Archived - TD Bank Financial Group's Submission in Response to Finance Canada's Tax and Other Issues Related to Publicly Listed Flow-Through Entities (Income Trusts and Limited Partnerships) consultation
Archived information is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.
Contact: Don Drummond, TDBFG
November 21, 2005
TD Bank Financial Group Submission to the Department of Finance on Flow-Through Entities
TD Bank Financial Group welcomes the opportunity to respond to the consultation paper published by the Department of Finance September 8, 2005 on flow-through entities (FTEs). We will comment on the impact of FTEs on government revenues and economic performance and make policy recommendations that will improve the structure of Canadian taxation and hence the performance of the Canadian economy.
The consultation document casts a very broad net with its rather open-ended definition of FTEs. We will focus our remarks on business income trusts, energy trusts and RElTs. We do not believe that the concerns and possible policy approaches the Department of Finance sets out in the consultation document are applicable to some business practices such as securitizations that may share some common traits with the general description given for FTEs.
First, we state the 4 key premises that shape TDBFG's views on FTEs:
- Decisions on what an entity should do with its earnings should be left to management and shareholders. The tax system should be neutral as to whether the earnings are retained or distributed to shareholders.
- The current tax system does not provide neutrality. For large corporations there is a bias against distributions due to the double taxation of dividends. And the lower tax rate on capital gains than dividends means it is more tax efficient to make distributions via capital gains (through share buybacks for example) than by dividends.
- Trusts have helped alleviate the bias against distributions and for that, and other reasons discussed below, have improved the efficiency of capital markets. The key question is whether this "market solution" is the best approach. We argue that trusts have a legitimate place in the Canadian economy, but efficiency and fairness require fundamental changes to remove the bias against distributions being made from corporate entities.
- "Do no harm". In most circumstances such a statement can be left unsaid. However, as the market value of trusts had risen to around $170 billion by the time the consultation document was issued, many approaches that might appear promising in theory, or could have been feasible in previous years before the market grew, would now cause large losses in market valuations. This in turn would cause unacceptable wealth losses to Canadian investors and hurt the Canadian economy. Further, some of the options for addressing concerns set out in the consultation paper would impact on other distortions present in Canada's tax system so great care must be taken not to inflict unintended consequences.
Revenues and FTEs
TDBFG has concluded that it is virtually impossible to estimate with any precision the impact of FTEs on government revenues. At any rate, we do not believe this is a principal consideration for policy purposes.
The FTE consultation paper gives a "central" estimate of a $300 million revenue loss to the federal government for the 2004 taxation. The impact of altering some key assumptions such as foreign and tax-deferred ownership is sensibly given. In addition to these uncertainties, there are many other considerations that taken in total surely undermine the credibility any particular estimate can have.
Among the additional uncertainties, consider the following:
- The $300 million "central" estimate is for the 2004 taxation year when the market value of FTEs averaged about $100 billion. The value had ballooned to about $170 billion by September 2005 when the consultation document was released. Everything else equal that would suggest the revenue loss had grown to an annualized rate more like $510 million. Moreover, in the absence of certain policy changes, the overall market values would likely have risen further and ownership from the tax-deferred sector would have grown. Therefore, the revenue loss would likely have continued to rise rapidly.
- Finance does not seem to have included the capital tax loss from FTEs. While this would raise the revenue loss estimate, it should not be a relevant consideration for policy as the capital tax is being phased out (the November 14, 2005 fiscal update commits to accelerating the phase out so that the tax is eliminated for 2006).
- On the other hand, just to cite one potential factor that could lower the estimate, Finance benchmarks the loss against the revenues that would flow from public corporate entities. However, some of the FTEs have been created out of private and foreign equity, in which case the government would not likely have collected revenues commensurate with the full corporate tax rate. Indeed, the record is clear that many of the entities that have converted to FTE structures in recent years were paying little or no corporate income tax because they were already highly leveraged.
