By Dan Hallett, CFA, CFP
It’s easy to be seduced by what appears to be a proverbial ‘free lunch’ in the investment world. Nowhere is this more prevalent than in the world of fixed income (i.e. bonds and bond-like investments). There are many potential topics to tackle in this context. But perhaps the most common misunderstanding I notice among investors emerges when equating monthly distributions with true yield.
YTM is what counts
Consider the case of short-term bond funds and the iShares CDN Short-Term Bond Index ETF (XSB/TSX). Its current interest distributions are roughly equal to an annualized 3.3%. While I have no real proof, I firmly believe that many investors equate this 3.3% distribution rate with XSB’s ‘yield’. The websites for iShares and other ETFs provide a lot of great information.
In the context of this topic, the yield-to-maturity (YTM) is the most meaningful piece of data on XSB’s profile page. The YTM was recently listed at 1.89% per year. The fund’s 3.3% distribution rate is real but it’s not the whole picture. A bond’s periodic coupon interest payments form part of its total return. The other part is the difference between the purchase price and the maturity or par value. And with interest rates (and bond yields) at historic lows, most bonds will fall in price between today and maturity.
XSB’s 3.3% interest distribution rate excludes the guaranteed capital loss between now and maturity on virtually every bond. The YTM counts both the coupon interest and the price change through to maturity. And that’s the figure you should care about since today’s YTM is a very good indication of the bond’s future return if held to maturity.
Note that the quoted YTM is calculated before deducting XSB’s 0.26% management expense ratio (MER). So the net YTM can be estimated by subtracting the MER from the quoted YTM. In XSB’s case the net yield is about 1.63% annually – less than half of the net interest distribution payments. A more striking example is Claymore’s 1-5 Yr Laddered Corporate Bond ETF (CBO/TSX). CBO’s 2% net YTM is less than half of its 4.6% distribution rate.
Yield-to-worst is better
At least with ETFs, fees are relatively low and YTM data is readily available online. This information isn’t nearly as accessible with short-term bond mutual funds. While a few firms like RBC/PH&N regularly update YTM figures, most do not publicly disclose this data. But mutual fund companies should be able to provide this information over the phone. For funds and ETFs investing in corporate bonds, high yield bonds and preferred shares, it’s ideal to obtain a figure similar to YTM called yield-to-worst (YTW) instead of bonds’ YTM or preferred shares’ current yield.
The next time you feel compelled to invest in a fund because of its unusually high distribution rate, understand that you’re either overestimating the fund’s true “yield” or its true risk exposure. Hopefully, this awareness will trigger enough questions such that you’ll be better informed prior to investing.
Related: The relevance of YTM & the impact of rising rates (July 2011)
[...] former blogger himself, Think Dividends sent me an article by Dan Hallet on this very topic: “Distribution rate does not equal yield“. I want to thank him for that article, reinforcing my conversation with Som [...]
Thank you for a good write-up on something that is not well understood.
As I understand it, I believe there are a couple of inaccuracies however.
1) re: “And with interest rates (and bond yields) at historic lows, most bonds will fall in price between today and maturity”.
We have been in an historic low interest rate environment for sometime (~5 years?). Interest rates and bond yields have not moved that much, so I don’t think most short term bond purchased in the last 1 – 5 years have fallen substantially.
If you are referring going forward from here – although inflation has modestly reared it’s head just lately, recent global economic growth risks will likely keep interest rates from materially rising, so it is not a foregone conclusion most short term bonds will fall in price – especially short term.
2) re: “XSB’s 3.3% interest distribution rate excludes the guaranteed capital loss between now and maturity on virtually every bond”
What “guaranteed capital loss” are you referring to? The price risk to maturity? To add to my points above, I am told an ETF such as the XSB rarely holds a bond to maturity – they will sell when the they market conditions are favorable (e.g. a particular bond is trading above it’s purchase price).
Regardless, would it not be better to characterize the YTM risk as “a variable risk not factored into the quoted current distribution rate (yield)”?
In that context, as you say, for a bond ETF, the YTM really represents the worst case.
If the XSB is yielding 1.86% worst-case, that is a good liquid yield in this environment – with potential upside.
