25 September 2017 CTF Finance Roundtable on 18 July 2017 Proposals
This summarizes the oral responses provided by two senior Finance officials at the Finance Roundtable held at the Infinity Convention Centre in Ottawa on 25 September 2017. The Roundtable occurred at the conclusion of a one-day conference sponsored by the Canadian Tax Foundation entitled "Tax Planning Using Private Corporations: Analysis and Discussion with Finance" respecting the 18 July 2017 proposals of the Department of Finance. The Department of Finance presenters were:
Miodrag Jovanovic, Department of Finance, General Director, Analysis, Tax Policy Branch
Brian Ernewein, Department of Finance, General Director, Legislation, Tax Policy Branch
This also briefly summarizes the questions posed at the Roundtable by Heather Evans (Executive Director, CTF), Cathie Brayley (Clark Wilson), and Mitch Sherman (Goodmans). Three other senior Finance officials presented at the opening session at the Conference, and their remarks are sumarized separately.
Q.1 – Effect on lower-bracket taxpayers
The rules appear to be based on the assumption that the shareholders are at the top marginal rate, and therefore there’s a significant arbitrage opportunity between corporate and personal rates – creating an incentive to defer tax by retaining and investing income in corporate solutions.
However, for shareholders at lower marginal rates, and depending on the type of investment income earned, the total tax liability is significantly higher than if the income were earned directly. Can Finance comment?
Miodrag Jovanovic: We recognize that the current refundable dividend tax on hand regime is quite effective at ensuring that there is a true-up of tax payable at the time dividends are paid. We recognize as well, however, that there are weaknesses in that system that result in a potential tax-deferral advantage. The question, therefore, is where to find the right middle ground – addressing these tax-planning strategies while also ensuring the proper level of integration across tax brackets.
We would like to highlight that, in the Consultation Paper, when we present examples of the result of the potential approaches, including the apportionment approach, we assume that the income attributable to the good source of financing (that is, income attributed or apportioned to holders’ contributions) is allowed to go tax-free to shareholders’ capital dividend account, instead of assuming that, in this case, the refundable dividend tax on hand regime would be kept, which would have allowed for better integration, or for a true-up at the shareholder’s level.
The fact that the Paper’s example suggests that we would use the CDA route rather than keeping the RDTOH on the good investments does not presuppose a preference. We are still open to questions as to how we can best keep the RDTOH system where appropriate, so we welcome any ideas.
(As a final point, we have other elements in the system already that, in order to prevent tax-planning opportunities, force the application of the top tax-rate in the context of trusts.)
Q.2 – Calculating deferral on investment income
Regarding the calculations on investment income, there has been a lot of modeling by various commentators, and some of the scenarios suggest that there is either a tax cost associated with earning investment income in a private corporation, or there is a negligible benefit that is only realized after an extended period in some provinces – sometimes more than a decade. Can Finance comment?
Miodrag Jovanovic: First, I think we agree that the tax-deferral benefit is presumably greater when the income used to finance the asset is coming from the eligible small business rate income and that, generally speaking, the tax-deferral benefit is significantly smaller where the assets are financed with general-rate income.
With respect to general-rate income, we also note that it depends on the length of the investment period and the rate of return, so we understand that if, for example, we start with 3% interest income, it may actually take five years before we reach the break-even point between either investing directly or through the corporation, and it is only after that where investing through a corporation becomes more promising.
The situation is slightly different for dividends than for capital gains because of the compounding effect where half of the gains would go tax-free through the CDA to the shareholders and there would be a greater initial amount to invest. Thus, capital gains invested through a corporation can show benefits after just a year or two.
For example, compare a five-year investment where the pre-tax rate of return is 4%. Investing through a corporation would yield an effective rate of return of 3.8%, while investing directly would yield 3% – and that is only after five years, so the difference could become quite significant over longer periods.
Another point worth raising is that, for most businesses, particularly those with a significant level of active income, with the application of the RDTOH and the lack of ordering in determining on exit whether there is access to the tax refund or not, most businesses that are distributing dividends at the same time as earning passive income are effectively able to get an ongoing tax rate on their investment income of 10% or 20%, depending on the nature of the income. For interest, it would be 20% on an ongoing basis – 50% minus the 30% refundable every year. Thus, the benefit for investing through a corporation could be significant.
Q.3 – Principles affecting the proposals
Where a corporation has a contingency fund or working capital for future investments, what are the factors that affect their characterization?
Miodrag Jovanovic: Let us first be clear about how the proposal would work with respect to capital gains. In a way, it would be a two-step approach.
