What’s up with income trusts?

Just weeks ago it was Telus that was the biggest ever conversion to an income trust. Now BCE jumps to the top. This mania for income trusts has me wondering how the rash of conversions from corporate entities to income trusts can make good economic sense.

Income trusts are clearly a vehicle by which corporate taxes can be avoided, that much is pretty obvious. The huge potential for lost tax revenue to governments has been the most pressing concern expressed to date (about $500 million per year according to Jack Mintz). Conversions are profitable for the Bay St firms doing these deals, for senior executives who proposed the move, and for investors who may get a capital gain out of the mania that is driving up prices.

So income trusts are in play for a number of follow-the-money reasons. But do income trusts actually make sense from a macroeconomic perspective?

A recent story in Canadian Business magazine sheds some light on the downsides:

[W]hile the deal may be good for the company, the business trust sector is a relatively new one, and that means there are a few loose ends that need to be tied up. Take your pick of the issues: erosion of the tax base, a turning back of the clock on transparency and governance or shady accounting around what is supposed to be a measure of net income. There is also the possibility of a serious decline in the value of income trusts, some of which seem to have been dumped on retail investors by private-equity firms hoping to get sad-sack companies off of their books. The worst-case scenario? Trusts may seriously underestimate the amount of money they need to maintain the business and choose to funnel it back to investors in what is, basically, a giant Ponzi scheme. And that sets up the sector for a spectacular crash.

It would be a political disaster for the federal government. After all, it’s not like they could say they didn’t know. The number of warning flares that have been fired over the business income trust sector is increasing. In the spring, Standard & Poor’s released an industry report card that suggested “credibility deficiency” appears to be the “underlying discourse” in the market.

… Trusts, of course, have existed for decades. They are an accepted methods of business in the oil and gas sector and the real estate sector. The peculiar tax traits of a trust–cash generated by assets is flowed straight through to investors–make sense for certain cash-generating assets such as apartment buildings, which throw off steady revenue streams without the need to set aside money for capital expenditure. Trusts, in fact, are built to do that. One of the positive aspects of trusts is that they facilitate leaving assets to someone who won’t be directly managing them, such as the case of a trust fund for a child not yet 18. … It’s a way to deliver a revenue stream from an asset to someone who isn’t its manager, while also keeping the hands of the day-to-day managers off of the asset.

One of the main traits of trusts is the legal requirement that any money generated has to be paid straight through to the unitholder. No tax is paid by the company on the distributed money. The government calls trusts “flow-through entities” for that reason. This is unlike corporations, which pay tax on net income.

The original intent of trusts, however, has evolved in interesting ways in the early years of the 21st century. Since 2000, an entire new sector–the business income trust sector–has bloomed in Canada. Investment bankers found that if they applied the trust model to all sorts of other businesses, they could find a ready market for cash-generating entities (like, say, the company that collects payments on the water heater in your basement). That market doesn’t change much. It’s basically the same set of transactions year after year (with some slight growth), and is steady as can be. In the wake of the dot-com crash, that steady flow of cash seemed perfect. Throw in the low-interest environment–which made the relatively high yields on these securities even more attractive–and the rush was on.

… Income trusts also take advantage of a tax writeoff on debt held in a trust. As a result, many ramp up debt as far as they can. (In fact, one of the key management challenges in a trust is to get the level of debt just right.) They also need to match distributions in their sector, and so trusts are often highly leveraged entities. “In a normal public company, you’ve got a balance between growth and distribution,” says Kerry Ho, CEO of Inhance. “You can dictate what yields you want, but once you go the income fund route, and you’re giving out 70% of your earnings, you have to compete against the other yields that like sectors are offering. You are forced into the yield for whatever is being offered in that sector, versus the corporate model where you dictate your growth rate and where you want to focus the cash.”That isn’t intrinsically wrong, it’s just a difference of structure. In fact, the loss of flexibility was promoted as one of the benefits of owning trusts; the thinking was that if managers can’t spend your money on bad acquisitions, the investor wins.

Well, at least some of the time. When that loss of flexibility is combined with current disclosure regulations–trust management does not have to inform unitholders about any of the numerous debt covenants that may apply to the company, not even if one is about to be breached, triggering payments to bondholders–trusts will regularly surprise the market and rapidly sell off.

A bigger macro concern could be around the impact on investment, which has already been declining as a share of GDP. Here’s The Globe’s Eric Reguly:

Finally, there is the issue of national competitiveness. Mr. Mintz, and others, argue eloquently that trust status does not suit all companies, that when you starve a company of retained profits because you have to pay most of that profit to investors, something’s got to give. Like enough capital expenditures to stay competitive.

Telus and BCE may come to regret their decisions. Telecoms is a dynamic business. Billions of dollars are being pumped into the companies’ networks so they can handle broadband services like TV and high-speed Internet. When the next must-have technology comes along in three years or five, will Telus and BCE as trusts have the financial firepower to invest?

They insist they’ll be capable of funding all the capital expenditures needed and keep investors happy with fatter payouts. We’ll see.

I’m open to good arguments as to why income trust conversions are more than just tax evasion and actually have economic benefits. Any takers?


