Below is a portfolio for you to look at, think about and hopefully weigh in on. I don't have this mix for any client so this is simply an academic study to try to learn from.
The portfolio;
iShares TIP Bond Fund (TIP) 25%
iShares iBoxx Investment Grade Corp Bond Fund (LQD) 20%
iShares Barclays Intermediate Credit Bond Fund (CIU) 20%
iShares JPMorgan USD Emerging Markets Bond Fund (EMB) 15%
Currency Shares Australia Trust (FXA) 20%
TIP and FXA are client holdings and obviously FXA is not from iShares and not a bond fund. Can a currency fund now yielding close to 4% be considered a bond proxy? The chart compares all five funds and the S&P 500. You can click to enlarge it or even better you can plug the symbols into whatever website you use for charts to take a better look.
Above a certain age heavy exposure to TIPS probably doesn't make sense. 'Certain age' is subjective but TIPS for someone in their 80s is probably unnecessary. For now there is very little inflation around and a compelling case for deflation but TIP yields 2.77% and the duration is only 4.08 years which is not too shabby.
LQD and CIU might appear to be the same fund by the description but under the hood they are very different. LQD has a yield of 4.80% ans duration of 7.17 years compared to 3.61% and 4.22 years for CIU. Another big difference is that CIU owns a lot of debt from foreign companies and even some sovereigns that are denominated in US dollars which means there are quite a few things you've likely never heard of. LQD owns a lot of things you have heard of. CIU is heaviest, sector wise, in industrials at 39% followed by financials at 30%. LQD is heaviest in in financials at 33% with the rest of the sectors being much smaller.
During the worst of it for bond funds in the immediate aftermath of the Lehman-weekend LQD did a little worse dropping 20% versus 15% for CIU.
EMB not surprisingly did worse still dropping 35% at its worst but has come back to where it was before then. The bonds in EMB are dollar denominated which has been a positive lately but will be a drawback in terms of opportunity cost if the dollar goes back down. The country exposure is spread around pretty well with a couple of countries around 9% and getting smaller from there. The fund yields 6.24% with a duration of 7 years which is a little long for where yields are these days.
FXA went down a lot before, about 35%, and is down now at $84 down from $94 and the dividend as been moving up as the RBA has been slowly increasing rates.
The yield of the overall mix is a little over 4% but of course that requires a grain of salt. ETF yields fluctuate, which is one of the drawbacks. The duration works out to a little over four years, assuming FXA at zero, which I think is a good overall number for a portfolio in the current environment. I did not intend to chase yield although maybe you would conclude I did.
Some questions I would ask is that for anyone interested in using ETFs exclusively for the fixed income portion of their portfolio. You can see from the chart what they did during one of the worst stretches for financial markets in modern history in terms of panics. One thing that is unknown is what these funds might do in a prolonged period where rates rise. Short dated paper is less sensitive to this but because bond funds have no par value to return to there can be a different dynamic which could hurt investors who are not on the lookout for this.
We do know that a short dated fund will move less than a long dated fund but the move in a short dated fund, depending on how things played out could be worse than the 15% or 20% that LQD and CIU dropped. They obviously snapped back fairly quickly but what was the panic level, if any, for holders as they were spiking down? What would you do if a bond fund you owned with no par value to return to dropped that much? Anecdotally it seems like during the panic people leave comments expressing real fear but after the fact the comments tend to lean toward 'no problem, I knew they'd come back.'
Another thing also is that FXA, relative to bonds, has been a volatile hold and clearly not right for plenty of people even if it makes for a good talking point. I think foreign exposure is important but wanted to add something a little more interesting than the sovereign debt funds that iShares has.
Your turn. What do you find interesting here, what would you do differently? Hopefully this can create a useful dialogue.
14 comments:
Other than FXA, we use the others in our portfolios as well. The lowered relative volatility of CIU vs. LQD have made it slightly more attractive as of late and is taking a higher % of our allocations. Thanks for the review.
I use bond ETFs exclusively for the fixed income part of my portfolio. At this time I am not comfortable enough with my knowledge to try to evaluate bonds on their own.
The Junk "bond" etfs (HYG and JNK) behaved nearly identical to the S&P500 for the past 2 years. For this reason I categorize them as high yield dividend stocks. Obviously they wont work exactly like them but thats the closest fit.
Continuing the diversification topic, TLT has a greater inverse relationship then BLV during the 2008 scary times.