I have had a small position in the XSB for about 8 months. I am gotten approximately 0.27% per month = 3.3% per year (that is after the MER). It has a consistent 1, 3, 5, 10 year track record.
As you point out, with historically low interest rates, at some point (nobody knows when) the bond prices will likely fall, but even then, the worst case is a decent yield (1.86%). Interest rates would have rise significantly, fast, and consistently for the worst case to materialize.
Thanks for stopping by Curtis and for taking the time to write such a thoughtful note. I’ll try to address all of your points, in order.
1. I mention that rates/yields are historically low not to make some future forecast but to support the notion that most bonds trade at a premium. For example, consider a bond that is issued at $100 with a 5% that matures at $100 in five years. Two years later (i.e. three years before maturity), yields on three year bonds from the same issuer are now 3%.
This will cause the original bond’s price to rise in price, say to something in the range of $105 (per $100 of maturity value). If you buy the bond at $105 and hold it to maturity, you will still get the 5% coupon but the fall in price between purchase (@ $105) and maturity (@ $100) will reduce that 5% coupon to equal a lower yield to maturity (3% in this hypothetical).
2. The guarantee capital loss I’m talking about is somewhat explained in my #1 above. If yields fall after a bond is issued – as has happened with most bonds trading today – its price will exceed the maturity value. So buying it at a price that exceeds the maturity or par value will guarantee a capital loss if held to maturity. It’s true that XSB tracks the DEX Short-Term Bond Index and, accordingly, it doesn’t just buy and hold its bonds. But XSB can’t escape the gravitational force that the maturity date exerts on the market price of the bond. And exchanging one premium bond for another doesn’t help.
You’re right that we don’t know exactly what the ultimate YTM will be if you don’t hold until maturity but the only variable is the magnitude not the direction. Still, the YTM is a good forward looking figure unlike simply looking at past returns which were achieved with higher yields and, importantly, with the price boost from falling yields. That YTM has fallen over the past few years tells you that the future return potential is smaller – certainly smaller than the pure coupon/interest portion that you see every month.
It’s important to note that the YTM is not a worst case scenario. Rather, it’s a very good estimate of a bond’s future return over the term of a bond. And that extends to a portfolio of bonds when the YTM is calculated on a weighted average basis for a bond portfolio.
I can’t stress this enough Curtis. If you’re looking at the 0.27% monthly or 3.3% yearly, you’re not getting the total yield picture. If you don’t believe me, you will if you hold XSB for another year or two (beyond the eight months you’ve held it). And I’m not saying that XSB is a bad vehicle. Just that investors must understand exactly what they’re getting into.
I understand how bonds work, but again I do not understand how it is *guaranteed* that the price at maturity will be lower than the purchase price for all the bonds in their portfolio – it all demands on the individual bond (when it was bought and what happens to bond yields after).
In the example you cite, what if the Bond ETF fund bought the 5% bond *when it was issued* at $100, and sold it 2 years later when it was worth $105? There is no loss, but a profit. Am I missing something?
I *think* what you are getting at, is in the last 5 years bond yields have mostly dropped, meaning that most bonds bought on the open market, after issue, are in a loss position? But even then:
1) What about the bonds that the fund bought at *issue* during those 5 years? Since bond yields have dropped, shouldn’t they have been worth more on the market and the fund had the opportunity to sell at a profit?
2) Recently (for the second straight month) bond yields have fallen, reaching their lowest levels of the year – wouldn’t bonds that the ETF bought earlier this year be in a profit position?
I am not trying to be a pain in the neck
, I am just trying to understand the ‘guaranteed loss’ statement.
You are right, that YTM is not the worst case scenario since it does not factor in any losses (or gains). I assume the NAV per unit will reflect that though?
re: “If you’re looking at the 0.27% monthly or 3.3% yearly, you’re not getting the total yield picture”. That is the yield *I have received* so far. If I sold my units today I would have a -0.9% capital loss on the units, and a gain of 2.16% from the distributions (over 8 months).
I fully realize that this could change going forward.
