In my latest podcast, I answer a series of frequently asked questions about bonds. The second of these came from a reader named Andrew: “I have been investing using your Couch Potato strategy for just over three years now,” he wrote. “However, does it still make sense to invest in bonds when they are continually losing money?”
As it happens, bond ETFs have not been “continually losing money” at all. Indeed, over the three years ending March 31, broad-based funds such as the BMO Aggregate Bond Index ETF (ZAG) and the Vanguard Canadian Aggregate Bond Index ETF (VAB) returned close to 4% annually, with positive returns in each calendar year. A $1,000 investment in either ETF would have grown to about $1,120 over that period. So why would an investor think he had lost money?
I don’t blame Andrew for being confused, as this one trips up a lot of investors. The problem lies in the way brokerages display the holdings in your account. Rather than calculating the total return on your investments—which would include both price changes and all interest payments and dividends—your list of holdings reflects only the change in market price. This makes sense for calculating capital gains and losses, but it can be highly misleading for investors who want to measure performance.
Let’s break this down to understand what’s happening.
Half the story
Say you bought 500 shares of ZAG about two years ago, on March 31, 2015. On that date the ETF was trading at $16.37 per share, so your shares cost you $8,185. Two years later, on March 31, 2017, ZAG was trading at $15.70, so your 500 shares were now worth $7,850, a decline of about 4%. When you log into your account, your holdings will look something like this:
Symbol | Quantity | Book Value | Market Value | Change ($) | Change (%) |
ZAG | 500 | $8,185.00 | $7,850.00 | -$335.00 | -4.09% |
At this point you’re cursing your decision to buy bonds, as this supposedly safe part of your portfolio has lost you $335. Right?
Wrong. The problem here is you’re ignoring all of the interest payments ZAG made over the last two years. It turns out those cash distributions amounted to about $0.93 per share, which more than offset the decline of $0.67 per share. That made the difference between a loss and a gain on your investment. You started with $8,185, and after two years you had 500 shares worth $7,850 plus $465 in interest, for a total of $8,315. That modest gain is hardly cause for celebration, but it’s certainly not a 4% loss.
Unfortunately, you may not have noticed this because the interest payments are paid into the cash balance of the account. At some point you probably reinvested that cash when buying new shares of some other ETF. But if you calculate your rate of return properly, using the total value of your account at the beginning and the end of the period, you’ll see a positive return.
Why this happens
As discussed in the previous post in this series, bond prices fall when interest rates rise. But even during periods when rates stay more or less the same, you will still see the price of most ETFs decline gradually, even over long periods.
This happens because most bonds today are premium bonds, which means they trade at more than their face value. This situation has come about because interest rates have trended downward for many years now, so most bonds issued in the past have coupons higher than prevailing rates. Investors pay more than face value to get those higher rates, but premium bonds will eventually mature at face value, resulting in a capital loss.
If your ETF is filled with premium bonds—at it almost certainly is—it will experience a series of small losses like this as the bonds approach maturity. That translates into a gradual drop in the ETF’s price during any period where interest rates do not fall significantly.
One way to anticipate this price decline is to look at two characteristics of your bond ETF, which you can find on its web page: the fund’s average coupon and its yield to maturity. The former tells you roughly how much you can expect in interest payments, while the latter estimates your total return, including interest payments and any price change.
If the coupon is higher than the yield to maturity—and again, these days it almost always is—then the fund is dominated by premium bonds. Today the average coupon on ZAG is about 3.35%, while its yield to maturity is 1.91%:
That means if interest rates don’t change, you should expect the price of this fund to decline by roughly 1.44% (that’s 3.35% minus 1.91%) a year. It will never be that tidy because interest rates change constantly, but the key point is that any bond ETF filled with high-coupon premium bonds should be expected to fall in price over time.
Where to get accurate numbers
If you want to know how your bond fund has performed in the past, the best method is to look it up on the ETF provider’s website. ZAG’s webpage, for example, reports the following total returns for the period ending March 31, 2017:
One important note: whenever ETFs report their returns, they assume all cash distributions are reinvested immediately. With an ETF this is impossible, even if you are using a dividend reinvested plan, because some portion of the interest or dividends will always end up as part of your cash balance. So your personal rate of return will never be precisely what’s reported on the provider’s website. But it will be close enough. And at the very least, you will no longer believe you’re losing money with your bond ETF during years when you’re actually netting a gain.
