Is bond diversification as important as stock diversificatio

[hoops777]

I have read many posts concerning bond funds and the result of prolonged interest rate hikes.The bogleheads are 100% behind owning the total stock market which can be owned by purchasing one fund.However the same does not apply to owning the total bond fund.So my question is wouldn't it be more appropriate to own some short term and inter term muni funds,a very conservative floating rate fund like Fidelitys,a high yield so called junk fund of the highest quality and some tips or Ibonds in addition to the total bond fund?I can look it up but I am not certain if the total bond has any international exposure,if not,maybe a small amount of high quality international as well??

[larryswedroe]

Hoops
Basically there is no reason to own a short, intermediate and long term fund as your risk is basically the weighted average maturity of the holdings, so might as well just hold the intermediate fund.
As to high yield, some like it because you gain a small amount of unique risk (much of it liquidity which you get in small stocks), but if you do own it you must recognize that you are holding a hybrid security with both stock and bond risk, and you pick up optionality (call risk) and the evidence is that it has not been well rewarded and you have to pay more for it and it's less tax efficient if you have choice of location.
Owning MBS takes optionality risk which I don't understand why people would take it for the very small premium
So bottom line is IMO you just don't need it, you don't get the geopolitical and economic diversification you get when you diversify internationally and EM. Adding EM bonds increases volatility, though adding some diversification benefit. Since IMO the main role of FI is to dampen risk, adding that risk goes the other way. But if you understand it and account for it, fine.
The answer is you can really keep bonds simple, owning only Treasury, agencies, CDs and AAA/AA rated munis and you're fine, taking all the risk you need on the equity side
Best wishes
Larry

[hoops777]

Larry ,thank you for your response.One last question concerning AAA and AA muni's.Do you recommend those individually or is it possible through a fund?Actually I guess the same thing applies to treasuries.Vanguards inter fund is about 50 % AA and 16% AAA.Thank you.

[ivesjl]

so might as well just hold the intermediate fund.

Would you mind commenting on how you select an intermediate fund? For example, Vanguard has two funds that seem very similar to me: VFICX and VFITX.

Would you consider consider the government fund, VFITX, to be less diversified?

How will this affect a portfolio's stability/returns over time? Why would these funds be superior to VBMTX?

Just for fun, I ran a comparison of a 2 fund portfolio (50/50, from 1985-2011, where the other 50% is in the total stock market), and here is what I found:

1. The two bond funds are highly correlated:
2. The portfolio that used the intermediate term fund had a smaller standard deviation with slightly higher returns:

But is this a reliable result? Is there a benefit to holding VBMTX over the long run?

[Bradley]

As to high yield, some like it because you gain a small amount of unique risk (much of it liquidity which you get in small stocks), but if you do own it you must recognize that you are holding a hybrid security with both stock and bond risk, and you pick up optionality (call risk) and the evidence is that it has not been well rewarded
Larry

Larry,

Using Morningstar performance comparison data, it shows VWEAX out performing BND over the past 1yr, 3yr, 5yr, 10yr and 15yr time periods. It appears those investor have been well rewarded.

Bradley

[Doc]

As to high yield, some like it because you gain a small amount of unique risk (much of it liquidity which you get in small stocks), but if you do own it you must recognize that you are holding a hybrid security with both stock and bond risk, and you pick up optionality (call risk) and the evidence is that it has not been well rewarded
Larry

Larry,

Using Morningstar performance comparison data, it shows VWEAX out performing BND over the past 1yr, 3yr, 5yr, 10yr and 15yr time periods. It appears those investor have been well rewarded.

Bradley

BND hasn't existed for that long. I assume you are referring to the mutual fund share class.

Take a look at this 1 yr rolling return chart. http://quote.morningstar.com/fund/chart ... %2C0%22%7D The big differences in the high yield and TBM funds really stand out. Using long term growth charts or worse yet long term average annual returns can often hide big differences in fund performance. After all if we wait long enough stocks are going to outperform bonds so why have bonds at all.

[magician]

Basically there is no reason to own a short, intermediate and long term fund as your interest rate risk is basically the weighted average maturity duration of the holdings, so might as well just hold the intermediate fund.

There are, of course, risks other than interest rate risk that will affect different maturities differently.

[nisiprius]

Bond diversification, at least among investment-grade bonds, is not as important at stock diversification, because investment-grade bonds are not as diverse as stocks.

We can see this from historical data in the Ibbotson 2010 Classic Yearbook, which says, 1970-2009

the correlation of intermediate government bonds was
0.88 with long-term corporate bonds
0.91 with long-term government bonds

In contrast, the correlation of large-company stocks was
0.72 with small-company stocks
0.66 with international stocks as represented by the EAFE index

We can also infer this pretty reasonably from the nature of the assets themselves. Stocks, which are tied to the year-by-year business success of their issuer, bonds promise specific numbers of dollars will be paid on specific dates. For investment-grade bonds, credit risk isn't a large factor--we assume the payments will be made. The value of the bond, then, is based mostly on bond math, which everyone can do. Since you don't know for sure what interest rates will be, you have uncertainty due to interest rate uncertainty, which can affect different bonds with different future streams of payments differently. But still, bond values, at least to the rough precision with which ordinary investors need to know them, are mostly determined by small number of factors, which everyone knows, and which affect most bonds in the same way.

[Bradley]

Doc,

Using Morningstar performance comparison data, it shows VWEAX out performing VBTLX over the past 1yr, 3yr, 5yr, 10yr and 15yr time periods. It appears those investor have been well rewarded. It is easy to go back in time and find periods of time that the TBM outperformed high yield. I do believe the Morningstar data contradicts Larry's blanket statement that holding high yield has "not been well rewarded".

Bradley

[nedsaid]

I agree with Larry Swedroe's comment that really what most investors need is an intermediate term bond fund.

I have plain vanilla intermediate term bond funds, TIPs, and a bit of foreign bond funds in my fixed income investments. I do have a balanced fund that owns a lot of high yield bonds. I don't feel strongly one way or another about high yield bonds.