- Finance appears to have counted the revenue implications from the tax-deferred sector entirely as revenue losses as opposed to calculating the net present value. Indeed, the consultation document repeatedly refers to the sector as tax exempt rather than tax deferred. This may stem from a belief that the rate of return to the tax-deferred sector is close to the discount rate. Under this condition a tax deferral is effectively a tax exemption. However, as the highest personal income tax rate is above the corporate tax rate and the tax-deferred sector effectively converts some capital gains (which are subject to a lower tax rate) to an income stream, the sector does likely pay taxes from income trusts on a net present value basis. Further, if defined benefit plans with tax-paying sponsors realize higher rates of return by shifting toward FTEs, then there would be a direct reduction of the deductible pension expenses of those sponsors. The experience of the Ontario Teachers' Pension Plan provides an illustration of how the net present value of taxes from the tax-deferred sector can be considerable. Over the past 10 years their plan has realized a rate of return that considerably exceeds the discount rate for their liabilities. As such, the net present value of taxes that will ultimately be paid is not far below the revenue that would have been collected at the highest marginal tax rate when the income was initially generated.
- Given the growth in the FTE market, it is impossible to do the calculation on a static basis. If the emergence of FTEs has created market wealth above and beyond what would have occurred in their absence, then additional revenues have been collected. Related to this point, Finance may have underestimated the size of the capital gain arising from conversions given the large premium US. residents are clearly willing to pay.
Some observers have suggested that the Department of Finance has overstated any revenue losses because they do not appear to have fully accounted for the fact that the top personal income tax rate exceeds the large corporation tax rate. This, according to some, suggests that FTEs actually raise the amount of revenue collected from the Canadian taxable sector. This would unambiguously be true if the benchmark were a tax system with proper integration between personal and corporate income taxes. But the benchmark is the current tax system with its double taxation of dividends. It is not clear that the margin of the personal income tax over the corporate tax is sufficient to offset the ("illicit") revenue gains the government receives from the double taxation.
Since the consultation document was released, the Alberta Government has suggested that energy trusts alone are costing their provincial treasury more than $300 million per year. On the surface, this may suggest that the federal revenue loss is understated. However, that is not necessarily the case. Some of the Alberta revenue loss may be a revenue gain (in a partial sense) to other provinces. For example, if an Ontario resident owns a unit in the Alberta-based trust, then Alberta and the federal government are deprived of the corporate income tax that would have flown under a corporate structure, but Ontario and the federal government receive higher personal income tax collections.
Due to the myriad of assumptions that must be made, many of which are next-to-impossible to quantify, TDBFG has determined that it would not be meaningful to attempt to provide an estimate of the revenue implications independent of the figures in the Department of Finance consultation document. More meaningfully, we do not believe that estimates of overall revenue impacts should be a primary motivator for policy change.
To the degree revenue implications are taken into account for policy considerations, the focus should be on where the losses might occur. If the benchmark is a tax system with integration of the personal and corporate tax regimes, then unambiguously the Canadian taxable sector is not a source of revenue loss. The impact of foreigners and the tax-deferred sector on revenues must be considered in the broader context of the economic impacts of FTEs. We return to this issue following a discussion of FTEs and economic efficiency.
Economic Performance and FTEs
TDBFG believes the central issue is that the Canadian tax system is biased against corporate distributions due to the double taxation of dividends. This bias has long distorted corporate behaviour, skewing business practices toward retaining earnings, buying back shares and leveraging high debt. Any time business practices are driven by tax considerations rather than business fundamentals there is inherently an economic inefficiency. FTEs in general and the explosion of market value in business income trusts over the last 5 years in particular are simply a recent response to the biases inherent in the tax system. Now in addition to the business practices of corporations being driven by tax considerations, the structure of entities is in good part being driven by opportunities to reduce tax.
FTEs increase efficiency because they address a distortion in the tax system that creates a bias against distributions. But a secondary distortion is created in that in order to distribute earnings in a tax-efficient manner, an entity must change its structure and plan on distributing all of its taxable income. This is costly and in itself leads to changes in practices that are driven for tax rather than business considerations.
TDBFG recommends that the focus be put on eliminating the bias against corporations distributing earnings. Put more broadly, the tax system should create neutrality as to what a corporation does with its earnings. Business and economic factors should be the drivers, not tax. If there were neutrality, there may well still be a legitimate role for FTEs. But it would not be for the primary reason of reducing taxes paid.
