In my previous post, I looked at the uneasiness many investors have about bonds when interest rates are poised to go up. Some readers have argued that this discomfort can be alleviated by buying individual bonds rather than index funds or ETFs. Let’s look at whether their arguments hold up.
Individual bonds do offer benefits: you know precisely how much interest you’ll be paid, and how much you’ll receive when the bond matures. The payouts from bond funds, by contrast, aren’t known in advance, and funds never mature. This can make planning difficult for those who rely on their bond portfolio for current income, or those who need a specified amount of money on a certain date in the future. No quibbles with that.
The other main argument in favour of individual bonds is much less convincing. It goes something like this: “If I invest $10,000 in a bond fund, its value will go down when interest rates rise. But if I buy an individual bond, I don’t have to worry about interest rate movements, because as long as I hold the bond to maturity, my principal is guaranteed.”
Behavioural economists love this kind of logic. While it’s true that holding a bond (or a GIC) to maturity allows you to collect the full principal, you’re fooling yourself if you think you’ve avoided a loss. Consider a bond that has two years left to maturity and pays 4% interest when the current rate for comparable bonds is 5%. Over the next two years you’re locked into an investment earning 4% interest, when you could be earning 5% with no additional risk. You may not have lost your capital, but you have forfeited $200 in interest. This opportunity cost may not feel the same as losing capital, but in the long run it affects your investment returns the same as any other kind of loss.
There are several other advantages of using index funds or ETFs rather than individual bonds:
Diversification. Canadian government bonds are highly unlikely to default, but corporate bonds certainly can. Just ask the investors who put their retirement savings in bonds from General Motors. You can dramatically reduce default risk by investing in an ETF like iShares’ Canadian Corporate Bond Index Fund (XCB), which holds 350 different issues. A wide selection of bonds with various maturities can also reduce interest rate risk.
Low minimum investment. Many bond issues require $10,000 or more, so building a customized bond ladder is out of the reach of many small investors. By contrast, Claymore’s 1–5 Year Laddered Government Bond ETF (CLF) includes a ready-made ladder of 25 bonds, and small investors can buy in for any amount. Bond index funds also have very low minimums ($100 to $1,000) and you can take advantage of dollar-cost averaging by adding money each month, something you can’t do with individual bonds.
Liquidity. You can redeem all or part of a bond fund at any time if you need the cash, typically with no fee (except perhaps the small trading commission on an ETF). While you can always sell individual bonds on the secondary market, many corporate, provincial and municipal bonds are not highly liquid and may have high bid-ask spreads.
Simplicity. Index funds and ETFs eliminate the need to choose individual bond issues, some of which include embedded options that are difficult to understand. Funds also conveniently reinvest all interest payments so you can take full advantage of compounding. (This is harder with ETFs, but Claymore and BMO do offer dividend reinvestment plans.)
Cost. Some investors say they prefer to avoid fund fees, seemingly unaware that buying individual bonds also carries costs. It’s difficult to know just how much mark-up you’re paying, as the bond market is far less transparent than the stock market, since bonds trade over-the-counter rather than on an exchange.
In his book In Your Best Interest, Hank Cunningham says the average markup on bonds is 1%, and points out that this fee is payable only once, while bond funds may carry fees of 1.5% or 2% every year. But Couch Potatoes know better than to pay those ridiculous MERs. Bond index funds and ETFs are among the cheapest in any asset class, with MERs as low as 0.10%. With fees this low, any cost benefit of buying individual bonds shrinks dramatically, and many investors will accept the small fee in exchange for the benefits outlined above.
Like all investment decisions, there’s no one right answer here. Building a ladder of individual bonds is a perfectly good strategy, particularly for income-oriented investors with a six-figure sum to invest, a good knowledge of bond features and access to a low-cost broker. But for Couch Potato investors who are in the accumulation phase of their lives, a low-cost bond index fund or ETF can deliver market returns far more easily.
Great posts on Bonds vs ETFs. On a related note, can you please comment on whether there are any advantages from holding preferred shares (eg: banks) rather than bonds, in a rising interest rate environment?
Thanks for the article, Dan. I still struggle with the decision but this was useful for me. Re: the performance of bond funds, I would just stress that when using free info providers such as Yahoo Finance or Google finance, it is important to use the Adjusted Price, not the closing price, of the security in question. Note that the default price used in most historical performance graphs (that I’ve seen anyway) is the closing price. If you compare the Adjusted Price of XBB (which takes into account dividends and splits), and compare it to that of, say, XIC, you will find that the two are not that far off over many conventional time periods (e.g. 1 year, 3 year, 5 year, etc.) in terms of returns.