I was going to suggest TLT also
Perhaps form a 4x25 like Permanent Portfolio blend of bonds
TLT (long duration)
EMB (stock like)
TIP (cash like)
FXA (gold like)
"Short dated paper is less sensitive to this but because bond funds have no par value to return to there can be a different dynamic which could hurt investors who are not on the lookout for this."
I see this sort of statement frequently and it implies that bond funds are somehow inferior to individual bonds. I challenge you to show this is true in a rigorous mathematical proof. Assume all interest from mutual funds is reinvested as rates rise. I think you'll find your statement to be untrue.
Also, due to the steepness of the yield curve, it may be that shorter duration bonds will be impacted more by rising rates, not less. This is a case where conventional wisdom may be wrong.
Rhianna32's observation on the correlation between junk bonds and stocks is right on.
Clive that is a very interesting mix. The thing with TLT is the extent to which it greatly increases the duration of the mix.
WH short dated individual issues being less sensitive is a mathematical thing. a with only three more coupons to go, for example, will not dramatically fall away from its par value (due to interest rate risk) because in that case the YTM would go through the roof which the market will not allow to happen. Maybe mechanical is a better word than mathematical.
As for you other point/question I would tell you to disprove it if you are so inclined.
Interestingly this would seem to disprove Karl Popper's idea of it only taking one negative to disprove a theory because I am sure there would be examples to disprove both sides of the point.
Hi Roger
"The thing with TLT is the extent to which it greatly increases the duration of the mix."
Is that good or bad thing?
Potentially the increase in duration instils more volatility against which periodic rebalancing can capture some gains
http://tinyurl.com/2waeg77
Between Jan 2008 and 2010 the straight four averaged a combined capital value -0.5% decline when ran as-is. Yearly rebalancing back to equal weightings uplifted the capital value of the set to around a +3.7% gain.
good or bad depends, i guess.
I view that part of the curve as very expensive and so I am not a buyer even if there is a good trade there.
I use funds for bonds - PTTDX, LSBDX and TGBAX. Somehow the bond market still seems somewhat opaque and individual issues are definitely out of my league. Somehow ETFs seem more approachable for stocks.
Isn't it logarithmic not linear?
US 30 year treasuries 4%, Japan 30 year treasuries 2%. If we enter a prolonged deflation type period such as Japan's and 30 year treasuries decline to 2% then the capital value of those treasuries will double and you've twice the average yield coming in.
Should instead 30 year treasuries rise to 8% then your income is half the average and you'd be sitting on a 50% capital loss.
With the whole world all deflating in near lock-step the risk of inflation would appear somewhat low.
Holding a blend of both deflationary and inflationary cover is the more neutral overall approach.
Interesting topic. A suggestion I would make is to add something to the portfolio that went up during the 2008 crash.
Here is a conservative portfolio that I use for money I need fairly soon.
45% Long Bonds
30% T-Bills
25% Commodities
http://www.riskcog.com/portfolio-theme2.jsp#58jfcsc749
The bond-bill barbell provides income, and the commodities hedge the bonds. I believe this portfolio doesn't have as high of current income as the other portfolio you presented, but the total return will probably be higher with smaller drawdowns.
I like Rhianni32's idea about high yield; near the bottom of the crash I bought some *high* yield closed end funds and SDS. The scheme was to lock in the high yields and stay market neutral if things crashed further or recovered.
I, too, prefer mutual funds for my bond holdings. Fees exceed those of etfs, but professional, active management is well worth the cost.
This is an interestingm post, Roger. Thanks.
I LOVE YOUR BLOG!
Keep up the great work!!
Common Cents
http://www.commoncts.blogspot.com
ps. Link Exchange??
Sorry - I can't get excited about bond ETFs or more specifically bond indexing. I'll pay a little more to have Gross and friends pick bonds for me.
I've dumped all my bond funds except for a small amount in int'l bonds and some TIPS. I have replaced the entire bond side of my portfolio (30% less the 3% comprising the above mentioned funds) with Stable Value Funds inside our 401(K)'s. WHEN interest rates start to rise, their yields will increase. For now, I'll accept the 3.5% or so yield. loating rate products are also interesting for those of us thinking along the lines that interest rates have nowhere to go but up.
Might I leave some return on the table? Yes, but the PIMCO Total Return fund returned better than 15% last year and I can't see that happening again this year soooo...thanks for the return but you're outta here!
DE
Post a Comment