Hello,
Question regarding ZAG – as it is a “fund of funds” (Aggregate Bond Index) – does this “hide” management fees? When the funds states a MER of 0.14% – do we need to add to this the underlying MER’s of the underlying ETF’s that ZAG is holding, thus drastically increasing the “true” MER of this ETF?
I have not been able to find clear answer to this question –
Thanks!
@Sebatien: No, there is no double-dipping on fees when ETFs hold other ETFs:
https://www.theglobeandmail.com/globe-investor/investment-ideas/is-my-etf-double-dipping-on-fund-fees/article4182696/
If the value of a bond fund is dependent on the ratio of “premium” bonds, then how can we compare the performance of each fund, and the future expectations without knowing the relative makeup of premium bonds?
I.e., two funds could have the same price history or same price today, but if Fund A is achieving that performance with lower rate bonds, while Fund B has a higher portion of premium bonds, then is it correct to prefer Fund A (looking at history)? Or conversely, is it better to prefer Fund B because it has a leg up on the potential future performance?
@Michael F: The most reliable way to compare the performance of two bond funds is to simply look at the performance figures published on their websites: trying to calculate it on your own (factoring in interest payments and price changes) is tricky. The published number is always a “total return,” which factors in both interest payments and price changes. For example, if the fund suffers a –2% loss but pays a 3% coupon, its total return will be reported as 1%. If there is a significant difference in the performance of two bond funds, it will not be because one has more or fewer premium bonds: something else would be causing that difference (long vs. short maturity, higher vs. lower credit quality, etc.).
In terms of expectations going forward, the most useful metric is the fund’s “yield to maturity.” If two funds have the same yield to maturity, then they should have roughly the same expected return, regardless of the amount of premium or discount bonds. The YTM factors in both the coupon (amount of interest paid) and the expected capital loss/gain expected upon maturity.
If a fund’s average coupon is much higher than its YTM, then the losses you will see on your statements will look larger, because that fund is full of premium bonds. If the average coupon and YTM are similar, then the losses will be more modest, as the bonds in the fund are trading closer to par.
Hope this makes sense.
I hold ZAG and as of the end of September 2018, the BMO website shows annualized performance of 1 year at 1.5% and 2 year at negative -0.84%.
https://www.bmo.com/gam/ca/investor/products/etfs#fundUrl=%2FfundProfile%2FZAG%23price%26performance
Can you explain why there is a negative performance over 2 years? With interest rates rising, should I switch out of ZAG and into GICs?
@Jen @Dan, this is a question that I also have. The latest (Dec 2018) 1yr and 3yr performance figures for ZAG and ZDB are both below the 2018 rate of inflation which was on average 1.7%. Not good news. On the flip side, I’m considering putting my fixed income allocation into a high interest savings account that returns 3.1%. Yes, I’ll lose the negative correlation benefit of bonds versus stocks, but to be honest over the past year I saw more of a positive correlation between ZAG/ZDB compared to stocks than any negative one.
So, with the availability of a 3.1% high interest savings is there any compelling argument to continue using ZAG/ZDB?
@Andrew: You can certainly make a strong argument for holding cash rather than bonds if you are able to get a higher interest rate that the YTM on the bond fund. In many cases, though, these rates (3.1%) are teaser rates and only available for limited periods. If interest rates were to fall, that offer would almost surely disappear, whereas bond values would go up. But as long you understand this, it’s a perfectly reasonable choice.
@Dan Many thanks for the quick feedback. In this case the rate isn’t a teaser rate but a 1 year term rate offered by Hubert Financial (https://www.happysavings.ca/products/terms/one-year-terms/). I’d have to leave the cash with them for one year to get the full 3.1% but this is something I’d do if I used a bond ETF anyway, after all we’re investing for the long term. I can re-evaluate in a years time and switch to bonds if things have markedly changed. Still seem reasonable?
Hi @Dan, great posting. Does this mean that my returns in my RBC brokerage account for equity ETF’s also do not take into account dividend returns, whether reinvested or distributed as cash?
Thanks,
Ivan
@Ivan: Yes, that’s correct. The gains/losses you see on most brokerage statements are price changes only. They are not “total returns,” as they do not account for dividends and other distributions. This is even true if you are using a DRIP.
Would bond index funds behave in a simliar way the as the etf funds? When I look at my cibc short term bond index premium fund it shows as a capital loss, so of course I am assuming that I have lost money.
@Cameron: Yes, the principle is the some with bond mutual funds. That said, bond returns really have been negative so far in 2021, so it’s possible that over the period you’re looking it there have indeed been losses.