There is no solid case for acting to prohibit or even limit the growth of FTEs. The record over the past 5 years shows that they have not behaved in a way that warrants the worst fears about how they could compromise growth. They have not been confined to low-growth industries. FTEs can be found in, among others, the following high-growth sectors: trucking, air freight and logistics; energy services; energy re-sellers; media; retail; and healthcare. Much of their growth comes from acquisitions where they have facilitated consolidation in some industries. Some growth has come from investments (the CBA cites re-investment at 27 per cent of earnings). And they have brought new capital into Canada.
Concern has been expressed that FTEs may tip the distortion in the opposite direction and encourage distribution of earnings for tax considerations. There may be some validity to this, although it should be noted that entities do not necessarily have to convert all their operations to the FTE structure. They can retain corporate status for parts of the business organically growing and convert the more mature, stable elements. Further, an entity can choose the timing of when it wishes to convert. It may wish to retain the corporate structure during the early, growth phase of business and only convert once the business has matured. The conversion process is not irreversible. Some trusts have spun out assets into a public corporation to finance an expansion.
There may be an inefficiency with FTEs returning to the market with secondary offerings as opposed to financing through retained earnings, but this is simply a consideration that an entity needs to factor in any decision about changing corporate structure.
Finally, the distributions of FTEs are not lost to the economy. Individual investors consume or more likely invest the proceeds back into the economy. Their investments may well be more efficient and growth-enhancing than the ways in which a corporation would have used the funds. Indeed, studies suggest that some fIrms with substantial retained earnings have a low ratio of market to book value and this ratio is often improved as dividends are increased. Implicitly this result suggests that the retained earnings are being poorly invested.
This does not mean that all is well with the structure the Canadian business world has meandered into. In certain respects FTEs have an advantage over corporations that goes beyond their ability to skirt double taxation of distributions. Even with integration for taxable investors income trusts would be more attractive to the Canadian tax-deferred sector and foreigners because these parties can receive the income from the FTEs without corporate taxation having been stripped out (as discussed below, a refundable dividend tax credit available to the tax-deferred sector would address this). Some argue that a taxable entity can lower its cost of capital by spinning out part of its business as a FTE. This depends on how the overall cost of capital is determined. For example, if the Canadian taxable entities, as the marginal supplier of capital, set the cost of capital, then all firms will face the same cost.
Some entities may perceive that they face barriers in converting parts of their business to a trust structure. For example, there may be a perception that the parts remaining under the corporate structure will be obligated to "cross subsidize" any drops in disbursements from the flow-through entity. Countering this perception may require that the entity detach the flow-through part to an extent that leaves them uncomfortable over the remaining degree of control. Nonetheless, there are many examples of corporations having spun-out assets into trusts where they continue to have an ownership stake without an obligation to provide indemnities or subsidies (in some cases there is a temporary back-stop as a marketing feature of the offer).
Our conclusion on economic efficiency is that the focus should be on removing the tax distortion that creates a bias against corporations distributing their earnings. If this were done, much, but not all, of the incentive to create income trusts would be eliminated. They would still be a legitimate structure in many circumstances. Finally, even in a more neutral tax system, there is a concern that FTEs would have an advantage over corporations because of their relative attractiveness to foreigners and the Canadian tax deferred sector. It should be noted, however, that such an advantage is not limited to FTEs. For example, private equity firms use tax-exempt equity to acquire mature assets through leveraged buy-outs. So private LBOs should also be a concern if the Department of Finance is worried about income trusts.
We have identified the principal issue as the bias in the tax system against distributions by corporations. So the focus of the policy thrust should be to eliminate double taxation of dividends. We would argue that this should be done even if there were no FTE sector. The double taxation distorts business behaviour and lowers the after-tax rate of return to investors. With the advent of FTEs, it also distorts the structure of businesses.
Properly integrating the personal and corporate income tax systems with respect to dividends paid by large corporations would require lowering the personal rate of taxation on dividends to remove the double taxation. A "pure" integration model would require going further and making the DTC refundable to both the taxable and tax-deferred sectors. This second step in particular would be very expensive and not, therefore, perhaps something the Department of Finance might contemplate for the near-term. In particular, this would exacerbate the revenue losses they appear to be concerned about with the tax-deferred sector. It is a worthy objective to work toward over time, however. Integration would make taxable entities indifferent between holding shares in a public corporation, equity in a FTE or debt. Under certain circumstances it would also lower the cost of capital in Canada.