The first step is to determine what the portion is of capital gains that would be associated with the “good pool” of shareholder contributions – the portion that would continue to receive the current treatment. In a way then, those capital gains (with capital dividend account) would receive the current tax refund.
The second step would be to determine whether there any particular types of assets that would also be excluded, regardless of whether they are financed with low-tax business income or financed with shareholder contributions. We raise that question in the Paper – there is an example where a holdco disposes of the shares of an opco that it controls, and the Paper states that there are good reasons to exclude that from the rule [denying a CDA addition].
For us, that example is the starting point for a discussion. The example assumes that you need control. The question is whether control is too high of a test. There are other tests in the tax system, such as 10% ownership, which is used for FAPI purposes. Regarding shares, there is also the question of what kind of control or ownership test is appropriate. Additionally, there is the question of whether we should distinguish between capital gains realized on assets that were used in the business directly and assets that were passively held.
We think that these are good arguments being put forward, and we welcome further comments.
Q.4 - Grandfathering
What comments can Finance offer about grandfathering of existing assets?
Miodrag Jovanovic: The government has stated recently that the intention here is to not have any impact on existing savings, or investment accruing on those savings. This leads to questions about how the government can ensure that outcome – how to provide grandfathering rules to protect those assets and investments and any compounding returns therefrom.
This is another important question for us. We are looking into options, but I do not want to get into the details right now. To give a general example of those considerations, though, there is a question as to whether we should allow corporations to segregate their assets into a distinct corporation in a way that would allow the distinct corporation to retain the existing treatment. For that to work, there would also have to be some kind of “walls” around that distinct corporation, or a system for dealing with the movement of funds between those corporations.
Again, we are continuing to consider these issues, and welcome further comments and ideas.
Q.5 – Complexity of TOSI
Does the complexity of policy objective of being fair warrant the complexity of the expanded tax on split income (TOSI)?
Brian Ernewein: A few observations: First, I think in fact the Consultation Paper acknowledged this issue. It identified this “evidentiary point” of this factual inquiry as one of the challenges of the sprinkling proposal. When we made the change in response to Neuman in relation to the so called “kiddie tax” back in 1999, it was limited in that respect because of the challenges we perceived with respect to adults where presumably would have a greater range of possible contributions to the business.
Having said that, it is not beyond the wit of humankind to price capital or labour. Markets do it millions of times each day. In this particular case, the question is not about coming to a price to the penny. First of all, we are dealing with a cap – you can always pay less, it is a question of not paying too much. Also, while this is essentially a factual question, there may be things that we and the CRA can do to help people conduct this inquiry. It might be possible for further legislative help or rules on this point, and you might want to think about what sort of submissions you might make to us. The Explanatory Notes may also give some guidance about how we think this inquiry would be conducted. CRA can almost certainly perform a role in this by offering guidance on the criteria they’d be looking at in terms in determining whether compensation or a dividend in lieu of compensation is reasonable.
The question was whether the policy objective warrants the complexity. We think the policy concern warrants a response. We also think that we could not reasonably invoke some of the responses that we have used in the case of minor children, such as disallowing dividends or taxing all dividends made to adult family members at the top marginal rate.
(Other approaches, like the Subchapter S approach, have also been raised in the passive income context. To digress for a moment, I am unsure how the Subchapter S flow-through approach would really answer the passive question, inasmuch as anybody subject to tax at over a 27% personal rate would still presumably want to leave the income in a corporation, and pay on the 27% rate until a possible late distribution. In the context of sprinkling, however, you still have to figure out where the income should go to, so it does not represent a likely solution here.)
We do think the policy concern warrants a response, and we do think the proposal we have is appropriate.
Q.6 – Difference in treatment of family transactions
The proposed rules could discourage certain types of transactions – for example, different tax-treatments for distributions to relatives vs. friends, or selling a business to a U.S. company rather than to a niece. Could Finance comment?
Brian Ernewein: We do not consider it to be exceptional to have different rules for family members or for people not dealing at arm’s length – it is more the norm than the exception in relation to valuation or other similar kinds of considerations. S. 67 tests the reasonableness of salaries in these circumstances; we have income attribution rules that apply only between family members generally, because we do not see much of a widespread concern with people giving money away to strangers; we have rules for shareholder loans; and so forth.
Conceptually, we are not breaking new ground with this. There are different issues regarding what tests are required, and there is greater complexity for dividends than for salaries, as there can be a lag when dividends are paid, and I do not mean to downplay that complexity. Nevertheless, there are many situations today where non-arm’s length situations need special handling.
