The Horizon Total Return Index (TRI) or swap-ETFs have several benefits that I wrote about previously. The most exciting of those is tax-efficiency. The risk premium from tax deferral and tax reduction of the TRI swap ETF structure is largest for those facing a heavy tax burden. The risk/benefit is particularly good for the bonds swap ETF (HBB) whether in a corporate account or a personal taxable account. Of course, that beneficial structure also carries some risks. The most obvious being the legislative risk.

Unsurprisingly, the T2 Legislator tried to blow up the TRI ETFs in their 2019 budget.
Fortunately, these funds have evolved in response. This past week, the TRI swap ETFs changed over to a corporate class structure. After spending some time to critically examine this and challenge some of the folks at Horizon with my questions/concerns [thanks for this], I want to share my understanding and thoughts about this change.
I think that these funds have emerged from the fiery miasma on a stronger footing.
You should form your own opinion if considering these products as part of your portfolio. This week, I’ll unpack what the legislative move actually was. That lays the groundwork for how the change to a corporate class structure addresses it and how robust that structural change is.
Disclosure: We have held HXDM, HBB, and HXX in our portfolio at various times. I don’t have any at present, but I likely will again in the future. I have no other financial relationship with Horizon at the time of writing this.
What about the TRI swap-ETFs made them vulnerable?
The swap ETFs were an obvious political/revenue target. They allowed a small group of investors (usually with high incomes) to reduce and defer tax via a hard to understand derivative contract structure. This is not something the average person would understand. Easy to sell politically as “sticking it some rich people using a tax loophole.”
The most obvious aspect (to a layperson) to use as the focus for the attack would be the derivative swap structure. While complicated, funds around the world commonly use the swap structure to invest. Derivatives contracts are regularly used to grease the wheels of our financial markets and that aspect is hard to attack without wide-ranging collateral damage. I fully explain the magic of the swap depicted in the graphic below elsewhere. It seems like a financial sleight-of-hand magic trick, doesn’t it? However, the Trudeaunator Government’s concern was actually not with the swap structure. On the contrary, they like magic – like magical self-balancing budgets.

The problem was with income allocation, not the swap.
The government’s objection was with how income was being assigned using the “allocation to redeemer methodology“. Allocation to redeemer methodology is a tax-rule that has been around for 20 years and is not really a loophole per se. It is a mechanism to prevent double taxation of capital gains from mutual funds. Mutual funds and ETFs are both “mutual fund trusts” – and are governed by the same rules.
Capital Gains Allocation – Implications for all Mutual Funds & ETFs but surgically targeted at the swap ETFs. For now.
The basic idea is that a fund trust may sell some holdings to get the cash needed to pay out to an investor who is redeeming their units. The capital gain on that would be allocated to the redeemer and the trust would get a tax credit for that.
In addition to capital gains generated by selling equities for cash when a fund unit is redeemed, fund trusts also generate some gains and losses while managing their overall assets. Under the previous rules, a trust could allocate these extra capital gains (from operations) to the redeeming unit holder and thereby keep the unrealized gains for the rest of the fund holders. The government felt this was an unfair tax deferral on the part of the fund trust and its remaining unit-holders. They wanted all costs/gains tracked to the redeeming unitholder level and all capital gains taxes paid immediately.
The realities of logistics and collateral damage reared their heads.

While swap-ETFs were clearly the target, this hamfisted legislation would impact virtually all mutual funds and ETFs. It is also practically impossible logistically to implement.
An ETF would have to track the adjusted cost base for individual investors who may hold a fund in multiple account types across multiple brokerages. How could an ETF provider track that!?! Like most attempts to modify the intelligent design of the tax code, there are predictably unintended consequences. It is like watching them modify the genes of dinosaurs in Jurassic Park.
Fortunately, the Federal Government tempered the final legislation due to the logistical impracticality and collateral damage. A threshold of “reasonable efforts” was set. Except for funds that are “not making regular distributions”. Like the Horizon TRI Swap-ETFs. The intended target.
Interest and Dividend Allocation – The Chink in the TRI ETF Armor
Similar to capital gains, other investment income can also be allocated using the allocation to redeemer methodology. What the swap ETFs were doing that the government objected to was allocating that income (interest/dividends) to their market makers.
Market makers are like the warehouse for stocks and bonds. They keep a ready inventory of these holdings and then build them into units to sell to an ETF provider when they need them. They also take them back when excess ETF units are sold. This provides liquidity to the market and the market makers make a small profit from the bid-ask spread on the turn-over. This profit is all taxed as business income.

So, technically the income generated was being allocated. However, the market makers were able to offset the income with expenses in their overall larger business. The government interpreted this as tax avoidance. The legislative response was to force income to be allocated to the unitholders or fund trust. Instead of a market maker.

Allocating income to the ETF fund trust ruins the tax efficiency benefit.

