HXDM ETF For International Exposure in a Corporate Account

Non-North American developed markets are an important part of a globally diversified portfolio. However, they also have high dividend yields. That could present a tax drag on investment growth in Canadian Controlled Private Corporations (CCPC). Both due to the magnitude of passive income and how foreign withholding taxes interact with corporate taxes. Horizons’ HXDM ETF looks to bypass those tax issues with its swap-based corporate class structure. However, that comes with some other compromises. Do the tax savings justify them?

The tax efficiency of investment income varies wildly depending on the corporate active income, dividends paid out, and the type of investment income received.

Corporate Tax Efficiency States

I detailed four main states of tax efficiency for corporate foreign passive income in my previous post analyzing HXS.

An Efficient Corp is paying out enough dividends as part of regular compensation to fund the owner’s lifestyle to release all of the RDTOH refunds. A corporation with RDTOH trapping has unreleased RDTOH and the owner would have to take extra dividends at a higher tax rate to get the smaller corporate refund.

Corporations with over $50K of aggregate investment income and enough active income to get bumped into paying some tax at the general corporate tax rates have two possible states. They could be over the limit, but the owner is able to use the extra eligible dividends generated by the GRIP. That represents a 60-75% corporate tax jump partially offset by 10-15% personal tax savings. If they do not use extra eligible dividends because they are unable to displace non-eligible dividend use, it is the full 60-75% tax bump. The situation is a bit more complex in Ontario and New Brunswick where the tax bump is lower and the savings may more than compensate for that, if the eligible dividends can be used effectively. So, I will use an Ontario and BC example in that section of the analysis.

Non-North American Foreign Withholding Tax & Corporations

zea xef viu hxdm

I detailed how US FWT interacts within a corporation in the preceding post. If using a Canadian-listed ETF that simply holds its US-listed version (like a wrapper), then the US FWT would reduce the nRDTOH available as a refund to the corporation. Plus, on top of that, the FWT from the non-NA dividends would be non-refundable.

That non-NA FWT averages out to ~8% tax drag for the indexes used. Then, the US FWT effect is layered onto the remainder. Two layers of tax. I teach my children about withholding taxes and how bad wrappers are every year.

The same thing happens if you use a US-listed ETF directly that holds the foreign stocks. The lower MER of the US fund may partially offset that, but it comes with the hassles of a USD account and currency exchanges. So, I will be using Canadian-listed ETFs that hold stocks directly for the comparison.

A Canadian-listed ETF holding the foreign stocks directly just has the FWT of the countries where the stocks are domiciled to contend with. Tax treaties make the FWT refundable as a credit to the corporation. However, that credit also shrinks the amount of nRDTOH refund available. In the top tax bracket in Ontario, the net effect is a total tax rate of 61.25% for flowing the dividend through a corporation compared to 53.53% if it were a direct personal investment.

The four main ETF providers in Canada all have ETFs that cover non-North American developed markets. Horizons’ HXDM tracks Horizons’ own EAFE total return index. It basically mirrors the MSCI EAFE index which is comprised of ~800 large-cap companies. BMO ETFs’ ZEA directly holds stocks that track the MSCI EAFE index. So, it is the most tax-efficient directly comparable conventional ETF to HXDM. Blackrock’s XEF also holds stocks directly, but tracks the MSCI EAFE IMI index. That index is slightly broader to include some smaller companies. Vanguard’s VIU holds stocks directly and tracks the FTSE All-Cap Ex-NA index of almost 4000 companies.

Conventional ETF Comparators

I will do a comparison of HXDM to ZEA and XEF. Both have MERs of 0.22%/yr. VIU’s MER is marginally higher at 0.23%/yr – not likely a significant difference.

Even though the detrimental effect of FWT is less for non-NA dividends vs US ones, the magnitude of the dividend also matters. The EAFE indices tracked have companies that pay big dividends. The yield varies, but an average of ~3.2%/yr over longer periods of time is a reasonable estimate.