For most people, a good old fashioned intermediate term bond fund works the best. The exception is that retirees and soon to be retirees who are worried about loss of principle might hold a big chunk of their fixed income investments in short term bond funds. The reason for this is not being able to reinvest the dividends when retired. I would presume that retirees would be taking the dividends to live on. Near retirees would not have enough years for the reinvestment to work its magic.

These low interest rates create quite a problem for retirees.

[Doc]

Doc,

Using Morningstar performance comparison data, it shows VWEAX out performing VBTLX over the past 1yr, 3yr, 5yr, 10yr and 15yr time periods. It appears those investor have been well rewarded. It is easy to go back in time and find periods of time that the TBM outperformed high yield. I do believe the Morningstar data contradicts Larry's blanket statement that holding high yield has "not been well rewarded".

Bradley

It depends what you mean by "rewarded". If your reward is only your FI portfolio taken in isolation you might have been well rewarded. But if you needed to take income every single year or rebalance into equities in a bear market then the high yield was not very rewarding.

Larry's philosophy is "take your risk on the equity side". Therefore taking equity like risk (aka high yield bonds) in your FI portfolio is not very rewarding.

I find it very interesting that in "normal" times much of the discussion involves how to rebalance between equities and FI. But when the yield on ten year Treasuries is less than two percent we throw out the whole idea of rebalancing and worry about how terrible it is to take a 2% below normal return on our FI for a few years.

If you never have to take money out of your portfolio you can take all the risk that you want. Why not forget FI all together and go with 100% stocks. Hey let's use only large growth also. In a 100 years or so you will probalby be well ahead of my 50/50 TSM/TBM portfolio.

[larryswedroe]

Hoops
Muni market is not as transparent as is Treasury so if going to buy individual bonds on own probably not a good idea for most investors. First, you have to know what you are doing. You cannot rely solely on ratings. You should also know to avoid highly rated bonds from sectors with poor persistence of ratings (health care, multifamily are two examples), sticking only to GOs and essential services and then you should be monitoring the ratings for changes. But if you have the time and knowledge then I would suggest buying at issuance so you don't have to pay a spread. We buy individual bonds for our clients, buying several billion a year, putting investment firms in competition, so we get essentially the same prices Vanguard gets when it buys, and we pay for a credit monitoring service as well. So we feel comfortable buying individual bonds and diversifiying them ourselves. In 15 years have not had a credit loss. Of course that is not a guarantee will never happen. The other thing IMO that is important is to avoid bonds with significant call risk. We try to buy without calls but will take up a bond say with 10 years remaining if it has at least 80% call protection (at least 8 years).

For treasuries or government agencies you can buy them on your own, the prices are pretty transparent. Suggest agencies over Treasuries as you pick up yield with no more risk. And CDs often better.

IVESJI
If buying corporates you should want to stick with an investment grade, and then consider say 15% of it as equity risk, otherwise you're just "lying" to yourself about the risks in the portfolio. And would also recommend avoiding callables, which I assume Vanguard buys. DFA does not. But if ST or perhaps even intermediate the risks not as great as for a longer fund.

Bradley
First your making the mistake of looking at the fund in isolation. The wrong way to do it. As I showed in my blog and in my books and with examples here, when combined with an equity portfolio the excess returns disappear, and in fact the portfolio has been a bit less efficient. And the risks show up at the wrong time, like 2008, which might push an investor over the cliff in terms of discipline, or max loss.
Also the excess returns to credit risk are by far the smallest (partly due to call risk), earning only a very small percent of the yield premium

Here are some links you should check out, hope you find them helpful
http://www.cbsnews.com/8301-505123_162-57559838/a-hidden-risk-in-high-yield-bonds/

http://www.cbsnews.com/8301-505123_162-37841232/high-yield-bonds-effect-on-portfolios/

http://www.cbsnews.com/8301-505123_162-57468875/why-junk-bonds-arent-a-good-portfolio-fit/

Magician
Yes besides rate risk you have call risk and credit risk, neither of which have been well rewarded and thus IMO best avoided

Larry

[magician]

Magician
Yes besides rate risk you have call risk and credit risk, neither of which have been well rewarded and thus IMO best avoided

There are also yield curve risk, liquidity risk, volatility risk, spread risk, and reinvestment risk that will likely affect different maturities differently, and several others that will more likely affect different maturities similarly.

Any way you slice it, there is a lot more to bond risk than merely duration (interest rate risk).

[Bradley]

Bradley
First your making the mistake of looking at the fund in isolation. The wrong way to do it. As I showed in my blog and in my books and with examples here, when combined with an equity portfolio the excess returns disappear, and in fact the portfolio has been a bit less efficient. And the risks show up at the wrong time, like 2008, which might push an investor over the cliff in terms of discipline, or max loss.
Also the excess returns to credit risk are by far the smallest (partly due to call risk), earning only a very small percent of the yield premium

Larry

Larry,
I can assure you that the excess returns did not disappear from my portfolio. VWEAX combined with VBTSX and VAIPX has provided for a much more efficient and productive portfolio. There is no doubt however that you can go back in time and create a series of events that may cause the excess returns attributed to VWEAX to disappear.

“My belief if that US corporate high yield bonds are a unique part of the US bond market and should be included in a portfolio. This view has worked very well for our clients over the years.”
-----------------------------Rick Ferri
-------http://www.bogleheads.org/forum/viewtopic.php?f=1&t=91779#p1322094



In All About Asset Allocation, Ferri states that, "Statistically, only about 25 percent of the default risk in high-yield corporate bonds can be attributed to the same factors affecting equity returns; however the results are not statistically significant. If you allocate 10 percent of your total portfolio to one of the B-BB rated bond funds listed at the end of the chapter, at most perhaps 2 percent of that could be considered equity related. ... That being said, very-low-quality bond funds (those with an average credit quality of CCC or less) do have a higher correlation with equity returns."