  • A bigger macro concern could be around the impact on investment, which has already been declining as a share of GDP.

    That’s not what I see:


  • Where did you get your numbers, Stephen?

  • Stephen, your graph looks fishy to me, especially the steady upward trend going back the 1960s. The CANSIM series I’m drawing on is 380-0002. I’ll email it to you but here is the quick summary:

    Machinery and equipment as a percent of GDP fell from 8.1% in 1998 (up a bit from the year before) and has dropped since then to 6.6% for 2003-2005 inclusive.

    Going back to 1961 there are ups and downs but they do not drop as low as you have them in the 1960s (5-7% range rather than your 3-4% range).

  • It’s investment expenditures on machinery and equipment (CANSIM series v1992056), expressed as a percentage of GDP (v1992067):


    They’re from Table 380-0002.

    I just looked them up again afresh, and I got the same thing.

  • I just got your e-mail. I don’t understand what’s going on. What are the CANSIM codes for the series you are using?

  • I just got the data through Table 380-0002, which is the one referenced via your v series. The only difference is that yours are in chained 1997 dollars and mine are in current dollars but that should not matter.


  • Huh. The price of M&E appears to be going up more slowly than the price of everything else.

    I suppose you could explain that if you thought that technical progress would be more pronounced in M&E than elsewhere in the economy. Not an implausible scenario.

  • That could well be it. Technological change raises some problems for price indices in general. But if I’m reading you right, M&E expenditures could be rising by less in constant dollars than real GDP because of the impact of (I’m guessing) drops in the cost of computers and other information technology.

    I wonder what this looks like in terms of how the price index for machinery and equipment is derived? This must have been a thorny issue when they were last revising the National Accounts.

    Anyone from Statscan care to jump in?

  • Oh yes, one other thing: I think that a disportionate amount of M&E is imported, so the appreciation of the CAD would also put downward pressure on its relative price.

    But that would only be a story for the past few years, though.

  • There is a bit of a debate going on over @ our blog on the tax implications of income trusts. We ran a simplified model using the Dept of Finances assumptions to assess the tax revenue implications of the BCE trust conversion. Our projection runs somewhere in the neighbourhood of 8 billion forgone over ten years in tax revenue.

  • Marc’s original point was that business investment has fallen to a low level relative to GDP. The subsequent discussion has focused exclusively on the machinery-and-equipment component of such investment. However, construction is also an important component. Even if Stephen’s machinery-and-equipment numbers are accurate, I think that Marc is correct that overall business investment is low relative to GDP. In other words, other components of business investment are probably down relative to machinery and equipment.

  • You get the same thing if you plot the series for (real) non-residential structures and equipment (v1992054).

    If you look at non-residential structures alone (v1992055) as a share of GDP, then you won’t find much, that’s true. But since a) M&E is something like 2 times as big as nonres structures these days, and b) M&E is more closely tied to improvements in productivity growth than are nonres structures, I don’t that matters so much in this context.

  • Oh yeah, income trusts. The key point was that it has been suggested that income trusts might put a damper on new capital investment, and would thus be bad for productivity growth in the future.

    Anyway, I’m inclined to want to use my numbers which are based on nominal figures for each of GDP and M&E. This is the most straightforward. What Stephen is showing is an artifact of the delating of those numbers by different price indices, which distorts the time trend. I think the calculation based on nominals also better reflects the historical growth record.

    One more thing. In Stephen’s graph, there is big jump in 1997 when they last did a major overhaul of the National Accounts, so that could explain the difference if some new M&E price index was invoked. On the other hand, 1997 was a take-off year for growth after a slow period from the 1990/91 recession up to that point. The only way to fully resolve this is to look more closely at the methodology and weighting of the M&E index — there is a good research paper in here for someone but it will not be me.

  • Pingback: Trust Calculation Debate Part 4: An Issue of Double Taxation or a Boondoggle? « Canadian Observer

  • Marc, I’ve taken your bait! But my response became too long and this post is now buried, so I posted my reply here:


  • Here is the comment I posted on True Dough’s site as a response:

    Thanks True Dough for some interesting fodder on income trusts. It will be interesting to see if your point one plays out in practice — it would be nice if income trusts were able to reign in excessive executive salaries, though I am skeptical that this will actually happen.

    I think the issue around taxation could readily be solved if there was the political will to do something about it.

    But the most compelling point I have seen is that conversion to a trust creates incentives that may lead to less investment, and therefore lower productivity in the future. This is just speculation not empirical evidence but if true does indeed have profound negative macro impacts that would dwarf any marginal efficiency gains suggested above.

    It may be the case that for some forms of activity income trusts are an appropriate governance structure — don’t get me wrong. But if the driving force behind conversions is indeed tax avoidance, facillitated by the economic incentives of corporate executives, bankers and shareholders, then this asymmetrical treatment may be leading companies to convert to income trusts who should not be doing so were the playing field level.

    I’d be interested in any empirical research on the investment implications of income trusts before I come down with a final decision.

  • So, This ‘is’ the time for risky experiments? (Income Trusts: Good for Canadian companies and Canadians or not? Let’s try them.) Then I guess Stephane Dion’s our man. Not.

Leave a Reply

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