In the case of the XSB, the 1, 3, and 5 year total returns have been 4.10%, 4.66%, and 4.78% respectively (http://ca.ishares.com/product_info/fund/performance/XSB.htm). If all of their short term bonds were in loss positions, I am wondering how they maintained those returns in a falling bond yield environment? The NAV has stayed within a fairly tight range during that time…
Not a pain at all Curtis. You may already know some of this but bear with me as I go through this for others who may be following the discussion.
Rather than batting around some of these finer details, let’s focus on yield-to-maturity. Simplistically, YTM is the current yield (i.e. coupon interest divided by the market value…your 3.3% annualized) + the annualized difference between the maturity value and the current market price.
So, the YTM is a total yield calculation and it is a very good estimate of the total return you can expect for the remaining life of the bond. While it’s most relevant for an individual bond held to maturity, it’s also a good estimate of a bond portfolio’s future return.
The nice thing about ETFs is that they all post the yield to maturity of their bond ETFs. So if you look at XSB’s profile, you’ll see a YTM of 2.08% (as of June 29) – up 20 basis points in less than a week. While you are getting interest payments (i.e. a current yield) of 3.3% annualized, your total return if you hold this for any length of time is likely to be closer to 2% annually.
To a few of your specific questions…
You wrote:
If you look at all of the bonds held by XSB, you’ll see that virtually every one will mature at a value that is less than its current value. For example, look at XSB’s 2010 Annual Report – see pages 75 to 77 of this link – and you’ll find that all but a handful of the bonds held at year end will mature at a loss compared to Dec 31 market values.
Absolutely correct and that’s the case with most. But I’m focusing on the future return, of which YTM is a very good indicator. And if they sell that bond at a gain today, they’re going to replace it with a bond with a YTM of 2% (on average).
Yes. But those profits aren’t sustainable. XSB’s compound annualized returns are going to keep falling because they won’t benefit from the same level of falling yields (even if they fall, they can’t fall by as much) and as yields fall there is a lower and lower coupon which will lead to lower returns. Investors can either wait for the lower returns to show up in the returns or they can look at the YTM today as a sign of what’s coming.
Exactly and that’s awfully close to the quoted YTM. Ignore taxes for a moment and consider a hypothetical scenario where XSB didn’t pay out its interest; but instead kept in in the fund. You then wouldn’t see the 3.3% interest payments. You’d only see the blended return of the interest +/- price changes. I wrote this blog post because I think investors would be less enthused about a fund that had a ‘running return’ of 1.8-2% annually as compared to the current situation of effectively separating the interest and price effects.
Sounds like you have a good handle on these two factors Curtis. We just have to get a bunch more investors onside. Thanks again for your great comments and questions.
I am very new to understanding how bonds really work!! I need to buy bond ETFs for my portfolio to achieve a balanced asset allocation. I want some short term bond in the portfolio and am considering laddered ETFs from Claymore — both govt and corporate — instead of choosing iShares XSB. I intend to hold these for at least 10 years and rebalance as needed.
Would you recommend this — hypothetically speaking?
I can’t say whether you should buy these ETFs but I don’t have a major issue with them. Just make sure to take your time, read up on ETFs in general and the specific ones in which you’re interested. It may be worth checking out this article by James Hymas, where he addresses common misunderstandings of bond ETFs.
Thanks Dan. Is it possible to subscribe to your blogs through email notifications?
[...] I can’t tell you how many investors I’ve come across who over-estimate the yield they’re getting on bond ETFs. Here’s how to tell what your true … [...]
I interpreted “fair value” (as quoted in the XSB’s 2010 Annual Report) to be the “current market value” – meaning if the fair value was greater than the “average cost” they are in a capital gain position (it looks like at least +35% of the portfolio falls into this category). This is why “guaranteed loss” did not make sense.
I assumed the bond ETF has the option to sell these bonds for a profit, but I am guessing the issue is that while they could sell, those bonds have good coupons (e.g. 5%, 6%, 9%), and they would then have to replace them with bonds paying much less in this environment. If the ETF elects to hold these bonds to maturity however, and bond yields do not move up materially, they would be getting a good coupon and be in a capital gain position at maturity wouldn’t they?