Bradley

[mdpsychcrnp]

[/quote]
Larry,
I can assure you that the excess returns did not disappear from my portfolio. VWEAX combined with VBTSX and VAIPX has provided for a much more efficient and productive portfolio. There is no doubt however that you can go back in time and create a series of events that may cause the excess returns attributed to VWEAX to disappear.
[/quote]

Bradley,
Indeed. With all due respect to Mr. Swedroe, there is a compelling case that taking some credit risk is rewarded in a portfolio over time. Ferri is a proponent of this (20% of bond allocation to high yield), and, it appears, so is Bogle, with his recent suggestion to take a bit of credit risk. It's a fair argument that a slightly higher percentage of folks on this forum do not feel that credit risk is rewarded. I would argue that most folks of that stripe have an equity heavy portfolio. In my estimation the rewards are negligible if not non-existent except in more balanced porfolios (60/40 - 40/60). I, myself, hold 20% of my bond allocation in VWEAX, and split the remainder 40% BND and 40% VCIT.

In the end, there are lots of very bright and capable folks who can make compelling arguments for any kind of bond/equity tilt in a portfolio. Folks can similarly make a compelling argument to have a small allocation to commodities or to gold. Or tilting to high yield corporate bonds, small cap value, REITS, emerging market small cap value (the BIG winner of the last decade), micro-cap stocks, dividend stocks, ad nauseaum. What is true is that for any strategy you can dream up, you will be able to find a particular historical period where that strategy trumps the others.

What matters most is an asset allocation that mirrors one's need and ability to take risk. That, and attention to expenses. Anything else is mostly dependent upon which strategy has been most effective between when you begin said strategy and when you choose to no longer follow that strategy.

[Taylor Larimore]

Hoops:

In my opinion, ANY Vanguard good-quality, short or intermediate-term bond fund will do the job of providing safety and income for our portfolios. Trying to forecast which bond fund will outperform other bond funds during some future period is unlikely to be successful.

Total Bond Market Index Fund is a good choice for its one-fund bond diversification. Whatever happens, the investor will not have all their bonds in a losing bond category.

Concentrate on stock funds for higher returns (and higher risk of loss). Stocks let us live well. Bonds let us sleep well.

Vanguard Bond Funds

Past performance does not guarantee future performance.

Best wishes.
Taylor

[Akiva]

Hoops
Basically there is no reason to own a short, intermediate and long term fund as your risk is basically the weighted average maturity of the holdings, so might as well just hold the intermediate fund.
As to high yield, some like it because you gain a small amount of unique risk (much of it liquidity which you get in small stocks), but if you do own it you must recognize that you are holding a hybrid security with both stock and bond risk, and you pick up optionality (call risk) and the evidence is that it has not been well rewarded and you have to pay more for it and it's less tax efficient if you have choice of location.
Owning MBS takes optionality risk which I don't understand why people would take it for the very small premium
So bottom line is IMO you just don't need it, you don't get the geopolitical and economic diversification you get when you diversify internationally and EM. Adding EM bonds increases volatility, though adding some diversification benefit. Since IMO the main role of FI is to dampen risk, adding that risk goes the other way. But if you understand it and account for it, fine.
The answer is you can really keep bonds simple, owning only Treasury, agencies, CDs and AAA/AA rated munis and you're fine, taking all the risk you need on the equity side
Best wishes
Larry

I generally agree with Larry, but I'll note that international interest rates don't move in tandem with US ones, so that if you have access to a *currency hedged* international bond portfolio, this can improve diversification. I think Vanguard had a thing showing that increasing your allocation helped up to about 40%. (Vanguard was supposed to launch one last year, but it got put on hold for some reason.)

Secondarily, the US has technically defaulted on its obligations twice in the last 100 years and other developed countries have had similar problems in various crisis situations. So I think it is a mistake to think that US backed debut is a 100% sure thing. It's just that default in a way that hurts investors seems *very very* unlikely. Still, to the extent that this worries you, having some international diversification would be good.

Also, it is worth noting that for some reason, the spread between Aaa bonds and Baa bonds seems uniformly too high given the additional risks assumed. (It averaged 1.2% historically and after you account for the additional default risk, this translates to a .97% premium.) So you can get somewhat higher returns on your fixed income portfolio because of this effect, *but* corporate bonds are much more correlated with the market and consequently, these higher returns may increase you portfolio's risk by more than it increases your returns. "Junk" bonds do not have this effect and are in fact much riskier than other bonds for little or no additional expected returns.)

Doc,

Using Morningstar performance comparison data, it shows VWEAX out performing VBTLX over the past 1yr, 3yr, 5yr, 10yr and 15yr time periods. It appears those investor have been well rewarded. It is easy to go back in time and find periods of time that the TBM outperformed high yield. I do believe the Morningstar data contradicts Larry's blanket statement that holding high yield has "not been well rewarded".

Bradley

If you put this in a reasonable portfolio, you'll see that it doesn't help your risk or your returns because all the risk shows up at the wrong time and because the returns have too much volatility. In return for all of this extra risk, you get at most an extra .64% annually. (That's if you invest directly. Evidence is that the funds that track this space don't do as well as the indexes used in academic studies because of illiquidity and other problems.) Furthermore, the difference between a .3% default rate at Baa and a .03% default rate at AAA is immaterial to most people (and yet the spread has historically wider than it should be based on these rates), but going from a .3% default rate to the 15% default rate on C rated bonds is huge. The .64% return premium you get isn't nearly enough to offset this. (I haven't run a study, but if you are willing to take the increased equity correlation and other issues with junk bonds, then I'd suspect that you are actually better off with preferred stock.)

[EDN]

When devising a bond allocation, the first step is to ask yourself what role the bonds will play. If you are looking to lower the risk of an all-stock portfolio, then you will logically look to the safest high-quality bonds of about 5 year durations (5yr has been about the sweet spot of the term spectrum and high-quality bonds have lowered risk more efficiently than higher-risk credit allocations).

For an accumulator, it is unlikely they would need more than Vanguard Int'd Treasury or Int'd Muni if taxable $ is involved and investor is in higher tax brackets. If the investor is retired and is more exposed to the risk of unexpected inflation, adding TIPS to either above is a worthwhile consideration.

