In February 2014, BMO released a more tax-efficient version of their flagship BMO Aggregate Bond Index ETF (ZAG). They called their new fund the BMO Discount Bond Index ETF (ZDB). The fund’s strategy was to buy Canadian government and corporate bonds that were trading at par or at a discount to their par value (for a more detailed discussion of ZAG vs. ZDB, please refer to Dan Bortolotti’s blog post, New Tax-Efficient ETFs from BMO). By purchasing lower coupon bonds, the ongoing interest received would be lower than traditional bond ETFs, resulting in a relatively lower tax liability for taxable investors). Clever strategy in theory, but let’s see whether this more tax-efficient structure actually worked out for investors.
The Results
Using my after-tax rate of return calculator, I’ve compared the returns of five plain-vanilla broad market bond ETFs that were available during the entire 2015 tax year. Before-tax, ZDB returned 3.60%, placing it in first place. After-tax, ZDB maintained its first place ranking, returning 2.52%. Its 2015 tax cost ratio (which is similar to a management expense ratio, but for taxes paid instead of management fees paid) was also the lowest of the group, at 1.04%.
2015 Before-Tax and After-Tax Returns
Broad Market Bond ETF |
AUM (millions) |
1-Year Before-Tax Return | 1-Year After-Tax Return (Pre-Liquidation) |
Tax Cost Ratio |
BMO Discount Bond Index ETF (ZDB) |
$184 |
3.60% | 2.52% |
1.04% |
iShares Core High Quality Canadian Bond Index ETF (XQB) |
$307 |
3.38% | 1.87% |
1.46% |
Vanguard Canadian Aggregate Bond Index ETF (VAB) |
$581 |
3.48% | 1.91% |
1.51% |
BMO Aggregate Bond Index ETF (ZAG) |
$859 |
3.24% | 1.61% |
1.58% |
iShares Canadian Universe Bond Index ETF (XBB) |
$2,042 |
3.14% | 1.39% |
1.70% |
Sources: CDS Innovations, BlackRock Canada, BMO Asset Management, Vanguard Investments Canada
In the chart above, it is also interesting to note that only $184 million is currently invested within ZDB (which is less than 5% of the total assets under management of all five bond ETFs). Hopefully this analysis will help to alert investors and their advisors to more tax-efficient alternatives to their current fixed income picks.
Hello Justin,
I have a balanced index fund in my corporate account and the balanced index is allocated 35% bonds. I have been considering if it would make sence to sell this fund and purchase the same amounts in separate etfs. I do like the idea of having seprate etfs for each asset as it would make rebalancing across accounts much easier. Would ZDB be more tax efficient in the long run? Just trying to figure out if this would make sense. The amount invested in the balanced index fund is over 220K.
Good afternoon Justin,
Would ZDB be a good candidate for a “forever bond’ in a non-registered portfolio?
I’m planning on holding on to this investment for the next 50 years.
I’m not interested in a GIC ladder. Just owning a bond that will not fall too much in value when interest rates normalize, whenever that may be, and remains tax efficient along the way.
Thank you!
@Nicolas – ZDB is a decent long-term holding in a non-registered account – you may want to wait until after the large 2020 year-end capital gains distribution before purchasing ZDB though (so you don’t get stuck with a big tax bill):
https://newsroom.bmo.com/2020-11-16-BMO-Global-Asset-Management-Announces-Estimated-Annual-Reinvested-Distributions-for-BMO-Exchange-Traded-Funds-and-ETF-Series
I’m curious about whether the existence of the tax advantage of discount bonds moves the market? Does the existence of this tax strategy result in the prices of discount bonds moving up at all relative to premium bonds?
@Moti: I guess it’s possible, but I doubt it. There’s probably more pension fund managers setting prices, and most of them could care less about discount vs. premium bonds (as they’re not taxable).
Justin what is meant by pre-liquidation return and post liquidation return and there significance to the discussion here. Also what tax rate are you assuming? thks
@Ronaldo:
Pre-liquidation return = After-tax return (from income distributions), but before capital gains taxes are realized from selling the security.
Post-liquidation return = After-tax return (from income distributions and from capital gains taxes realized from selling the security).
I believe the tax rate was 54% (the highest combined federal/provincial tax rate of any province during the time period).
