Q: Can you share your thoughts about the BMO Discount Bond (ZDB) and the Horizons Canadian Select Universe Bond (HBB) as long-term holdings in a taxable account? – D. F.
Earlier this year BMO and Horizons both launched bond ETFs specifically designed for taxable accounts. These two funds have very different structures, and each has its strengths and weaknesses. So let’s dig more deeply into each fund to help you decide which might be right for your portfolio.
Before we discuss these specific funds, let’s review the problem with holding traditional bond ETFs in non-registered accounts. Most bonds these days trade at a premium (higher than their par value), because they were issued when interest rates were higher. Premium bonds are perfectly fine in your RRSP or TFSA, but they are notoriously tax-inefficient and should not be held in non-registered accounts.
 Do you want a discount or a swap?
The BMO Discount Bond (ZDB), launched in February, is similar to traditional broad-market bond ETFs, such as the iShares Canadian Universe Bond (XBB), the Vanguard Canadian Aggregate Bond (VAB) and the BMO Aggregate Bond (ZAG). All of these funds include about 70% government bonds and 30% corporate bonds (VAB has slightly less in corporates), and they have an average maturity of about 10 years, a duration around seven years, and a yield to maturity in the neighbourhood of 2.35%.
The difference is that ZDB specifically selects bonds trading slightly under par. That means its average coupon (that is, the amount of interest paid out in cash) is lower than its yield to maturity. So while ZDB can be expected to post similar pre-tax returns to the ETFs mentioned above, it would likely come out ahead on an after-tax basis because of those lower distributions.
The Horizons Canadian Select Universe Bond (HBB) has a completely different structure. It’s also designed to mirror the characteristics of the broad-market funds mentioned above, but rather than holding bonds directly it gets exposure through a total return swap. (For more about HBB works, see this post.) The ETF pays no distributions at all: the swap is designed to deliver the total return of a diversified bond index, and all of the growth is reflected in price changes. Investors in HBB won’t have any tax liability until they ultimately sell their shares, at which point any growth would be taxed as a capital gain.
What’s your priority?
So if you need to hold bonds in a taxable account, which ETF should you choose?
Let’s begin by comparing costs. The BMO fund has a management fee of 0.20%, while the Horizons ETF has a management fee of 0.15% plus another 0.15% payable to the swap counterparty. So ZDB is about 10 basis points cheaper. But if HBB performs as expected, it seems clear it would have the edge after taxes, at least for high-income earners. By eliminating cash distributions the ETF could allow investors to defer taxes indefinitely. Moreover, future capital gains would be taxable at half the rate of interest income, and you’d have the flexibility to choose when to realize them.
However, HBB has a couple of additional risks. The first is the possibility—inherent in any swap-based product—that the counterparty will fail to meet its obligation. Granted, this seems like a very small risk: the counterparty is National Bank, and Canadian banks have a long track record of stability. There’s also a collateral arrangement in place to ensure that even if the counterparty did default, the loss to unitholders would be no more than 10%.
The second risk is that the good people in the Ministry of Finance may one day take a dim view of swap-based investment products. It’s impossible to know how likely that might be. The government famously changed the rules on income trusts back in 2006, and in 2013 it put an end to mutual funds and ETFs that recharacterized interest and dividends as capital gains. Swap-based ETFs do not recharacterize income in this way, but one can never ignore the possibility that tax-advantaged products are at the mercy of our elected officials.
What would happen if swap-based ETFs did eventually fall victim to a zealous finance minister? Chances are the funds would be liquidated, which could force investors to realize capital gains they had deferred to that point. That could result in a hefty tax bill you weren’t counting on all at once. But even if that were to happen, it’s possible you’d still come out ahead had you been able to benefit from the fund’s tax advantages over several years.
In the end, your decision comes down to which you prefer: the maximum tax deferral of a total-return swap with some additional risk, or the more transparent structure of a discount bond ETF with less tax-efficiency.
@Kevin: CDIC insurance would not apply here. But the assets held by any ETF are protected:
https://canadiancouchpotato.com/2014/06/02/ask-the-spud-how-are-investors-protected/
Hello,
A few months ago now I finally took the initiative to do a bit of research on investing which eventually brought me to the CCP. I now have my TFSA maxed out in with TD e-series funds instead of in a 1% savings account.
I’m now considering taking the next step and moving about $60k I have with Sun Life through a work savings (paycheck deduction and company matched) program to my TD Waterhouse account to set up a CCP inspired portfolio, Ideally this gives me more control and easier access to my money, while lowering my MER.