A series of measures could be taken in the near-term to address some of the concerns with FTEs at considerably lower cost. The starting point is to reduce the large corporation tax rate from 21 to 19 per cent. The commitment was re-iterated in the November 14, 2005 economic and fiscal update and it so happens that it helps considerably in reducing double taxation of dividends. This can be seen in the Department of Finance's estimates of the revenue loss from FTEs. The $300 million estimated with a 21 per cent corporate tax rate drops to $135 million under a 19 per cent rate. The second move would be to raise the DTC (or some combination of changing the gross-up and credit). The overall (combined federal-provincial) rate of personal taxation on dividends would need to drop from 31 per cent to 20-21 per cent (provided the federal corporate income tax rate has been dropped to 19 per cent; otherwise the tax rate on dividends would need to go to about 15 per cent). As the small business tax rate is so much lower than the general rate, this would result in over-integration for dividends paid by small businesses. Hence, it would likely be necessary to have two dividend tax regimes distinguished by whether the dividends were paid by large or small corporations.
To be effective in addressing some of the concerns the Department of Finance has expressed with FTEs, the corporate tax rate should be dropped to 19 per cent effective 2006 and the capital tax elimination, as per the undertaking in the fall economic and fiscal update, should be eliminated next year. Together with the reduced rate of taxation on dividends, there would be a revenue loss to the federal government. This would likely be at least partially offset through economic effects. In recent years moves in the TSX Composite Index have been highly correlated with the changes in the dividends paid by the companies in the index. A lower tax rate on dividends would make it more tax-efficient for companies to distribute earnings through dividends. They would likely raise dividends and that would provide greater investment income to individuals, in turn raising the revenues collected by the government.
If the double taxation of dividends were removed, some existing FTEs may wish to convert back to corporate status. However, in many cases this would trigger a taxable event. To facilitate the transformation, there should be a "roll-over" provision that would accommodate the switch in status for a prescribed period, say 5 years.
A cut to the corporate tax rate and a higher DTC would remove the bias against corporate distributions. If an entity wished to payout all of its earnings to shareholders it could do so in a tax-efficient manner through dividends. This would, not, however, completely eliminate the advantage of FTEs. The Canadian tax-deferred sector and foreigners would still prefer to receive income from FTEs because the income will not have been reduced by corporation income taxes.
The September 8, 2005 Department of Finance consultation document refers to a distribution tax on FTEs as a possible policy approach. TDBFG does not think that such a tax would improve economic efficiency or stem any revenue losses. First, it should be categorically stated that if steps are taken to remove double taxation of dividends then there are no grounds to have the Canadian taxable sector incur any amount of a distribution tax. The sector would unambiguously not be any part of revenue losses to the government. That leaves foreigners and the tax-deferred sector as possible bearers of such a tax. We will argue below that a tax effectively levied on these sectors would be counter-productive.
A superficial appeal of a distribution tax effectively passed to foreign investors in FTEs is understandable. Lowering the taxation of dividends in Canada would not alter the fact that a foreign investor from a country, such as the United States, with a treaty-determined withholding rate of 15 per cent, would remain attracted to Canadian FTEs over Canadian public corporations and U.S. corporations. If they bought a unit of a Canadian public corporation their income would be stripped of the Canadian corporate income tax and the withholding tax. If they bought a US. equity their income would be stripped, at a minimum, by the 35 per cent U.S. federal corporate income tax rate and the U.S. effective tax rate on dividends of 9.75 per cent (9.75 per cent being the corresponding "gross" tax rate after subtracting the 35 per cent corporate tax rate from the "net" dividend tax rate of 15 per cent - note as it is unclear how states apply taxes to income received from Canadian FTEs we have not included state taxation). So they would gain about a 30 percentage point edge by holding a unit in a FTE and paying only the withholding tax. (If the 15 per cent (net) US. tax rate on dividends is not extended beyond the current expiry date of 2008 then the American interest in Canadian FTEs may dry up because American investors would then need to pay additional U.S. taxes. The same would apply if the US. tax authorities deem that income received from Canadian FTEs does not qualify for the special 15 per cent rate). Hence, in order to benefit from the interest of foreigners, there would still an advantage to FTEs over corporations and hence an incentive to convert. There might be a temptation to just let this go if there was symmetry with Canadians' investments in the United States. In that case, Canadian investors would get an edge by holding American FTEs and we might consider the game a tie. But in general the US. does not have the counterparts to our FTEs. And at any rate, it would not be advisable for Canada to mimic what is likely a policy mistake on the part of the Americans of taxing trust distributions as if they were tax-paid corporate dividends.