I take bond diversification one step further by putting about 30% of the allocation (50% of nominal segment) into fully-hedged international government bonds (DFA World exUS Government) for a bit more diversification beyond a US-only allocation (which matches my typical allocation in equities), but I wouldn't for a minute say this has much impact overall on a portfolio. For a retiree at 60/40 (assuming 100% IRAs), this is only 12% of a portfolio (along with 12% US Government and 16% TIPS), so it doesn't have a significant impact -- but IMO the additional diversification outweighs the added complexity.

The US-only allocation is also just fine.

Eric

[athrone]

Do you live in a country where the Bonds are free of interest rate, inflation, or default risk? If so, then diversification is not necessary.

[Bradley]

If you put this in a reasonable portfolio, you'll see that it doesn't help your risk or your returns because all the risk shows up at the wrong time and because the returns have too much volatility.

Akiva,
I have put VWEAX in a reasonable portfolio and I have seen it has improved returns. What would you have replaced VWEAX with in Rick's "tight fisted" portfolio to improve return performance?

http://online.wsj.com/article/SB1000142 ... %3Darticle

I and many of my colleagues use this model(minus BRSIX) adjusted for personal risk tolerance and have been well rewarded.

Bradley

[Akiva]

If you put this in a reasonable portfolio, you'll see that it doesn't help your risk or your returns because all the risk shows up at the wrong time and because the returns have too much volatility.

Akiva,
I have put VWEAX in a reasonable portfolio and I have seen it has improved returns. What would you have replaced VWEAX with in Rick's "tight fisted" portfolio to improve return performance?

http://online.wsj.com/article/SB1000142 ... %3Darticle

I and many of my colleagues use this model(minus BRSIX) adjusted for personal risk tolerance and have been well rewarded.

Bradley

You may have gotten better returns in recent history, but given what the covariances are, and given the higher default rates, the historic return premium to junk bonds cannot possibly be sufficient compensation. Your back-test merely proves that you got lucky, not that you knew better than the weight of academic evidence.

[Bradley]

Your back-test merely proves that you got lucky, not that you knew better than the weight of academic evidence.

Akiva,
There were no back-tests. The results are actual results. I'll ask you again. What would you have replaced VWEAX with in Rick's "tight fisted" portfolio to improve return performance?

Bradley

[Akiva]

Your back-test merely proves that you got lucky, not that you knew better than the weight of academic evidence.

Akiva,
There were no back-tests. The results are actual results. I'll ask you again. What would you have replaced VWEAX with in Rick's "tight fisted" portfolio to improve return performance?

Bradley

The issue isn't "return", it's risk-adjusted return, and you haven't shown that it doesn't increase risk more than it increases return. Furthermore, given the length of time we are talking about, even if it performed better in the recent past, that isn't proof that it actually does this over long periods of time or will continue to do so going forward. We do know the historic default rates on junk bonds and we know the historic credit spread they've enjoyed and we can see that this spread just barely covers the increased default risk, leaving next to nothing to cover the increased stock-like behavior, call risk, etc.

Similarly, lots of people invested in MBS for a long time and saw better returns, and when people like me pointed out that the few tens of basis points they were getting for the massive risks they were taking could not possibly be worth it, these concerns were brushed aside. Then the financial crisis hit and lots of people got screwed. Had the government not bailed out the GSEs even more would have been hit. Again the long history of better returns wasn't nearly enough to compensate for the risk that was being taken, outstanding recent historical performance not withstanding.

As for what I'd do instead, I'd suggest that you stop chasing yield and focus on risk adjusted returns. I think the suggestion to diversify internationally is a good one because it does this.

If you want to take extra credit risk, investment grade bonds like those held by DFA's DFTEX fund are fairly compensated. (But again, correlations with the rest of the portfolio are problematic.) And if you want to go after even higher yield, preferred stocks have similar risk characteristics to junk bonds without the massive default risk.

[Bradley]

As for what I'd do instead, I'd suggest that you stop chasing yield and focus on risk adjusted returns. I think the suggestion to diversify internationally is a good one because it does this.

If you want to take extra credit risk, investment grade bonds like those held by DFA's DFTEX fund are fairly compensated.

Akiva,
Now your just being plain silly. Rick, his portfolio designs nor his strategy chase performance. VWEAX is part of a strategic allocation. Further how can you conclude that DFTEX fairly compensates investor? 2011-2012 are it's only full years of data. How can you say DFTEX fairly compensate investor while VWEAX does not?

Bradley

[Akiva]

As for what I'd do instead, I'd suggest that you stop chasing yield and focus on risk adjusted returns. I think the suggestion to diversify internationally is a good one because it does this.

If you want to take extra credit risk, investment grade bonds like those held by DFA's DFTEX fund are fairly compensated.

Akiva,
Now your just being plain silly. Rick, his portfolio designs nor his strategy chase performance. VWEAX is part of a strategic allocation. Further how can you conclude that DFTEX fairly compensates investor? 2011-2012 are it's only full years of data. How can you say DFTEX fairly compensate investor while VWEAX does not?

Bradley

1) No person named "Rick" has posted in this thread. You linked to some article that claims to report what Rick Ferri invests. He's on this board and can speak for himself. If he's aware of research that refutes the mainstream view that junk bonds aren't compensated, then he can bring it up. Otherwise there's no point in me arguing with it because I can only speculate about his reasons for doing that. All I can do without more information is to say that it represents a deviation from the general consensus on bond investing.

2) I know that a corporate bond fund like that compensates the investor because the abnormally high compensation for taking that spectrum of credit risk is a documented established phenomena in the academic literature. (Much like the value effect or the small cap premium.) The evidence is exactly the opposite for high yield bonds over a long period of time.