Since this money would be non-registered, I can’t decide if GICs or bond ETFs like VAB / HBB would be the best decision. Otherwise I was thinking matching the Option 4 of the model portfolios since I’m investing more than $50k.
Does this seem like a wise move to you? Could you offer some guidance to an investing noob like myself as to weather bonds or GICs make the most sense for this situation. For the TFSA this is all so simple but in the land of unregistered money things are very murky for me.
@Michael: In a non-registered account you probably don’t want to use an ETF like VAB or a bond mutual fund. A discount bond fund (ZDB) would be more tax efficient. GICs are also a good choice but tend to be more appropriate for larger accounts.
Excellent, I’ll look into ZDB. Excuse my ignorance but why do GIC become more appropriate with larger accounts?
@Michael: It’s really just a practical issue. Discount brokerages have minimums on GIC purchase, such as $5,000 or $10,000. So you may not have enough funds on hand to build a proper ladder. It’s also impossible to add money to a GIC, so if you’re contributing small amounts to the account every month or two and ETF is more flexible.
Thank you very much for your answers, and for this great website. The information provided here has been such a great help to me in taking control of my savings.
@Michael: Thanks for the comment, and good luck with your investing.
Hi CCP, I noticed that these two products have very low net assets, do you consider this to be an issue with liquidity of these Etfs?..I don’t want to get products that are difficult to trade. Thanks
@Carl: Assets under management and trading volume don’t necessarily have a major effect on ETF liquidity. As long as the bid-ask spread is just a few cents and you use limit orders you should be fine.
https://canadiancouchpotato.com/2012/09/10/etf-liquidity-and-trading-volume/
Hi Dan,
I understand that it is relatively tax-inefficient to place bonds in a taxable account, but if VAB is at over 8%, and you are in a low tax-bracket, doesn’t that still provide a better return, and better liquidity than a GIC at 2.95%,
which is the highest I could find (on a five-year term)?
Thanks for any clarification you can offer.
@Ros: Sorry for the late reply. The issue here is that you can’t know what VAB’s return will be ahead of time.
With a GIC you know exactly what the interest payments will be and the principle amount never changes. With a bond fund you have no such predictability. In 2013, VAB lost almost 2%, while this year it is up almost 8%. No one knows what its returns will be in 2015.
@CCP: The math in this situation is always pretty convoluted (to me, at any rate), but as I understood it, the problem is that not only is the future return unpredictable, but also, for the return of 8% this year, let’s say it works out to $x, the tax you pay is not the same as the tax you would pay on regular income in the amount of $x at a given marginal rate. Whether it is higher or lower depends upon whether the underlying bonds that are held in the ETF are on average premium or discount, isn’t that it? At this point I get confused, and can’t say for sure whether the 8% this year is associated with discount or premium bonds.
I am doing some rebalancing of my portfolio and I had a look at ZDB again out of curiosity. Unfortunately, what I saw tells me it still isn’t ready for prime time. Specifically, it has low trade volume so you’re always trading with the ETF marker maker. Because there is no robust multi-party market, one market maker can arbitrarily set the BID/ASK spread so it is currently still very large; one trade will likely cost you much more than a year’s worth of MER. HBB has the same problem.
I think some ETFs providers sometimes think we are fools and don’t look at the BID/ASK spreads. E.g. they claim low MER’s but hide the fees in the BID/ASK spreads the generate and/or collect. I’ve seen this with a lot of new and/or niche ETFs.
@Brian G: While the bid/ask spread situation can certainly play out to the investors’ detriment as you describe, and the ETF providers may indeed consider us naive, I don’t understand how you can accuse them of “hiding the fees in the BID/ASK spreads they generate”. You imply that the low MERs that they advertise understate what they are raking in. I agree that the high bid/ask spread generates an extra expense for the investor, but this extra money that is extracted from the investor does not end up in the hands of the ETF provider, as you imply. Or does it, and I am missing something?
@oldie, look at the mechanics of how an ETF works behind the scenes. My understanding is that when a company (the ETF sponsor) makes a new ETF it must also enter into a contract with a “authorized participant”/”market maker”. That agreement can have any terms negotiated in it or the sponsor and the market maker can even just be the same company as the sponsor. If they are the same company, then clearly the bid/ask spread can make their way back to the ETF company. If they two are different companies then I suspect it would be up to the terms of the agreement.
When you go to buy a lightly traded ETF, you are trading with the market maker most of the time. Market makers are there to provide liquidity AND to make a profit doing so. When the volume is low, to make enough profit in absolute terms, clearly they must charge more on each transaction and hence the desire to have the larger spread.