HXDM ETF Structure Costs & Risks

HXDM has two main features that make it unique. It uses a swap contract and is also part of Horizons’ family of corporate class funds. The swap contracts allow Horizons to realize the total return of its index (capital gains and re-invested dividends) instead of holding the stocks directly or via another ETF. That means that it doesn’t collect foreign dividend income. There is no FWT, no income distributions, and just a capital gain. The taxable half of the capital gain is only taxed when the ETF is sold. That structure comes at the cost of a higher fee. Up to 0.30%/yr in swap fees. Plus, the 0.22%/yr MER. So, I will use a 0.52%/yr fee drag for the analysis.

Swap Contract Risk

The complex structure also brings some extra risks with it. There is counter-party risk because National Bank holds the contract and the amount owing grows as the index’s total return accrues. HXDM‘s current counterparty exposure sits at 18.55%. That is not very high when you consider the amount of exposure and that the probability of National Bank defaulting is low. However, if the counter-party exposure were to take off as markets take off, then Horizons could be forced to settle some contracts to manage the growing risk. That is the most serious risk (in my opinion).

Corp Class Income Management Risk

Swap income is treated as regular income for Horizons’ mutual fund corp. Not capital gains. If Horizons’ MF Corp has net zero income, then there is no tax. That is functioning as intended and can also be reduced by previous losses. Horizons’ non-capital income loss pool looks to have about 5 years of runway at its current level and burn rate. If there is net income, then the taxes within their corporation could get really nasty. We’ll have to see how Horizons manages that. They’d likely switch back to a conventional structure if the risk was growing rather than blow themselves up. The risk is if they get caught off guard.

When the money is flowing through, private corporations are relatively tax efficient. Still, in that setting foreign dividends are not treated kindly relative to capital gains. The net fee and tax drag reduction for buying and holding HXDM vs its best competitors is 0.47%/yr.

When the HXDM is eventually sold, the “income” does flow through and is taxed as a capital gain. Capital gains are functionally tax deferral and taxed at a lower rate (50% inclusion). That gives a slight advantage compared to regular income in the corporation of 0.21%.

However, if you are able to make use of the CDA to give a tax-free capital dividend instead of other income – that lowers the current year’s tax bill considerably. To effectively use a capital dividend, you’d need enough capital gains across your portfolio to justify the accounting fees. You could harvest them while staying fully invested. However, you also need to have a use for the personal cash. It lowers the amount of money paid out of the corporation in the current year. So, it is really another form of tax deferral until you take the savings out in the future. Still, a powerful tax planning strategy and corporate-class ETFs are built for it.

It could be possible for a corporation with a high passive income and owners that require few dividends to live on to develop RDTOH trapping. In that situation, the drag on foreign dividends increases because the 50% upfront tax is not attenuated by the RDTOH refund. That 50% tax rate is applied to the large dividends from international companies to make for some serious tax drag each year. If HXDM is bought and held instead, it has a 1.2%/yr lower fee & tax drag advantage.

When eventually sold and the capital gains realized and taxed, the advantage shrinks to 0.63% that year. However, you would have also had the larger 1.2%/yr advantage for the intervening years before then. You may not be using the CDA room generated since it would worsen RDTOH trapping. However, as I have mentioned in other posts, this type of “too much passive income” problem is solved by more personal spending or investing. Using the capital dividends generated would be an efficient way to move money out of the corporation and do that.

xef zea ccpc account

When a corporation is affected by the passive income limits, dividend income is punitively taxed. Except sometimes in Ontario and New Brunswick. These developed international markets spit out a lot of income. That makes the tax advantage of HXDM potentially massive for most provinces.

Even in Ontario, where a tax anomaly usually favors conventional ETFs, HXDM has a slight advantage of 0.15% if you are taking advantage of the break in tax integration. If unable to use the extra eligible dividends, it is a 1.5%/yr advantage.

In other provinces, like BC, it can be absolutely bonkers. If unable to use the extra GRIP generated by the corporate tax bump, there are more taxes than dividends received. Punitive Canadian taxation plus the unfavorable FWT treatment. If able to take advantage of the extra personal eligible dividends, HXDM has a 0.53%/yr advantage by avoiding the income distribution. However, if you can’t use the big GRIP balance from getting bumped to the general corp tax rate, the difference is a massive 2.87%.