As the withholding tax rate is set by treaty and Canada does not have the discretion to make a unilateral change, a distribution tax that is credible to Canadians is the only viable option for addressing the concerns above. At this point we must return to one of our original points of "do no harm". A distribution tax would inevitably reduce demand for FTEs and hence lower their market value. Further, it would provide a competitive edge to private LBOs, performed mainly by US. equity firms, at the expense of Canadian-made income trusts. Lower market values would wipe out the wealth of Canadians and cause a revenue loss to the federal government. Hence, we must be very careful that any policy move does not have the ultimate result of hurting Canadians. This really comes down to who has significant influence over the valuation of the FTEs. In the case of energy trusts this is likely foreigners. For energy trusts as a whole, foreign ownership is nearly 50 per cent and for some trusts it greatly exceeds 50 per cent. So if the distribution tax is very large, the market values would plummet and Canadians would be badly hurt. Foreign ownership of business income trusts is quite low. Therefore, a distribution tax on this form of FTE would not likely greatly affect market values, but then neither would it address the concerns set out by the Department of Finance. Foreign ownership of REITs lies between the situation of business income trusts and energy trusts.
If the advantage to an American investor of buying a Canadian FTE is about 30 percentage points over an American equity, then on the surface there would seem to be scope to levy a distribution tax without greatly impacting foreign demand. A 15 per cent tax, for example, would still leave half the original edge. But it is possible that any distribution tax could result in a reduction in Canadian welfare. In other words, the capital losses from lower market values would exceed the present value of higher federal tax revenues (which themselves would be lowered by the tax implications of the capital losses). A corollary of this point is that when one considers a broader welfare perspective, it is not clear that the holdings of foreign investors in FTEs is causing a revenue loss to Canadian governments.
The above arguments suggest that a distribution tax effectively targetted at foreigners would be counterproductive to Canadian interests. The same arguments can be made for a tax that would target the Canadian tax-deferred sector. Again the superficial appeal of such a tax is understandable. With the federal corporate income tax rate cut, the tax-deferred sector would enjoy about a 32 percentage point edge with income from a FTE (depending upon the province and its corporate income tax rate). However, as with the case of foreigners above, we must consider a broader context.
While foreign demand is concentrated in energy trusts, the Canadian tax-deferred sector likely looms large in determining valuations of business income trusts. If any distribution tax were not creditable to the tax-deferreds, market values could greatly fall and badly hurt taxable Canadians. If non-taxable entities essentially determine the market for business trusts and overall equity, such a distribution tax could raise the cost of capital for all firms. Further, any tax on income trusts might simply cause a substitution to entities avoiding corporate taxes by purchasing debt and using LBOs as a substitute for an income trust. A welfare loss would also be inflicted on Canadians if the pension funds substituted offshore holdings for Canadian FTEs.
A broader welfare context casts doubt on the benefits of a distribution tax on either foreigners or tax-deferred. Nevertheless, for the sake of completeness, we considered whether there are any additional arguments for or against a distribution tax on each of the 3 types of FTEs considered here. There are some additional considerations for not applying a distribution tax to REITs. First, their investments tend to be of a "passive" nature. Second, the REIT structure is common in other countries and without it here we would face a competitiveness issue in this sector. Third, while the Department of Finance estimates that REITs cost the federal treasury $80 million per year, the reality is that most real estate companies were managed not to pay taxes so this figure may be high. Fourth, if the pension funds had to pay a distribution tax, they might divest of REITs and purchase real estate directly. Their demand could exceed the total value of real estate available, leaving little left over for the REIT sector. That would hurt individual investors because it is difficult for them to have real estate holdings outside REITs.
