY added any more equity
- and we get virtually the same risk/return profile. I think the
comparison between P-A and P-B is your strongest argument against the
credit risk being rewarded - moreso than comparing P-A and P-C, which
is interesting but argues a different point.


The annualized return for P-C is 11.74% as well. However, the annual
standard dev is 8.66 - significantly less than 10.65 for P-A. Note that
this portfolio has the most fixed income - 50%. I think this supports
our stanard "take more risk w/ equities and have a higher allocation to
conservative fixed income" approach.

Draw your own conclusions.
I must be completely confused because I ran a total returns portfolio using M* to compare TRs for VBIIX verses VWEHX from 6/24/96 to present. (This is the only method I know that will include reinvested dividends if you set up a transaction PF). VBIIX outperformed the hi-yield VWEHX, although with much greater volatility according to M* risk figures. However, during the bear market period 2000-2002 VBIIX was a stellar performer, while hi-yield VWEHX just managed to hold level. Does this make sense? Not to my way of thinking. Am I totally off-base here for some reason?

Post Reply