Flowing the Capital Gains Through

The reality is that a corporation with enough passive income to be running afoul of the passive income limits also likely has a large enough portfolio to make annual capital gains harvesting and capital dividend usage worthwhile. That boosts corporate tax deferral. So, the main reason to avoid that would be if you expect a higher tax rate in the future. Not an impossibility with a large corporation either.

If harvesting and passing the capital gains through, HXDM has a major advantage. The extra tax deferral of using the CDA makes for a 0.4-1.7% advantage in ON and a 0.2-2.5% advantage in a province with normal tax integration, like BC.

If you had to realize the capital gains of HXDM for some reason and couldn’t use the CDA to your advantage, HXDM would still have a 0.8-1.5% advantage in BC. In Ontario, conventional ETFs would have a slight advantage if you were able to take advantage of the tax anomaly. Otherwise, HXDM would still have a 0.58% advantage.

Of all of the corporate-class ETFs, HXDM is the one that has the most potential advantages compared to its conventional peers when used in a Canadian corporation.

HXDM in an efficiently running Corp

During the buy and hold years, there is a moderate ~0.5%/yr advantage in an efficiently running corporation. If the corporation is efficiently passing out dividends and large enough to harvest and pay out capital dividends, the advantage of HXDM is huge. In the 1.2-1.6% range for that year. Plus, the advantages in the intervening years.

HXDM in a Corp with “too much” passive income

This is where HXDM shines. Avoiding the large dividends that international equities pay, the FWT inefficiency, and punitive Canadian taxes is a big advantage. That could be in the 0.2-1.5%/yr range depending on the province and the nuances of the passive income problem. A corporation large enough to have a passive income issue could also likely take advantage of capital gains harvesting and the CDA generated. That enables a 0.2-2.5% corporate tax deferral advantage in that year.

What about the risks of HXDM

I think that the risk of income within Horizons’ corporate class structure is a real risk. I wrote about it here. That said, it may be a ways off and the potential advantages of HXDM are significant. There could be up to 1.5% net mutual fund corp income attributed to HXDM before the advantage in an efficient corp is undone. For a corporation experiencing passive income problems or able to take advantage of capital dividends, there is a larger margin for safety. Like 3-4% net mutual fund corp income attributed to HXDM. I would be shocked if Horizons let that happen. However, it is a risk and the Titanic was thought to be unsinkable too.

The potential advantages of HXDM might make it worth considering if you are into using a more complex ETF investing strategy. However, whether to use a complex strategy is also an important question to ask yourself. For most people, the advantages of an asset allocation ETF strategy as a whole may be better than making it complicated for one slice of your asset allocation.

None of this is specific investing advice. It is just information to help you consider your own due diligence processes. While I have tried to be complete and accurate, there is no warranty to that effect. If you do find an error, let me know so that I can investigate and correct it as needed.


  1. Great analysis LD! I’ve been following your Horizons posts closely and HXDM is the one that intrigues me the most. The punitive nature of how non-eligible dividends is taxed in a corp, and the sizeable dividends distributed, is what makes the savings so great with HXDM. However, that’s also the reason why I don’t invest in foreign developed ETFs with a corp. I keep those ETFs in a TFSA + spouse TFSA + overflow in RRSP. How does the post tax end amount differ between using ZEA in a TFSA/RRSP vs HXDM in an efficient corp? Love to hear your thoughts!

    1. Hey Mark,

      We don’t have any non-NA developed in our corp either. If I did, I probably would probably use HXDM. However, we have ours stashed in TFSA/RRSP as conventional ETFs. In those accounts, I think a conventional ETF makes more sense. The FWT is puny compared to the swap fees and added risks. That is the fee and tax drag part (effect on the rate of growth).

      You raise another important point. A TFSA is after-tax, an RRSP is 100% pre-tax, and a corp is about 88% pre-tax. That is the tax-deferral part (effect on the flow through to get the money in the end).