3) I really don't think you have a good grasp of the fact that past performance does not predict future returns. You keep arguing that since someone got lucky and had good returns in recent history with junk bonds, they should be okay going forward, and similarly you say that we can't know if corporate bonds compensate you for the additional risk because the particular fund I used as an example hasn't been around very long. It takes an enormous amount of data to find a statistically significant effect, and you can't find it simply by looking at a chart of past returns. The academic studies people rely on to say the things that I'm saying have extremely large data sets and are extremely careful with their statistics to deal with this problem. Only a similarly careful analysis should be used to deviate from the consensus advice.

[EDN]

In response to the question "how to improve results without HY bonds", I use a basic rule of thumb: hold more in equities and tilt more to small and value within equities. HY is both equity like and the companies that issue HY bonds tend to be smaller and more value oriented firms, so this doesn't result in a meaningfully different portfolio profile.

I'll use the same investments in each example and just vary the weights slightly based on the attached article from above:

The "base" allocation (60% stock and 40% "bond") is 27% Russell 3000 Index, 9% S&P 600 Value, 6% Wilshire REIT Index, 5% Europe Index, 5% Pacific Index, 3.5% DFA Int'l Small Value Index, 4.5% DFA Emerging Markets Core Index, 30% Barclays Aggregate Bond Index, and 10% Barclays HY Bond Index.

The "adjusted" allocation (70% stock and 30% bond) is 21% Russell 3000 Index, 21% S&P 600 Value Index, 7% Wilshire REIT Index, 14% DFA Int'l Small Value Index, 7% DFA EM Core Index, and 30% Barclays Aggregate Bond Index.

From 1994-2012 (2003-2012), the return of the "base" allocation was +8.3% (+9.3), and +9.1% (+10.4%) for the "adjusted" allocation. Both allocations were within 0.5 annual SD of each other, and had similar downside losses during server bear markets: in 2002 (2008), the "base" allocation lost -5.2% (-24.2%), while the "adjusted" allocation lost -3.6% (-24.9%).

I am of the opinion that the "adjusted" allocation is a better overall mix of assets for a given level or risk and return as it sticks with stocks and high-quality bonds without anything "in-between".

Eric

[Bradley]

You linked to some article that claims to report what Rick Ferri invests. He's on this board and can speak for himself. If he's aware of research that refutes the mainstream view that junk bonds aren't compensated, then he can bring it up. Otherwise there's no point in me arguing with it because I can only speculate about his reasons for doing that.

Akiva,
The article I referenced was/is a published article in the WSJ.

http://online.wsj.com/article/SB1000142 ... %3Darticle

The journalist published a model portfolio that Rick constructed for many of his clients. Why would he use VWEAX if he thought investors were not fairly compensated for their risk? Rick manages over a $Billion, is the head of research at his firm, is a CFA, a respected investment advisor yet you do not believe his clients are fairly compensated with their VWEAX allocation?

Bradley

[Akiva]

You linked to some article that claims to report what Rick Ferri invests. He's on this board and can speak for himself. If he's aware of research that refutes the mainstream view that junk bonds aren't compensated, then he can bring it up. Otherwise there's no point in me arguing with it because I can only speculate about his reasons for doing that.

Akiva,
The article I referenced was/is a published article in the WSJ.

I don't see your point. Articles that professionals would consider misleading are published in major news papers all the time. Laywers complain that legal reporters can't cover cases correctly. Doctors complain that they misreport medical findings. Why would you think that they'd be more competent when it comes to finance? If you took the financial media seriously, then you'd probably think that most of what Bogleheads teaches is wrong...

If you want Rick Ferri's opinion on this you should PM him since he frequents this forum and ask him to respond, but I'm not going to criticize him on the basis of what some reporter may or may not have accurately reported about him and I'm not going to make assumptions about what his thought process might or might not be.

Why would he use VWEAX if he thought investors were not fairly compensated for their risk?

I don't know why he uses a high yield fund. You should ask him that. I'm just reporting, correctly, that the academic consensus is that these investments do not pay-off in risk adjusted terms.

Rick manages over a $Billion, is the head of research at his firm, is a CFA, a respected investment advisor

All moot for purposes of this discussion.

yet you do not believe his clients are fairly compensated with their VWEAX allocation?

All of the actual evidence is that long-term investments into junk bonds have not been appropriately compensated for the risk you are assuming. Maybe he knows something the rest of us don't, but I can't give someone advice based on hypothetical knowledge that might or might not exist. I'm going to report the best available evidence as I understand it until I find convincing evidence to the contrary.

[Doc]

My belief if that US corporate high yield bonds are a unique part of the US bond market and should be included in a portfolio. This view has worked very well for our clients over the years. Larry has a different opinion about high yield, and we agree to disagree. That's what makes a market.

viewtopic.php?p=1322094#p1322094

I'm in the Swedroe camp but I'm only a camp follower.

[Akiva]

My belief if that US corporate high yield bonds are a unique part of the US bond market and should be included in a portfolio. This view has worked very well for our clients over the years. Larry has a different opinion about high yield, and we agree to disagree. That's what makes a market.

viewtopic.php?p=1322094#p1322094

I'm in the Swedroe camp but I'm only a camp follower.

Thanks for linking to that. It seems, based on the context and the later part of that same post (follow the link), that he thinks high yield is good *for now*, much like emerging market bonds once were (but are not right now). This is a fundamental analysis and his view could change when market conditions change. (There's nothing wrong with this.) But that's a whole different matter than saying that such an allocation would always have been rewarded and should be part of a purely passive portfolio.

As for why he thinks they are a good investment at this time, his research is proprietary and I don't expect him to share it in the same way that an academic would share their research by publishing. And while he's a smart guy, I can't just take him at his word on this. Maybe he's right, maybe he isn't, but barring overriding persuasive evidence, I'll stick with the academic consensus.

[larryswedroe]

I would just add this
It's important to note that when you look at historical returns series you need to be careful of starting and end points, especially if valuations were particularly high or low at the start or end points
Anyone now looking at HY today is looking at end point with valuations at historically high levels as investors chase yield in any form due to Fed policy of "repression."
In my research I have used the longest period available, from start of Vanguard fund, and it has shown in each case inefficiency, And it doesn't even consider location issues for those with choice.
And it's also what the peer reviewed literature has shown. Not my opinion's or Rick's.
For anyone interested in subject I suggest reading Martin Fridson's work --he is generally considered IMO the "expert" on high yield

Best wishes
Larry

[magician]

Rick . . . is a CFA . . . .