To your point, the YTM doesn’t lie, so is my interpretation of fair value incorrect?
Maybe your point is better summarized as follows: “Given the preceding and current environment for bond yields, future distribution payments (on short term bond ETF’s) are likely to be lower than previous and current distributions, and the YTM is a better indicator”.
BTW, my 2.16% return is over 8 months; it’s closer to +3% annualized. Something tells me you don’t think I’ll see that going forward
.
Regardless, your intent of the article – YTM is a better indicator of future distributions than what is being paid now – is insightful and appreciated.
It would be interesting to go back in time to determine how closely the YTM followed the ETF distribution.
One correction – if held to maturity the majority of those bonds are in a loss position – their cost is more then par value. The ETF could sell for a profit now, but would have to find suitable coupon replacements to maintain the yield.
Thanks for a thought-provoking article, Dan. At the very least, the YTM data can be used for comparing various bond ETFs when making a purchase.
I’d appreciate it if you could clarify something. As I understand it, YTM (less MER) is an indication of the yield one would receive if an ETF were to hold the entire bond portfolio to maturity. Since (a) an ETF does not have a maturity date, and it keeps buying and selling the securities; and (b) most investors do hold ETFs for the long term as the fixed income component of their portfolios, doesn’t today’s YTM lose significance in the long term?
The only exception of this would be BMO’s Target Maturity ETFs (ZXA, ZXB, ZXC, ZXD) that do mature in the specified years. In the case of these ETFs, one would indeed recieve today’s YTM. Isn’t that right?
Thank you.
Be’en – there isn’t an email subscription per se but you have two other options. Toward the top right is a link to a RSS Feed. This is similar to an email subscription. When you subscribe to the RSS Feed, a “The Wealth Steward” sub-folder will be created in your email folders under your RSS Feed folder. Whenever we post a new article, a new message will appear in your “The Wealth Steward” folder, along with a link to the full article.
Also, on Twitter you can follow The Wealth Steward or me (@DanHallett)
Sorry guys, my last response was lost so it may be best to simply refer you to the most recent post, which addresses some of the issues raised in your great comments. See The relevance of YTM & the impact of rising rates.
This is an old conversation but, exactly what I was looking for. I want to buy some individual bonds and can find “Semi-Annual Yield”. I would like to buy Provincial and corporate bonds that mature in 2 years. I plan to hold the bonds to maturity. The majority of bonds I can find have a par value of $110 and “SA Yield” of 2%. So, if I buy “$10,000″ worth and the cupon payments are approximately 2% in an environment where inflation is rampant would I lose the difference between the premium I paid for the bonds and their maturing price but, be guaranteed the 2% cupon payments and $100 par value. Where can I find or how do I calculate YTM?
Thanks for your note Johnny. To answer your question about calculating YTM, this is best done using a financial calculator (do they still make these?) or Microsoft Excel. But there’s also a simpler formula that you can manage with a basic calculator. See this Q&A from Yahoo! Answers. You’re also likely to find YTM calculators around the internet. Finally, for Canadian bonds try this website, which lists bond quotes with yield information (and the yield there appears to be YTM).
Johnny, you had another question…
Par value is the value upon which coupon payments are based and it’s almost always equal to the $100 maturity value. Let’s use an actual bond to illustrate using real numbers. The Perimeter Financial Bond Quotation page lists a Quebec bond maturing on 2014-Dec-01 with a 5.5% coupon rate and a current market price of $110.09. The site lists this bond’s yield (which appears to be equal to YTM) at 1.63% as I write this.
(Since brokers sell bonds at prices higher than the price at which they buy from you – i.e. the spread – as a buyer you need to look at your broker’s quoted “offer yield”.)
In this example, you pay $111.87 ($110.09 market price + $1.78 of interest accrued through today) for a bond that will mature at $100. So you have a guaranteed capital loss of almost 12% if you buy at this price and hold until maturity (~2.7 years away). But you’ll get two interest payments totalling $5.50 per year for every $100 of par value. So, yes you’ll get a bigger coupon payment but you will see a loss on the purchase/maturity prices of the bond.