Expenses, like management fees, offset some of the interest or dividend income allocated to a trust. However, the rest is considered net income for the trust. Further, each ETF is a separate mutual fund trust. Only the expenses of that fund count against its income.
Unlike a large company, such as National Bank, an individual fund has limited flexibility to offset costs. Also, unlike a corporation, a mutual fund trust must pass all net income through to its shareholders each year. Retained earnings are punished.
So, this change effectively means that swap ETF “fund trusts” will now have to make annual distributions of the investment income net of expenses. Just like a regular ETF.
The swap ETF structure still has some advantages like reduced tracking error. However, its main benefit was the tax efficiency from not distributing income and only triggering a capital gain at redemption instead. Without an adaptation, this benefit would be… terminated.
Rise of the Resistance: Change to a Corporate Class ETF Structure

The organization of each ETF as a separate fund trust underlies the forced income distributions. The legal classification as a “mutual fund trust” also underlies the allocation to redeemer methodology. Fortunately, there is another option.
Change to corporate class funds.
How does that work exactly? What are the potential pitfalls? Is that a robust solution or just delaying the inevitable?
These are the questions that I immediately had. I spent some time grappling with them and had some detailed conversations with some folks from Horizon. Find out what I learned about the evolution to corporate class ETFs.
Thank you for an outstanding explanation of this very complex area.I have been following this topic over several months in the print media but there is nothing to compare with your post complete with pictorials.I look forward to Part 2 next week! ! Your readers should be aware that corporate Class structures are a $ 130 billion industry in Canada so the feds will be taking on the giants if they want to dismantle that area of tax law !
Thanks again for your great posts, you provide an invaluable education to people like me interested in this stuff.
Lyndon
Thanks, Lyndon. I know that Horizon has brought up the industry size as reassurance, but I am always reluctant to accept that at face value. The government is desperate and the only thing that would deter them is a messy intervention making them look bad from collateral damage. So, I think that the details are really important in assessing the risk of further intervention. I hope that I made some of the details a bit more clear. It gets really interesting next week when I take a deep look at the solution. I think the corporate structure is robust and a more straightforward business model that would be hard to dismantle without disrupting how businesses, as a whole, function in Canada. Along with that, a corporate group of investments can be set up efficiently or inefficiently. On our end, that means a big difference in return due to the fees, the options for investing, and the frequency and tax efficiency of distributions. I pick that apart next week and this is where I think it gets really cool. It is quite a clever set-up that they’ve done.
-LD
Great post LD!
I am waiting with baited breath for next week’s installment.
Damn – great cliffhanger. Waiting for next week.
Great post, Mark. I’m curious as to why you no longer own the swap based ETFs you had? I’ve had a chunk of HXT since 2012, but not the others. I also cancelled a medium sized universal life insurance policy about that time and have since put the annual premium into HXT designated for a charity I support. I figured out I only need to get to age 82 for the charity to get the same amount on my death, and of course more if I live longer, a reasonable bet considering my family history.
Hi Grant. It is pure coincidence actually. We held it all in my wife’s taxable account and decided to sell and use that money to buy a house last month. It was the best way to bridge finance ourselves (along with a small shareholder loan from our corp) until we sell our current huge house (which could take a while). The capital gains weren’t astronomical (I only discovered the swap ETFs a couple of years ago) and I had done some tax-loss harvesting last fall which offset it decently. When our house sells, our taxable account will be plumped back up and I plan to use them again. Probably more so, actually. I am much more comfortable with the new structure.
-LD
Now that there is the possibility of (eligible) distributions, will these mutual funds be tax deductible if funded by an investment loan or HELOC?
Hey AlphaDoc. That is a great question! I honestly don’t know the answer. I suspect it will be a grey area of interpretation.
-LD
https://www.canada.ca/en/revenue-agency/services/tax/technical-information/income-tax/income-tax-folios-index/series-3-property-investments-savings-plans/series-3-property-investments-savings-plan-folio-6-interest/income-tax-folio-s3-f6-c1-interest-deductibility.html#p1.70
This is the closest folio (see 1.70 and example 11) I can find. I do wonder if it will fail to meet the criteria because of the public expression by horizons to try to avoid dividends.
Thanks AlphaDoc. I think that example describes it.
-LD
You wouldn’t believe if I told you I got way more thrill reading it and you might’ve had writing it. Such is your explanation. May your tribe grow.
Given what happened to HOU and HOD with the price of oil futures… if these are in the same corporate structure and they had losses greater than the NAV of the derivative funds, wouldn’t the index CCMF’s or other non-derivative products be on the hook to bail them out? This was my concern and why I held off on re-buying any of the restructured funds – counterparty risk at Horizons itself.
Hi AlphaDoc. The oil futures moves really was something, eh. Generally a move up in HOD should have an equal move down in HOU. However, it was not 1 to 1 when things went wonky. So, there would be a potential loss for corporation overall. However, the loss could still be assigned to the fund that lost it first minimizing the spillover. The big concern would be if the NAV could go below zero (ie make a debt liability). They did decrease the leverage, The NAV changes based on the day to day % change not the value of the underlying asset. Generally, that means that an ETF NAV cannot actually go below zero. However, here we saw changes >100%. I don’t know the answer, but will see if I can find one.
Edit/Update: I asked some managers at Horizon about this. The NAV cannot go below zero. If it did go to zero, then it would be a loss booked to the corp. That loss would impact the shares of the decimated fund. However, it would be a benefit to the shareholders of the other funds. The corp would be able to use that to offset income from dividends/interest and not have to pay tax or make distributions to the other fund holders (the intent of a corp class structure). Currently, they have booked $2 billion in losses which means that they have a long runway before the corp class funds would need to make distributions. Actually a good thing for the index trackers.
-LD
Thanks for looking into this!
Sorry to beat a dead horse. I tried contacting horizons and got a form response… why are the counter party exposures to all their indexes so high and does this not increase the counterparty risk above the maximum of 10% exposure for a maximum of 30 days they quote?
Right now HXDM is at 17% at end of June (as of end of may it was around 15% too) and HXT and HXS are both at 27% now too.
Hmm. Interesting question!
-LD