Rick is a CFA charterholder. (If he calls himself "a CFA", it's a violation of the CFA Code of Ethics. Weird, I know, but them's the rules.)

[Bradley]

Took Larry’s advice to read Fridson, below is what I found. Maybe he has changed his mind?



"Still, Martin Fridson, chief high-yield strategist at Merrill Lynch, gives junk the thumbs up: "The yields more than compensate for the risk." Why listen to Fridson? Because he's usually right..............................................Betting on junk bonds is no sure thing. But listening to a market pro like Fridson is one way to stack the odds in your favor."
------- By Susan Scherreik

http://www.businessweek.com/stories/200 ... o-buy-junk

[Akiva]

Took Larry’s advice to read Fridson, below is what I found. Maybe he has changed his mind?



"Still, Martin Fridson, chief high-yield strategist at Merrill Lynch, gives junk the thumbs up: "The yields more than compensate for the risk." Why listen to Fridson? Because he's usually right..............................................Betting on junk bonds is no sure thing. But listening to a market pro like Fridson is one way to stack the odds in your favor."
------- By Susan Scherreik

http://www.businessweek.com/stories/200 ... o-buy-junk

That was his evaluation of fundamentals back in 2002. Surely his view has changed since then. And again, you keep confusing a tactical allocation based on fundamental outlook with the long run performance of an asset class. Most investors on this board aren't willing to engage in fundamental analysis to dynamically adjust their allocations -- it goes against the spirit of passive index investing.

Put another way, Fridson thought they were a good deal back in 2002. Ferri thought they were a good deal a year ago. Even assuming they are right, how are you going to know when these guys stop thinking they are a good deal and close out their allocation (like Ferri did with emerging market bonds)? If you don't understand the thought process that goes into their analysis, you can't replicate it and are just relying on experts of a different sort than the gurus and cranks that the people on this forum have learned to avoid.

Yes, sometimes things like junk bonds can pay off. And sometimes (like between 1896 and 1932) stocks have negative returns for a very long time. But the general long-run performance characteristics show that junk bonds don't pay and that stocks do.

[Bradley]

But the general long-run performance characteristics show that junk bonds don't pay and that stocks do.

Where did you get your data from to arrive at your conclusion?


The opposite is true over the long run. Using Vanguards S&P 500 fund vs Vanguards HY the long run performance data shown by M* is as follows;


VWEAX 5yr...9.25%
VFIAX 5yr..4.65%


VWEAX 10YR...8.08%
VFIAX 10YR...7.26%


VWEAX 15YR...6.02%
VFIAX 15YR...4.82%

[larryswedroe]

Bradley as I said you keep making a basic mistake, looking at things in isolation which is totally irrelevant unless it's the only asset you own

From my blog post which had the data for the longest period then available.[url][/url]

Looking only at returns (while not considering risk), the Vanguard High Yield Corporate Fund looks like a good investment compared to its benchmarks. (Please note that the Vanguard fund's annualized return is measured from its inception on Dec. 27, 1978 through 2009. The benchmarks' annualized returns are measured from 1979 through 2009.)
Vanguard High Yield Corporate -- 8.72 percent
Five-Year Treasury Notes -- 8.23 percent
Gov't Bond Index (1-30 Years) -- 8.45 percent
Barclays Capital Credit Bond Index Intermediate -- 8.70 percent
Using Morningstar's database, which goes back 20 years, we can look at how the addition of the Vanguard fund affects portfolios. During this period (1990-2009), the high-yield fund returned 7.4 percent with a standard deviation of 12.9 percent and five-year Treasuries returned 6.65 percent with a standard deviation of just 6.2 percent.

As we did on Wednesday, let's look at two portfolios that are rebalanced annually:
Portfolio A has a 60 percent allocation to the S&P 500 Index and a 40 percent allocation to the Vanguard fund. This portfolio returned 8.1 percent per year with a standard deviation of 16.1 percent.
Portfolio B has a 60 percent allocation to the S&P 500 and a 40 percent allocation to five-year Treasuries. This portfolio returned 8.2 percent but did so with a standard deviation of just 11.6 percent.
Thus, while the high-yield fund produced higher returns during this period, its addition to the portfolio not only produced a lower return, but it resulted in a portfolio with almost 40 percent greater volatility. Portfolio B was a much more efficient one. Note that the quarterly correlation of the high-yield fund to the S&P 500 was 0.62, while the correlation of the five-year Treasury to the S&P 500 was -0.28, making the five-year Treasury note a much more effective diversifier of the risks of equities.

[Bradley]

Portfolio A has a 60 percent allocation to the S&P 500 Index and a 40 percent allocation to the Vanguard fund. This portfolio returned 8.1 percent per year with a standard deviation of 16.1 percent.

Larry,

You keep making up portfolio examples that are unrealistic and irrelevant. Who among us uses 100% HY as the fixed portion of a portfolio? No one. Rebalancing 100% HY with equities? This is one of the most extreme and outlandish examples you could come up with. Try testing Rick’s “tight fisted model portfolio” or nearly any other properly constructed portfolio and tell us how they perform compared to your two fund examples. When combined with VAIPX and VBTSX, VWEAX will test much better than your poorly constructed portfolio examples.

Bradley

[Akiva]

But the general long-run performance characteristics show that junk bonds don't pay and that stocks do.

Where did you get your data from to arrive at your conclusion?

I'm using the same extensive data that respected academic studies rely on. It is much more extensive than the short-term history of a couple of mutual funds. At a minimum, you can get a bond database from Moody's that includes every bond they've rated since 1975. (You could argue that the data before the junk bond revolution of the 80s is structurally different and should be excluded, but that only hurts the outlook for junk bonds.)