I have been spending time trying to understand how a bond ETF like XSB will respond with rising interest rates and found the discussion at Bogleheads very good and your readers may wish to check it out:
The summary takeway from this is that if you hold the bond ETF to the average maturity time then it should have a total return equal to the average yield to maturity at the time of your purchase and so it should act just like the portfolio of individual bonds that you would assemble yourself as a substitute to the bond ETF. In the case of the ETF XSB the weighted average duration is 2.65 years so if you hold the ETF for this length of time and reinvest the distributions as you go the you should end up with the weighted average yield to maturity which is 2.35%. Now, the distributions are providing roughly 3.5% annual return which is higher than your planned (expected) yield when you purchased so this is the fund giving you some of your own money back so it should not be a surprise if the price of the fund drops accordingly.
By thinking of the ETF this way it provides some confort that even though the price is falling I still get what I planned when the purchase was first considered.
LRM: Thanks for linking to the Bogleheads article, which is a great resource.
This is my first time visiting your blog. I really like it. This was a thoughtfully written post, and your readers contribute nicely to it with their intelligent comments. Keep up the great work. You seem to have a smart following.
Thanks, Andrew, and welcome to the blog!
Great site. Thank you.
I’ve been racking my brain on this and am hoping you can help.
Let’s say I am trying to decide between 2 bond funds. One is a ladder with 5 year duration and one is a straight bond fund also with 5 year duration. How would the sensitivities of the fund NAVs differ to changes in interest rates. Would the two funds be identical if the entire interest rate curve changed identically?
Bif: In theory, two bond funds with the same average duration would behave very similarly to a change in interest rates. For example, Claymore’s CLF (laddered) has an average duration of 2.4, while iShares XSB (not laddered) has an average duration of 2.8, so they should move together pretty closely.
However, for a laddered fund to have a five-year duration, the buckets would be something like three to seven years. Does such a product exist? Are you comparing real funds or just wondering about this in the abstract?
Thanks for the response – actually I was thinking of the two funds you mentioned, but i wrote 5 years for no particular reason.
In that case, then, yes, these two ETFs should perform similarly. Note, however, that CLF is only government bonds, while XSB has corporates too, so there will be some differences for that reason.
I am curious if “opportunity cost” takes into account the losses in the bond fund that occur when interest rates rise? The funds will eventually replace bonds as they mature with higher yielding ones but, the investor whom initially purchased an individual bond is guaranteed to start earning the cupon rate of interest immediately and a guaranteed return of principal on maturity. Seems like a good short-term strategy to buy individual bonds with 1-2 year maturities and hold them until they mature. When interest rates and yields have increased buy bond funds like XBB.
@Johnny: You could buy short-term bonds now and wait for interest rates to rise, and many investors have been doing just that. However, over the last couple of years, XBB has dramatically outperformed short-term bonds: that is the opportunity cost. There’s no easy answer here, because no one knows when rates will rise: they could stay low for a long time, and maybe even dip lower.
Do you know a long term bond fund with low MER and buy/sell spread? I ask because on the short end, I can get better rates with GICs and savings accounts. I just sold my house in Toronto so for the first time I have enough cash to warrant this kind of micro management.
@Michelle: There are only three long-term bond ETFs in Canada:
http://ca.ishares.com/product_info/fund/overview/XLB.htm (gov’t and corporate)
http://www.etfs.bmo.com/bmo-etfs/glance?fundId=80004 (gov’t only)
http://www.etfs.bmo.com/bmo-etfs/glance?fundId=75747 (corporate only)
Just be aware that long-term bonds are the most sensitive to rising interest rates.
Hi Dan- thx for the post & to all the comments.
Like the fellow who just sold his house I’m trying to figure out what to do with a larger sum I took out of something else. My 1st goal is to preserve capital & then to stay even with inflation[before taxes]. I’m doing that with some, where I have it in a 5 yr ladder of GICs averaging 2.75%. I also have some in XLF & CPD ETFs. I’ve looked at XRB yet see it’s near an all time high at $25.89. I don’t know how to asses an ETF like this which holds RRBs as to the risk to my capital of putting it there? How should I evaluate that? And what other options are there.
I should say I’m retired, and this is Non reg money which can stay put for 25- 35 years.
@BradTS: RRBs are an excellent portfolio diversifier for the long term, and if you really do have a 25-35 year horizon, they certainly will provide inflation protection. However, like all long-term bonds, they can be fairly volatile. As you noted, they are expensive now (i.e. yields are very low) and if long-term rates rise, their value will fall. So they’re definitely not a short-term investment. I would also be aware of the tax issues: if you’re in a high tax bracket, then there isn’t going to be a lot left for you. GICs tend to be a better choice in registered accounts.