That’s what they need to do, but there is nothing saying I need to go down that path. Instead, if I stick with highly traded ETF’s, I avoid all this extra expense.
Hi CCP Great website.
I am new to investments and would like to do direct trading instead of investing in mutual fund from adviser. I have a Q. about TFSA. I have whole of my contribution unused and want to do direct trading in that account so in that case what ever income(dividends, gain, interest) i will be getting thru my investment that will be tax free as i am planning to open TFSA direct trading account?
And would also like to know is Scotia itrade is right choice as a broker?
Do we have to have TD waterhouse account to buy TD e series mutual funds at low cost?
And last but not the least about traditional DRIP and SPP can be done in TFSA?
@Beginner: Yes, any investment gains you earn inside a TFSA are tax-free. Scotia iTrade is a perfectly good brokerage, however only TD Direct offers the e-Series mutual funds. You can sign up for both DRIPs and systematic investment plans at any brokerage. Good luck!
Hi Dan,
Just checked out HBB, and noticed that there was a 9 cent bid/ask spread. Pardon my ignorance, but would you please explain why that is?
@Ros: That is indeed a pretty wide spread. The ETF seems to have low trading volume and the market makers don’t seem to be keeping that spread where it should be. Unfortunately, ZDB seems to have a wide spread, too.
Hello – I am compariing bond ETF’s and wanted to know if the performance numbers shown (annualized 1,3,5 year) are based on just the change in the unit value of the ETF or if the annualized performance also factors in the Yield return that is paid? (i.e. – I would have expected a high yield bond fund to have an equal or slightly better return than a standard bond fund over the past year but that does not appear to be the case. CHB vs XBB as an example).
I would appreciate the input to help with my rebalancing.
Thanks
@Randy: ETF performance numbers are total returns: they include price changes and reinvested distributions.
Som Seif founded Purpose Investments to get back into the game after his Claymore got sold to Blackrock. We jumped into his dividend ETF (PDF) as a Can/US replacement for our CDZ.
Now we have a large GIC maturing next month in non-registered and want to keep it in FI, but the opportunity to Swap with holdings in registered accounts (and buy equities in the margin accounts to replace the equity holdings in the RSP) does not appeal.
Do you have comments on his total return bond fund (PDB), which uses the corporate class structure to turn interest into capital gains or dividends for tax benefits.
When measured against PH&N Total Return D (RBF1340) or Can Bond Universe (XBB), the total returns since inception in September 2013 do not compare well, so a tax advantage may get lost in the much lower yield with the PBD.
What’s your current thoughts on the ZDB and HBB, as bid/ask spreads and liquidity may still be factors.
We got severely burned buying RBCs EAFE Quant (RID), as spreads remain over 12 cents, which is a brutal money grab for RBC when compared with BMOs ZEA of three cents.
@Davie15: I have no experience trading the Purpose ETFs, so I cannot comment on their bid-ask spreads. But in general I consider many of the funds to cross the line into active management, and therefore I don’t feel they are compatible with the strategy I recommend. The reason the Purpose bond fund underperformed last year is because of its “overlay of portfolio duration management”: in other words, they guessed wrong on where interest rates were headed.
Corporate class structures sound intuitively appealing, but they always come at a cost. I would be interest in seeing some analysis of their tax savings in the real world.
We have used ZDB with some clients. The bid-ask spread can be a little wider than we’d like, but the tax efficiency is likely to compensate for any small additional trading costs.
Hi Dan,
Wondering if you can tell me why VAB is doing well while HBB is not. Shouldn’t their performance be very similar?
Thx for any light you can shed on this
@Ros: Their indexes are not exactly the same: VAB has significantly more in government bonds, which have done better recently. A better comparison would be XBB and HBB, and they look to have performed very similarly.
By any chance are you looking at Google Finance?
https://canadiancouchpotato.com/2015/05/29/how-bad-data-leads-to-poor-investment-decisions/
Thx Dan
I have to hold some of my fixed income allocation in taxable. I’m considering switching from VAB to HBB in my taxable account.
I understand how duration works for regular bonds and bond funds ( https://canadiancouchpotato.com/2011/07/07/holding-your-bond-fund-for-the-duration/ and http://www.bogleheads.org/wiki/Bonds:_advanced_topics#Duration ). If interest rates go up, and you hold for the duration, while reinvesting dividends, you will be made whole.
I was trying to figure out if duration works the same way given the swap structure of HBB. Do you know of any quirks that HBB’s structure might have, which might affect its returns even if held to duration in a rising rate environment?
Thanks for all your insights.