      When trying to optimize asset location, minimizing the fee and tax drag is the key to determining location. Non-NA ETFs usually end-up tax-sheltered for that reason. The tax-deferral influences how much you need to achieve your goal asset allocation. If using a post-tax method, then you’d have to apply a discount to the value of each holding in an RRSP or Corp.

      It is pretty complicated (why I build Robocorp SWAT to try and automate the math). It allows the option of corp class ETFs.

  2. The following assumes all other things being equal. I’ve felt that Canadian stock exposure is better inside a CCPC than outside a CCPC. And after reading this post, I’m starting to wonder whether if you’re going to have nonCanadian stock exposure in a CCPC, it’s better for it to be American than EAFE. That’s because US stocks tend to have lower dividends, so less exposure to the increased tax on nonCanadian dividends in a CCPC. And lower dividends means less concern about exposure to passive income limits.

    However, all other things being equal is uncommon. And such an asset location strategy assumes that taxation and dividend policies don’t change. Finally, such an asset location strategy may result in diversification being distorted at the level of the CCPC or other account.

    1. Hey Park,

      I agree. The first big question is whether to attempt asset location at all (with the complexity and assumptions about the future that brings). If not, then an asset allocation ETF is pretty hard to beat in terms of cost and executing the plan effectively.

      If an asset location strategy is attempted, it has to be easy enough to actually implement reliably. When I’ve run longer-term simulations with my asset location calculator, the EAFA ends up in the registered accounts. The corporation is usually mostly Canadian and some US. Sometimes EMM and bonds. Depending on account size and asset allocation. As you mention, you’d need to keep asset allocation as the priority and not distort it for asset location. A lot of moving parts and tough in practice.

  3. When it comes to reaching an objective of not losing money after inflation, taxes and expenses, taxation of 53.53% is a significant headwind to overcome. Taxation of 61.25% only makes it worse. As you may have noticed, the goal here is modest; I’m just trying not to lose money.

    All of the potential ways to mitigate this problem have drawbacks. Horizons products are one solution, and like the rest, they are not without problems.

    Other ways are to avoid fixed income, tilt to Canadian stocks, asset location, maximizing tax efficient withdrawal from a CCPC and invest in low dividend stock. When it comes to low dividend stocks, a tilt to small and/or growth will do that.

    Any other suggestions?

    1. That summarizes the issue nicely. Minimizing our loss of money to fees and taxes is about as close to “control” as anything we have with investing. We can also control our behavior. Sort of. The options that you mention are the big ones from an investment portfolio standpoint. That is largely what I do personally, and I don’t have anything else to add. There will be compromises and the right answer varies for the individual for sure.

      A few others that I have on my list depend on personal situation and preferences too. They are more tax planning oriented. The easy one is to use all of your registered accounts. Seems obvious, but there are many people who don’t! If already using an advisor and that is a sunk cost, then an IPP could be useful to create more tax-sheltered room and shift money out of the corp. Income smoothing over your lifespan is another one (ie siphoning money out of the corp to invest personally or spend slowly) rather than deferring so much from a moderate tax bracket now only to be forced to take it at a higher one later. Keeping the money moving out of a corp keeps it pretty efficient. Using capital gains harvesting and capital dividends helps. We did that over the years with my lower-income spouse for her to invest personally, and it has helped keep our tax rate and corp passive income down immensely. Planning charitable donations with appreciated stock from the corp is another good way to shift money out as long as you have more than enough and have charitable inclinations.

      The other “Good Life” planning strategy that fits with the corporate investing taxation is to reflect and adjust work and spending (if you have that flexibility). By the time things like corporate inefficiency really ramp up with RDTOH trapping or the passive income limits, it implies a lot of income/investments and proportionally less spending.

      Those are the big ones that I can think of, but they aren’t portfolio management per se. Notably not on that list is permanent life insurance. It gets marketed as a way to invest tax efficiently and avoid passive income, but a lot of costs (and even taxes and other risks) buried in it that make the real life return not live up to people’s expectations when they are sold the product.

Leave a Reply

Your email address will not be published. Required fields are marked *