More generally, 5, 10, and 15 years are far too short of time periods to make the kind of judgments you want to make. If you had done that in 1932, you'd have concluded that stocks were a bad investment since they had done so poorly over the last 30+ years. And then by the time that you'd have had enough data to overturn that, you'd have bought into yet another long bear market. And by the time you'd have had enough data to overcome the bad returns again, you'd have missed most of the largest bull market in history. Doing what you keep trying to do does not lead to correct inferences about future expected returns.

[larryswedroe]

bradley
I don't think it makes a bit of different what percentage is used. I've shown that in other examples. And the point is to show that it is irrelevant what an asset does in isolation, which is what you keep showing. It only matters what the impact is on the portfolio's risk and return.

Note the literature is very clear, might try reading it.

Here is a new paper that supports Fridson's work on HY, basically the only place high yield has been rewarded has been the once investment grade that is newly distressed. The reason IMO that it has been rewarded here is you remove the I lose/I don't win game because of the distress which minimizes the call risk. Also you pick up the liquidity premium when the stock loses it's investment grade--making the call risk is less of an issue.

Best wishes
Larry

[Akiva]

bradley
I don't think it makes a bit of different what percentage is used. I've shown that in other examples. And the point is to show that it is irrelevant what an asset does in isolation, which is what you keep showing. It only matters what the impact is on the portfolio's risk and return.

Re: this. He doesn't actually need to go look up your other examples, since EDN already gave a perfectly good one in this thread that he's chosen to ignore. EDN pointed out that slightly increasing your equity exposure and your small/value tilts (so that your risk exposure was closer to the risk exposure you'd get with HY bonds), had actually done better (both in terms of return and risk) than simply including a small allocation to HY bonds:

In response to the question "how to improve results without HY bonds", I use a basic rule of thumb: hold more in equities and tilt more to small and value within equities. HY is both equity like and the companies that issue HY bonds tend to be smaller and more value oriented firms, so this doesn't result in a meaningfully different portfolio profile.

I'll use the same investments in each example and just vary the weights slightly based on the attached article from above:

The "base" allocation (60% stock and 40% "bond") is 27% Russell 3000 Index, 9% S&P 600 Value, 6% Wilshire REIT Index, 5% Europe Index, 5% Pacific Index, 3.5% DFA Int'l Small Value Index, 4.5% DFA Emerging Markets Core Index, 30% Barclays Aggregate Bond Index, and 10% Barclays HY Bond Index.

The "adjusted" allocation (70% stock and 30% bond) is 21% Russell 3000 Index, 21% S&P 600 Value Index, 7% Wilshire REIT Index, 14% DFA Int'l Small Value Index, 7% DFA EM Core Index, and 30% Barclays Aggregate Bond Index.

From 1994-2012 (2003-2012), the return of the "base" allocation was +8.3% (+9.3), and +9.1% (+10.4%) for the "adjusted" allocation. Both allocations were within 0.5 annual SD of each other, and had similar downside losses during server bear markets: in 2002 (2008), the "base" allocation lost -5.2% (-24.2%), while the "adjusted" allocation lost -3.6% (-24.9%).

I am of the opinion that the "adjusted" allocation is a better overall mix of assets for a given level or risk and return as it sticks with stocks and high-quality bonds without anything "in-between".

Eric

[Bradley]

bradley
I don't think it makes a bit of different what percentage is used. I've shown that in other examples. And the point is to show that it is irrelevant what an asset does in isolation, which is what you keep showing. It only matters what the impact is on the portfolio's risk and return.

Note the literature is very clear, might try reading it.

Here is a new paper that supports Fridson's work on HY, basically the only place high yield has been rewarded has been the once investment grade that is newly distressed.
Best wishes
Larry

Larry,

Basic math is all that is needed to prove the percentage used does make a difference. You keep saying read the literature. Specifically what should we read? Do you have a book title? Research paper? You say here is a new paper that supports Fridson but give no reference.

[William4u]

When I think of [safe, low risk non-corp] bond diversification, I just think of adding TIPS. Corp bonds are too risky and too like stocks to be a part of the "bond" part of a Boglehead asset allocation.

[Akiva]

When I think of [safe, low risk non-corp] bond diversification, I just think of adding TIPS. Corp bonds are too risky and too like stocks to be a part of the "bond" part of a Boglehead asset allocation.

In fairness, a large number of Bogleheads do hold substantial allocations to both MBS and corporate bonds via the Vangaurd Total Bond Index fund. Generally speaking I agree with Swedroe, but plenty of people think that you can at least arguably justify an allocation to investment grade corporate bonds. The rest of this stuff is substantially more questionable.

[larryswedroe]

Bradley

First I know it makes a difference what % you use in the outcomes, but when I've run the data with different percentages in past high-grade lost, no matter what the percentage.

Try this piece I wrote which also cites the literature, two papers including Fridson's
http://seekingalpha.com/article/58172-d ... gh-returns

Here is some other information

“Explaining the Rate Spread on Corporate Bonds,”3 by Edwin J. Elton, Martin J. Gruber, Deepak Agrawal, and Christopher Mann, posed three pertinent questions:
• How much of the spread between corporate and Treasury bonds is explained by expected losses from defaults? Some percentage of corporate bonds is likely to default, and defaults generally result in recovery rates below 100 percent. Furthermore, the lower the credit rating, the lower the recovery percentage. Investors must be compensated for the greater expected losses.
• How much of the spread is explained by the tax premium? Interest on U.S. government securities is not taxed at the state or local level. However, interest on corporate debt is taxed. Investors must be compensated for the tax differential.
• Is the incremental risk of corporate bonds systematic (nondiversifiable) or unsystematic (diversifiable)? Investors are compensated with risk premiums for systematic risks, but they are not compensated for unsystematic risks.