As we indicated at the outset, great care must be taken in defining what the appropriate universe is for FTEs. For example, some instruments in financial markets have elements in common with FTEs but are essentially forms of securitization. There would be no rationale to apply a distribution tax to these.
There may be some resistance to lowering taxation of dividend income because it is skewed toward the wealthy. This is not a valid perspective for several reasons:
- While 55 per cent of dividend income goes to individuals with incomes above $100,000, 30 per cent goes to those with less than $50,000.
- Holding equity is one of the few genuine hopes for most Canadians to get ahead economically. Real after-tax returns on fixed income recently have been zero, at best. It is unfair to so heavily tax the financial hopes of so many Canadians.
- Reducing the taxation of dividends is not giving anyone something for free. We are merely talking about removing double taxation and letting Canadians pay taxes on dividends at a normal rate.
- Canadians did not protest the two cuts to the tax inclusion rate on capital gains that the Liberal Government introduced and capital gains are even more skewed up the income scale than dividends. Canadians realize they need to hold equities if they wish to get ahead.
Finally, a word on the option the Department of Finance put in the consultation document we have not addressed so far. We see restricting interest deductibility for FTEs as a non-starter. It took years of consultations and deliberations to come up with a proposed set of new rules in 2003 and they still have not been implemented. Restricting interest deductions seems like a form of thin capitalization rules that presumably would have to be applied to all businesses in Canada. TDBFG attaches a very low probability of success in crafting new rules that would address issues with FTEs without unintended and unacceptable consequences for other entities such as corporations with high debt loads.
The focus should squarely be on the bias in the current tax system against corporate distributions. To a large extent FTEs are a "market" solution to mitigate this distortion. Cutting the corporate tax rate, eliminating the capital tax and lowering the taxation of dividends to eliminate double taxation should be the heart of any policy approach. The cut in the corporate tax rate should be accelerated relative to the plan put in the February 2005 Budget and re-affirmed in the November 14, 2005 economic and fiscal update. The update's commitment to eliminate the capital tax for 2006 should be implemented.
There would still, however, be an advantage to FTEs and hence an incentive to convert because they would be more attractive to foreigners and the tax-deferred sector that Canadian corporations. The most effective, but also most expensive, way to address this would be through a "pure" integration of the corporate and personal income tax systems including a refundable dividend tax credit available to all Canadians, including the tax-deferred sector. No doubt consideration will be given to alternative, less expensive, approaches, at least on an interim basis toward this more definitive solution. Our sense is that any distribution tax, even if creditable to the Canadian taxable sector, would likely result in a loss of Canadian welfare. In other words, the net present value of any revenues received would be exceeded by the capital losses (which in turn would reduce the amount of revenues received).
FTEs have a legitimate place in the Canadian business structure and they would no doubt continue to exist and possibly to grow under the proposals in this submission. As such, consideration should be given to the governance rules, including what is necessary to better inform Canadians of the nature of their investments in these entities. Some FTEs may wish to switch back to corporate status if dividend taxation is reduced. A "roll-over" provision should be introduced for a finite period, say 5 years, to accommodate such a conversion which otherwise could trigger a taxable event.
A good part of the difficulty in designing any policy approaches to the concerns the Department of Finance has set out in the September 8th consultation paper is that FTEs bring to a head so many distortions in the Canadian tax system. Pulling a thread in one place could cause an unravelling elsewhere. For example, if the attractiveness of FTEs to foreigners and tax-deferreds is reduced, there would likely be a shift to the use of private equity where many of the same tax-reducing advantages are enjoyed. Even if a distribution tax addressed the advantage of FTEs for foreigners and tax-deferreds, there would still be the issue of differential tax treatment of debt versus equity. If Americans do not hold the amounts of Canadian energy resources they desire through FTEs, they will likely turn to direct foreign take-overs. Concern must be expressed about the possibility of any move to apply a tax on tax-deferreds for income from FTEs being the "thin edge of the wedge" that could ultimately undermine the foundation of this critical form of saving in Canada. For example, income from bonds is also tax deferred. The question must be posed of why action would be taken in the case of one asset class and not others.
The complexity of the issues around FTEs suggests that beyond moving quickly to lower corporate income taxes, eliminate the capital tax and remove double taxation of dividends, additional action must be put in the context of much broader tax reforms that will require additional study and consultations.