We do not think that someone foregoes hiring their spouse because of the need to justify their salary, nor for hiring their children. It does not strike us that a family’s decision about whether or not to carry on their business, or a person’s decision about whether to invest in a company with another family member, would realistically change because they might need to justify the return that a family member gets.
Q.7 – Comparison of friends in business vs. family members
Cathie Braley: Consider this example: Suppose that there is a year with strong financial results, and the shareholders want to pay out a dividend. The two shareholders both get 50% of the dividend.
Is there any difference in the function that the CEO provides compared with the Manager of Customer Relations and HR? Does that invite questions about whether the 50-50 distribution of dividends is appropriate?
Brian Ernewein: I think the framework for the analysis is right. It is a question of examining the contributions that each has made, and boils down to whether there is any difference in the labour or sweat equity contribution. It is not necessarily the case that the CEO provides more value, it would depend on the facts.
I should touch on an important related point. In the case of a start-up, particularly in the case of venture capital, the first round of funding is often from friends and family. We were given an example where son asks his mother for $100,000 to put into the company, and the assertion that we have heard is that there is essentially no ability to pay the mother a return on those shares, either by way of dividends or capital gains.
We do not see any basis for such an assertion. The mother has put her money in a high-risk investment and is entitled to a high rate of return reflecting the riskiness of that investment. The same reasoning applies to investment by friends. The risk in these investments is not proof-positive, but is very compelling evidence of the reasonableness of high returns.
Q.8 – Pipeline transactions
- Has the Department considered whether a carve-out from the proposed rules could be made for pipeline transactions?
- If no wholesale carve-out is provided, would the Department consider relaxing the requirements of s. 164(6) to more easily facilitate estate-planning?
- Many estates created before the Announcement Date fully expected to be able to implement a pipeline transaction and now may be precluded from doing so – could Finance comment?
Brian Ernewein: These are all the right questions, and they have been put to us on more than one occasion since July 18th.
Let me be clear – I do not intend to make pronouncements on any of these questions today. That necessarily awaits the conclusion of the consultations and the government’s assessment of the input we have received. However, I will make some observations.
Regarding the first question, it is difficult to see how the current results obtained using pipelines do not invoke the same policy concerns that other non-arm’s length transfers do – for example, the inter vivos 84.1 transfers that are more explicitly covered in the changes to s. 84.1. I get the point – that death is a steep price to pay in order to get a lower tax bill – but death is also inevitable, and people certainly do make plans that take death into their planning equation.
On the premise, therefore, that there will not be a complete carve-out for pipelines, what other changes might be appropriate? One of the first things we have identified is the double-taxation that can arise – where you could have tax on the deceased and dividend tax on the wind-up by the estate. That is probably not a result we are trying to achieve.
The point has also been made that there is a provision to integrate the deceased’s capital gains tax with the distribution tax paid by the estate through the mechanism of s. 164(6) – of course, only within the first year and subject to some constraints. The question is whether or not that might be the mechanism that we should consider more generally to respond to the “capture” of pipelines by the proposals. If so, should we allow more than 12 months? We understand that it is not always possible to get that done within 12 months. The period would not be unlimited, but this is an area worth thinking about.
Finally, regarding the transition, I think the best example is where the death has occurred before July 18th, s. 164(6) is not available or the clock has run out, and the estate was counting on using a pipeline to alleviate its taxation, or to wind up the corporation, effectively paying only one level of tax. I think there is also reason to consider the application of a transition in those circumstances. More precisely, it is not necessarily the right answer to have the capital gains tax paid by the deceased and the second layer of tax paid by the estate on the distribution. Whether that means that we should just accept the capital gains tax in that case, or whether we should provide integration for it with the dividend tax, still needs to be discussed.
Q.9 – Effects on intergenerational transfers of family businesses
How does Finance intend to balance the need to avoid surplus-stripping with the possible effects of deterring transfers of a business within the family?
Brian Ernewein: I do not consider our changes to s. 84.1 to represent a policy-change. I know there is a lot of sensitivity to the “loophole” label, but it fits what we are doing here.
S. 84.1 has been around for a long time. It seeks to treat as a dividend non-share value extracted from a corporation through a non-arm’s length sale. It would apply when I sell my company to another company and I take back boot, and I cannot see a policy reason justifying a different result just because the extraction takes place in two steps – first by selling and stepping up ACB, and second by selling again and extracting the boot.
Where there is a material difference in the tax rate between dividends and capital gains, I think one of three things has to happen. You either have to get the capital gains rate up to the dividends rate, you have to get the dividend rate down to the capital gains rate (which raises questions about whether you are then over-integrating corporate income tax), or you have to do something to limit the ability to convert one into the other. I am not aware of any plans to increase capital gains rates or to lower dividend tax rates, so I think the system has to adjust and try to maintain or reestablish the rules to prevent the conversion of surplus into capital gains.