@NG: The concept is the same: HBB is designed to deliver the same return as the broad bond market (like VAB), so it will rise or fall with interest rate changes in the same way.
Do you have any concerns over the liquidity of HBB or ZDB?
HBB has a volume of ~5000, assets of ~250M.
ZDB has a volume of ~13000, assets of ~500M.
Is there a maximum amount you would into either of these based on their relatively low liquidity?
@gocanada: Once an ETF has a couple of hundred million in assets it’s likely to be sustainable, so no worries there. In terms of volume, this is much less of an issue than it is often made out to be. As always, just watch the bid-ask spread, get Level 2 quotes and place limit orders:
https://canadiancouchpotato.com/2015/09/12/the-etf-volume-you-cant-hear/
Hello,
I have a question. I have the proceeds of the sale of my house (150 000 $) that I want to invest in a taxable account. My TSFA and my RRSP are maxed out. I’m a retiree of 71 and want a conservative portfolio. Would you recommend me to use your model portfolio, with 70 % in a bond fund even if it is non very tax-efficient?
What would be my other options?
Thanks,
Peter
@Peter: It’s impossible to make a specific recommendation for your situation without knowing all the details. But in general terms, it is not tax-efficient to use a traditional bond ETF in a taxable account. If you want to hold fixed income in a taxable account, GICs are likely to be a better choice, though you need to be aware they cannot be cashed out before maturity. If you want fixed income with liquidity and tax-efficiency, one of the funds mentioned in this post would be more appropriate.
Hi Dan,
As of March 5, 2016, XBB has Weighted Avg Coupon = 3.19%, and Weighted Avg YTM = 2.62%.
In the future, if XBB ever holds a portfolio where Weighted Avg Coupon < Weighted Avg YTM, will you be prepared to recommend it in taxable accounts?
@DM: Yes, absolutely. The tax-inefficiency of traditional bond ETFs today arises because they are full of premium bonds (coupon > YTM), and this is the result of a long period when interest rates trended down. If that trend reverses, then at some point these funds may be filled with discount bonds, which would make them quite tax efficient.
Hi Dan,
I didn’t know about the tax inefficiency of bonds until recently (from one of your podcast episodes) and I have about $10,000 of VAB and $80,000 of ZAG in my taxable account. I maxed out my TFSA and loaded my stock indexes more heavily into that account as I figured they would go up more (they have) and I would want them to have the tax-sheltered. (though I realize I may not be understanding the full picture)
My bonds are currently sitting with small capital losses.
Should I sell them and buy GICs, or the ZDB or HBB ETFs listed above?
And am I making a mistake by having a more securities-heavy portion of my portfolio in my TFSA and a more bond-heavy portion in my taxable account?
Thank you very much.
LOVE the podcast by the way. I think I listened to every episode in an afternoon once I discovered it.
@John: Thanks for the comment, and glad you’re enjoying the podcast. I would agree that it is usually a good idea to hold all equities in a TFSA to take full advantage of tax-free growth.
I cannot tell you what to sell or buy, but I do think you are correct to look for an alternative to traditional bond ETFs in taxable accounts. The small capital losses should not be a deterrent to selling.
Hi,
A slightly off topic question here perhaps, so apologies in advance.
I just finished reading Millionaire Teacher by Andrew Hallam (an excellent and entertaining read, as I’m sure many others have noted). I noticed that when he suggests a model index based portfolio he uses yours as a starting point, but makes one change, which is to use a short term bond fund in place of the aggregate fund you recommend. He cites inflation risk as the reason. This would seem to make intuitive sense, and yet you haven’t chosen it. I assume this is because there are pros and cons involved that I’m not aware of, and it’s not a straightforward decision. Can you explain why your bond fund recommendations differ?
Thanks
@Dave: The decision between short-term bond ETFs and broad-market bond ETFs is really about your comfort level with risk. In general, longer-term bonds can be expected to outperform (since investors are rewarded for locking up their money for greater periods), but long bonds are more sensitive to interest rate movements and are therefor less volatile.
In my opinion, investors with a time horizon of at least 8 to 10 years would benefit from using a broad-based bond fund, assuming they are comfortable with the slightly higher volatility.
This should help:
https://canadiancouchpotato.com/2015/05/18/how-changing-interest-rates-affect-fixed-income/
I was curious to know if VBG (Global ex-US fixed income) carries premium bonds. In other other words, is it equally bad holding in a taxable account?
@melwin: You can always check this by looking on the ETF’s website and comparing the average coupon to the yield to maturity. If the former number is higher, then the fund has a good number of premium bonds. As of today, the coupon is about 2% and the YTM is about 1%, so not very tax-efficient.