Hi Justin,
In the current landscape where there’s probability of an interest rate hike, is ZDB still a good choice for fixing income in taxable accounts? I’ve got all my fixed income currently sitting in ZDB.
Thanks!
@Jeremy: The pundits who have been calling for an interest rate hike have been wrong for over 7 years – I wouldn’t put much stake in their predictions.
If you prefer to reduce your interest rate risk, you could consider a 1-5 year GIC ladder or a tax-efficient short-term bond ETF (such as BXF).
@Justin: Granted interest rate predictions have been pretty much worthless lately, but if rates SHOULD go up, is there any potential liquidity risk to a discount fund like ZDB or BXF (of which there was an article on in the Financial Post a couple days ago at http://business.financialpost.com/investing/funds/friends-with-tax-benefits-how-etfs-can-help-keep-the-taxman-at-bay). If other comparable duration bond funds to BXF such as VSB say became neutral overall instead of premium, could you not have a problem unloading BXF? In fact, the premium aspect of VSB already seems pretty small, and with the MER and yield differences, any after tax advantage is already negligible (or negative even) – unless I’m missing something (which is entirely possible).
@Dave: ZDB now has a weighted-average yield to maturity of 2.63%, and a weighted-average coupon of 2.06% (meaning that it is truly a discount bond ETF). A portion of the fund’s growth is now expected to consist of capital gains, making the ETF still more tax-efficient than a premium bond ETF, or even a par bond ETF.
VSB has a coupon of 2.5%, higher than it’s yield-to-maturity of 2.2% (so it is still considered a premium bond ETF). BXF would still be more tax-efficient than VSB at this time. If VSB eventually becomes a par bond ETF, First Asset may want to consider changing the investment objective of BXF to hold discount bonds instead.
For what it’s worth, I’ve been moving to a mix of 1-5 year GICs and ZDB in our client accounts for a few years now.
@Justin – In the table above, do the “1-Year Before-Tax Return” numbers have the Return of Capital distributions removed? When I look at the 2015 CDS Innovations spreadsheet for ZDB, it seems ZDB has much larger RoCs than VAB for example. If that 3.60% for ZDB includes RoC, then the real return would be about 3.28% I think.
The more I look at that ZDB spreadsheet, the more that 9% RoC looks fishy to me. Doesn’t that just artificially inflate the yield? (or is it the return – I can never keep them straight).
Oops! I’m pretty sure my calculation of 3.28% is wrong, but I’m still curious if the ROC was factored out or not.
Also I just read an article that said newer Bond ETFs will sometimes have higher ROCs if they’re growing fast from new investors so maybe the 9% ROC is not so fishy after all. Although I have to wonder how many investors don’t realize their ZDB distributions aren’t all interest and that 9% is their own money coming back and they better reduce their ACB before they sell any off!!
@Dean: For purposes of before-tax calculations, the full dividend is assumed to be reinvested back into the fund, so no matter what amount of return of capital is paid out, it is included back into the fund (so it does not artificially deflate or inflate the return). As you mentioned, the ROC can inflate the quoted distribution yield on the fund’s website.
So would you choose zdb over vab in the model portfolio?
@Pete – I think you may have missed the point of the article. It’s not that VAB (or any plain-vanilla bond ETF) needs to be replaced in my model portfolios, it’s that ZDB may be a more tax-efficient solution if you are holding bond ETFs in a taxable account. For RRSP and TFSA accounts, VAB would still be fine to hold.
@Justin – What would explain ZDB having a higher before tax return than VAB, while the MER of VAB is lower than ZDB? – Thanks!
@Que – the difference is likely due to sampling error from ZDB not holding all of the bonds in its index (and also due to the fact that ZDB and VAB follow different bond indices).
I bought $157K of the BMO Discount Bond Index ETF (ZDB) in May 2015 – it pays me $255 per month in interest and has appreciated 1.48%. I am happy with that. If only ZPR was even close to that. ZPR has lost 25.84% over that same time frame – fortunately I only put $59K into ZPR.
Would have been nice to also include Horizon HBB :)
@Sandra – HBB’s different tax structure and counterparty risk makes any comparison difficult. The closest you could get is to calculate a post-liquidation return for all ETFs (i.e. assume that each one is sold at the end of the 1-year holding period). I would estimate that HBB would come out slightly ahead of ZDB post-liquidation (by about 0.22%).