To get to the question you posed, s. 84.1 does not apply between an arm’s length sale and a non-arm’s length sale of a corporation when it is a sale from an individual to another family individual. That may be an obvious point, but it is worth emphasizing that a parent can sell the family company to the child and claim capital gains treatment, including the LCGE, in the same way as on a non-arm’s length sale. There is also a difference when the parent wants to sell not to the child, but to the child’s corporation. To the extent that the parent takes cash or other non-share consideration back, those proceeds under s. 84.1 would be recharacterized as dividends to the parent – and that contrasts with the sale of a corporation to an arm’s length company which generally will be treated as a capital gain.
I think this distinction and treatment is defensible. Again, if there is a difference between dividends and capital gains with no constraint, the parent in my example would almost certainly set up his or her own corporation to sell to in order to take the dividends out in the form of capital gains. If we only prevented the person themselves from doing that, then the parent would use the parent’s spouse; if we prevented that between spouses, then we think the couple would seek to employ their children or another non-arm’s length person.
Therefore, I do defend s. 84.1 generally. What I think that does leave is the point that we have raised in the Budget, and again in the Consultation Paper itself, as to whether there is any scope for relief on s. 84.1 on what we have vaguely called “legitimate” intergenerational transfers of family corporations. We still struggle with whether it is possible to do something in this space without creating pretty significant revenue losses – and when I say pretty significant, I am talking in the hundreds of millions or perhaps a billion dollars. There was a private member’s bill in the House of Commons voted on earlier this year that sought to accomplish something along these line, and our own revenue estimate was half a billion to a billion, plus cost. The parliamentary budget office made its own estimate, that was also in the hundreds of millions of dollars.
The Consultation Paper also identified some rules in the U.S. that relate to this area. Essentially it requires you to divorce yourself or the parent to divorce themself from the company in order to support the characterization of the sale of the company as a capital gain rather than a distribution. We are of the view that the U.S. rules are too tough. They do not really fit the family model, where you often want to transfer the property on a transitional basis, perhaps in part, and stay involved with the company. Our conundrum is that if we make the rule less stringent, it is hard to tell stripping from a genuine sale.
Thus, we need to reflect further on what possible approaches we can take. I think some of what we heard today was really interesting - whether or not you can do something to lift underlying capital gains and not apply the deemed dividend distribution to that extent. I was reminded when I was looking at this on the weekend of the rules that we have in our foreign affiliate regime, (since we seem to be tapping into those anyway, according to some of the speakers here today), and the s. 93(1) election for sales that allow you to recharacterize as a deemed dividend, which is interesting. Again, you have to turn that around, reflecting the difference between dividends and capital gains for individuals versus corporations, but I think that is interesting as well. All useful suggestions, and I am hoping that the tax community will develop these further in the commentary they provide us with before the consultation period ends.
Q.10 – Scope of s. 246.1
The text of s. 246.1 is broad, and the explanatory notes do not provide much guidance. There is a concern that even very basic transactions, such as the return of invested capital, or the payment of a CDA dividend even when the CDA dividend is sourced from the sale of arm’s length sales of marketable securities, might be subject to rules. Could you give some examples of transactions that should clearly be caught by s. 246.1, and some examples which clearly should not?
Brian Ernewein: First, I should explain how we got to s. 246.1.
We had the s. 84.1 gap, discussed above. We also had the impression that the courts were not observing the scheme of the Act to prevent surplus-stripping, or at least the conversion of dividends into taxable capital gains. Having addressed the two-step transaction problem in particular, we were of the view that the courts might permit other similar transactions. For example, an individual not selling a corporation to another corporation but selling an interest in a partnership that owns a corporation, or a corporation not being the buyer, but a partnership owned by the corporation is the buyer. Perhaps these are caught today, but the fisc would have a challenge if tax-planning were based on the notion that a taxpayer is only caught if they are within the four corners of s. 84.1.
To give a more substantive example, suppose that you have a company that is sitting on cash, you would like to take that money out, but it would be subject to tax at over 40% as a dividend. Instead, you drop that company down into another company, triggering a capital gain. Maybe even bring that money out with your capital dividend account, and the other half comes out taxable effectively at a 25% rate.
Those types of situations are not literally caught by s. 84.1, and we were not certain they would be caught by GAAR or other rules. That is the backdrop to s. 246.1.