You may want to have a look at some of these options:
Thanks for the link…I know you mention tax issues, however the ETFs in that link are equities & I’m not comfortable putting my Core savings in equities & “riding that wave”. That’s fine for a small part of my holdings but not for my Core savings…at least not for me in these times we’re in.
I just want something where I firstly: preserve my capital & then…if I can earn a return BEFORE tax of around the rate of inflation… like with my GIC ladder, I’m happy.
I was just wondering if there’s a way to preserve my capital AND do a little better than I am with my 5 yr ladder of GICs averaging 2.75%? without any direct investment risk to the capital invested?
I accept the slight tax issue as it’s definable. Also I figure that if inflation goes up my average GIC return will go up as I roll a GIC over each year.
Thanks again for your thoughts.
@BradTS: Actually I’m a bit surprised that your GICs are averaging 2.75%. You must have bought them several years ago!
As you know, there really is no way to improve your returns without taking more risk. But you might consider something like a corporate bond ladder that goes out 10 years. As you know, GICs only go out to five years, so extending the maturity might earn you a little more yield.
If you stick to investment grade bonds, the risks are quite low, and the yield to maturity should be right around 2.6% or 2.7% for a one- to 10-year ladder. (If you use only bonds with 5 to 10 years to maturity, you can get a little more.) iShares even has an ETF that builds the ladder for you:
Another option is RBC’s target maturity corporate bond ETFs, which go out as far as 2020. Their website even includes an explanation of how to build a bond ladder using these ETFs:
What do you think of First Asset DEX 1-5 Year Laddered Government Strip Bond Index EFT? Is there a spot for it in a Couch Potato Portfolio?
@Jonathan: Please see this post:
Dan, I understand the opportunity cost example you gave in the post, and the need to hold a bond fund for the duration of the fund in order to, worst case, get your capital back, but the following example has me stumped.
If I buy $1000 of XBB today it has a duration of 6.76 years, so worst case, I would get back my $1000 in 6.76 years. If I buy a 7 year government bond at par today (assuming I could find one) it is currently paying 2.14%. Worst case scenario, I could hold this bond for 7 years then get my $1000 back plus the interest I would have received each year of 2.14%. In this instance, doesn’t that make holding an individual bond better than a bond fund?
@Tristan: I think you’re confusing “term to maturity” with duration. A bond’s (or fund’s) duration takes into account both the term to maturity and the coupon, so the duration is always less than the term.
In the case of XBB, its average term to maturity is about 10 years and its duration is only about 7 years. So a fairer comparison would be to a 10-year bond with a similar coupon rate to XBB, not a 7-year bond with a much lower coupon (not sure where the 2.14% figure comes from).
It’s also important to remember that bond index funds are designed to maintain the same duration indefinitely. So next year, XBB will still have an average term of 10 years and (assuming interest rates don’t change much) a duration of about 7. Same thing two years and three years from now. Bond funds never mature, and that makes them inappropriate if you have a known liability on a future date. Individual bonds are more predictable in that sense, but that doesn’t necessarily make them “better.”
Thanks. I think I understand the difference (maybe not!) between duration and term to maturity, and that bond funds never mature, and that individual bonds are more predictable if you have a known future liability. What I mean is, say, 7 years ago I had had a known liability of $10,000, so I bought a 7 year bond at par (the interest rate doesn’t matter). Today the bond matures, I get the $10,000 back plus whatever interest has accrued over the 7 years. Also, 7 years ago I bought $10,000 of XBB which at the time, for arguments sake, had a duration of 7 years. Because of rising interest rates over the last seven years, the worst case scenario has occurred and today my XBB, with dividends reinvested is only worth $10,000 – i.e. I get my capital back only. In this case then, holding an individual bond was better, in terms of return, than the bond fund. I guess the answer to my question then is that returns of comparable bonds funds and individual bonds are expected to be the same, but there may be times of large interest rate changes at particular times that could lead to one performing better than the other?
@Tristan: The key misunderstanding is that your 7-year bond does not have a duration of 7 years: its duration will be less than that (exactly how much less depends on its coupon rate). So the comparison is not fair. Yes, in your example the 7-year bond has less interest-rate risk than XBB, but that’s because its duration is lower, not because it’s an individual bond.
OK, I get it now. Thanks! So to make a fair comparison you’d need to rebalance the duration of the bond fund each year with a shorter term bond fund/cash to match the time left of the term of the bond, i.e. when you needed the money. Then you would be expect the bond and the bond fund to perform the same.