The study’s conclusions regarding these questions can be summarized this way:

• The difference in tax treatment and expected default losses does not adequately explain the spread. For example, the authors found that expected losses account for no more than 25 percent of the corporate spread. In the case of a 10-year A-rated corporate bond, just 18 percent of the spread was explained by default risk. The difference in tax treatment accounted for 36 percent of the spread. In other words, default risk accounts for a surprisingly small fraction of the premium in corporate rates over Treasuries. There must be another source of risk premium demanded by investors.
• The Fama/French Three-Factor Model (see Glossary) explains as much as 85 percent of the spread that is not accounted for by taxes and expected default loss. This means most of the spread is compensation for three risks: the risk of the overall stock market; the risk of small (versus large) companies; and the risk of value (versus growth) companies. The lower the credit rating—and the longer the maturity—the greater the explanatory power of the model (its inherent accuracy). Thus, much of the expected return to high-yield debt is explained by risk premiums associated with equities, not debt. These risks are systematic risks that cannot be diversified away.

A 2006 study, “Personal Taxes, Endogenous Default, and Corporate Yield Spreads,” by Howard Qi, Sheen Liu and Chunchi Wu, focused on the tax issue and found that personal taxes explain almost all of the spread between shorter-term high-grade bonds and Treasuries. That is exactly what we should expect, because there is almost no credit risk in short-term, high-quality bonds. Taxes explain about 60 percent of the spread for longer-term, high-quality bonds. This is logical, as the credit risk of corporate bonds—even high-grade ones—increases with time. But the picture is quite different for high-yield bonds: Personal taxes explain about 60 percent of the spread for shorter-term bonds and less than 40 percent for long-term bonds.4
These findings demonstrate three important facts. First, corporate debt contains very little risk that is not explained by other factors. As we have seen, most of the returns are explained by term risk, taxes and equity risks. Second, despite its low correlation to other portfolio assets, high-yield debt provides little benefit in terms of portfolio diversification. Third, it also helps explain why corporate bond spreads appear to be so large. (There is a tax component, as well as an equity component, in the security.)

And this paper
That high-yield bonds are hybrid securities is also supported by the findings of “Co-movements of Low-Grade Debt and Equity Returns of Highly Leveraged Firms,” a 1994 study by Hilary Shane that evaluated 208 low-rated bonds and an index of the stocks of the issuers represented in the bond portfolio. Shane concluded: “Significant positive correlations of the all-inclusive, low-grade bond portfolio with the matched equity portfolio and with the Treasury bonds support the intuition that low-grade bonds are hybrid securities.”

Here is another paper on the issue of hybrid security
One 2002 study on default risk, “Is Default Event Risk Priced in Corporate Bonds?”, by Joost Driessen, supports this theory. The author concluded that an event (default) risk premium is a significant determinant of excess corporate bond returns. Clearly, if there is an event that increases the likelihood of default, it will negatively impact the equity of the company, as well.7 That is why the Fama/French Three-Factor Model explains much of the spread not determined by default and taxes. High-yield returns move systematically with equities, and the risk of default increases when equity prices fall (and vice versa). The important point to remember is that correlations are not static values.

And other negative features of high yield
The problem for investors is that high-yield debt exhibits negative skewness. Mark Anson researched the period from January 1990 through June 2000 and found that high-yield bonds (as measured by the Salomon Smith Barney High Yield Composite Index) exhibited skewness of -0.43; similar to that exhibited by stocks. For the same period, the S&P 500 Index had skewness of -0.50.9
High-yield debt returns also exhibit high kurtosis.

Also from my book on alternatives

For the 29-year period from 1979 to 2007, the fund provided investors with a return of 9.01 percent per year. For the same period, the Lehman Brothers Intermediate Credit Bond Index, an index of investment grade bonds with maturities of one to ten years, returned 8.88 percent per year, almost matching the return of the Vanguard fund without taking the credit risks of junk bonds. Thus, while the spread between high-yield bonds and investment grade bonds can often be quite wide, the incremental return realized by investors was just 0.13 percent. This demonstrates investors should never confuse yield with returns. And that doesn't even take into account correlations. Nor the 2008 collapse.

As I showed in my book a more efficient alternative is simply to lower equity exposure and add small value tilt and stick with investment grade or Treasuries. You get superior results.

And here is what David Swensen, pretty smart CIO of Yale Endowment had to say

David Swensen, the highly regarded Chief Investment Officer of the Yale University endowment, stated: “Well-informed investors avoid the no-win consequences of high-yield fixed income investing.”

Here are some papers on subject

Martin S. Fridson, “Do High-Yield Bonds Have an Equity Component?” Financial Management (Summer 1994), p. 83.
Bradford Cornell and Kevin Green, “The Investment Performance of Low-grade Bond Funds,” Journal of Finance (March 1991), p. 32 and 47.
Dale Domain and William Reichenstein, Returns-Based Style Analysis of High-Yield Bonds, Journal of Fixed Income, Spring 2008.
Mark J.P. Anson, The Handbook of Alternative Assets, (Wiley, 2002), pp. 99-100.
Antti Ilmanen, Rory Byrne, Heinz Gunasekera and Robert Minikin, “Which Risks Have Been Best Rewarded?” Journal of Portfolio Management (Winter 2004), p. 54.
Martin Fridson, “Original Issue High-Yield Bonds,” Journal of Portfolio Management (Fall 2007).

Hope that helps
Larry

[hoops777]

Larry could you please be a little more specific

[Taylor Larimore]

Larry:

The time and research you put into your comprehensive reply to Bradley (whether we agree or not) is very much appreciated.

Thank you and best wishes.
Taylor

[hoops777]

Larry,I thank you as well.You went far beyond the call of duty

[larryswedroe]

Taylor/hoops
My pleasure, having done the research for my books it's not that difficult to dig up the sources.
Best wishes
Larry

[LazyNihilist]

I find Bonds very difficult to understand compared to Stocks.

I just put 75% in Total Bond Market and 25% in TIPS, for the Bond portion of my portfolio. (with some IBonds as emergency funds).
I hope this is right.

[Taylor Larimore]

I find Bonds very difficult to understand compared to Stocks.

I was once a Director of the Miami Housing Authority which issues bonds. We had to hire a housing bond specialist because no single bond expert fully understands all kinds of bonds.

Understanding Bond Documents

Knowing what he/she doesn't know is the mark of a good investor.

Congratulations and best wishes.
Taylor