Let us move on to examples on the other side. Suppose you have owned marketable securities in your company forever – you have owned the company forever, the shares have been there forever, they have appreciated in value, and you sell them. Can you pay up a capital dividend? Yes. We do not intend to interfere with that kind of situation.
Another one, which is a little bit more involved, is where it is already within the corporate group, and then you want to drop down to trigger a capital gain as part of a restructuring in an arm’s length sale. Yes, I think that is an example too. But there are, as I say with my first example on CDA, there are I think some nuances that make it difficult for us to map out in a sound bite where the rule starts and stops.
Q.11 – Non-ancillary s. 246.1 application to NAL transactions
Some commentators have suggested that the rule could inappropriately apply to bona fide arm’s length situations. For example, suppose a purchaser sets up an acquisition company, and gives a note or borrows money to pay the purchase price, and later uses the earnings or assets of the target in order to repay its own borrowing. Can Finance comment?
Brian Ernewein: The general answer is, the surplus stripping changes are intended to focus on non-arm’s length transactions. We are generally not targeting the arm’s length purchaser refinancing example in the question.
I want to offer some qualifications. We do not think that the existence of an arm’s length sale in a series of transactions is intended to bless any non-arm’s length elements in the series. To take kind of a silly example, imagine that I have a company worth X, I take out 90% of X through some transaction that technically meets s. 84.1 and then I sell the remaining 10% to an arm’s length purchaser. The arm’s length sale ought not to bless the 90% strip.
The second thing is that s. 246.1 is not intended merely as a backstop to s. 84.1. We intend to import into the Act a scheme in terms of non-arm’s length surplus stripping. So, for example, if a taxpayer says “I really wasn’t seeking to beat 84.1, this was more like an 84(2) avoidance,” then that is not given a free pass. The same policy considerations would inform the application of 246.1 to that situation.
Q.12 – Implementation
What can Finance tell us about the next stages of advancement?
The first point is that we will not be considering our advice to the government, in terms of its own response, until we have actually closed the consultations and had a chance to receive and review the submissions. I do not know precisely how the legislative proposals – I’ll focus on income sprinkling for the moment – will go forward – that is to say whether the next step is another draft, tabling, or something else. We are aware of the coming-into-force date of 1 January 2018, and we also recognize that, if we are proposing to make any changes or think that any changes are warranted, it is important that we communicate that before they are intended to have effect. So I hope you will take some comfort in that point.
On 84.1 and 246.1, they of course have a July 18th coming-into-force. I have grouped comments on those provisions into two groups. We need to land in the short term on transition rules regarding pipelines, for example, and perhaps other issues which people might raise with us in terms of transactions straddling the July 18th date or happening soon thereafter, that are in their mind caught by either rule and ought not to be. There is that sort of short-term pressure that we need to review the situation and make appropriate adjustments.
There is also the longer term – not years, but more than the next couple of weeks or months – where we need to consider whether to add exceptions or refinements to the rules. This is not just an examination of the changes we made to 84.1, but rather a holistic review of how s. 84.1 has been applied for a long time.
Finally, on the passive income proposal, I should emphasize there is no proposal in the July 18th package. We had draft legislative proposals in relation to sprinkling and in relation to character-conversion but not in relation to passives. It was essentially a diagnostic of what we viewed as a problem. We discussed a couple of approaches, an old one and a new one that we have not seen before, but there is no legislative proposal and no coming into force, so the next stage would necessarily be to take in comments to see what sort of proposal might be developed, and whether the government is interested in advancing that, and to move to sharing that with the public if and when that decision is taken.
Additional comment on tax advantage computation
Miodrag Jovanovic: I have received comments that dispute our calculations on capital gains from large corporations and the tax advantage from using a private corporation, so I will take a moment to explain our approach.
We compare two individuals – one is an unincorporated professional (“Brian”), the other incorporated (“Miodrag”). They each have enough money to put aside $100,000 of earnings.
Brian puts aside $100,000, after paying personal income tax, he will be left with $46,500. He decides to invest for 10 years in RBC shares, with an expected 6% rate of return in the form of capital gains. His rate of return on that 6% pre-tax rate would be 4.7%.
Miodrag’s income within the corporation is taxed at a general rate – let us assume that leaves $73,500 to invest. He buys the same RBC shares and holds them over 10 years. He distributes the portfolio to himself after 10 years as an eligible dividend. His benefit comes from the half-inclusion rate when determining the amount of capital gains at the corporate level, and the other half will go tax-free to him through the capital dividend account. In this example, his effective rate of return is 6.4% compared with Brian’s return on the same starting amount of $100,000 – in fact, the amount is even above Brian’s 6% nominal rate of return, meaning that Miodrag has received an implicit subsidy.