It has been just over a month since the Federal Court of Appeal rendered its decision in JP Morgan Canada Asset Management v. AG Canada. In the wake of that decision, I thought there would never be another judicial review application in the tax context.
By way of background, decisions of federal employees (such as those within the Canada Revenue Agency) can be subject to judicial review under administrative law principles as a remedy of last resort. In the tax context, judicial review applications are common in decisions on interest and penalty waivers, and can encompass other types of decisions as well. The JP Morgan case resulted from a challenge to the CRA’s decision to not follow its own policies.
It turns out that those predicting the demise of the judicial review remedy were premature. This week, the Federal Court released its decision in Conocophillips Canada Resources Corp. v. Minister of National Revenue Result: taxpayer was successful on a judicial review application.
The facts are simple. The CRA alleged that it sent a Notice of Reassessment to the taxpayer. The taxpayer denied having received it. When the taxpayer filed its Notice of Objection, the CRA said, sorry, you’re out of time, we can’t consider your Notice of Objection. The taxpayer sought judicial review of the CRA’s decision to not consider the Notice of Objection, saying that the Federal Court had jurisdiction to consider the decision, and the decision was unreasonable in light of the fact that CRA could not prove when the Notice of Reassessment was in fact mailed. Essentially, the taxpayer was trying to force the CRA into considering the objection on its merits, even if it resulted in the CRA confirming its audit position. Then, at least, the taxpayer would have recourse to the Tax Court of Canada by way of appeal.
The Crown’s counter-argument was that the taxpayer was already free to appeal to the Tax Court. The Tax Court then had jurisdiction to decide whether the Notice of Reassessment was properly sent, in time, to the taxpayer.
The Federal Court agreed with the taxpayer and found, firstly, that it had jurisdiction to undertake a judicial review because the taxpayer had no other adequate legal remedy (by way of Tax Court appeal), and secondly, that the CRA’s decision was unreasonable. The judge expressly considered but distinguished the JP Morgan decision. In my view, the Crown will almost certainly appeal this decision to the Federal Court of Appeal on the jurisdictional issue.
The so-called “foreign affiliate dumping” rules are now a reality, notwithstanding the heavy criticism they received from many quarters. A Quick Update Report provides both a high level discussion and detailed look at these comprehensive new rules.
In 2013-0507881E5, the CRA usefully confirmed its view that a payment made pursuant to a valid price adjustment clause – in respect of preferred shares that were redeemed before the payment – is taxed as a dividend in the year of the payment. In the course of asset transfers within a corporate group, it is common that preferred shares of the acquiring company be issued as consideration in order to access a rollover under s. 85(1). Such preferred shares have a redemption amount that is typically set at a bona fide best estimate of the value of the property transferred, which is subject to adjustment if this value is later determined to be wrong (see Income Tax Folio S4F3¬C1, Price Adjustment Clauses). If the preferred shares are later redeemed, the subsequent adjustment may come as an additional payment to the former holder. In 2013-0507881E5, the CRA considered a case where the preferred shares were issued in year 1 and redeemed in year 5. A payment was made to the former holder (of the preferred shares) in year 7 pursuant to a valid price adjustment clause. The CRA confirmed its position that such a payment would be taxed as a dividend to the former holder in the year of the payment (i.e., year 7), such that no penalties or interest would apply.
In 2013-0491571E5, the CRA confirmed its view that a “partnership interest” is one property notwithstanding that a partner may hold separate “units” in the partnership. Accordingly, where a partner sells only some of its units in the partnership during a fiscal year of the partnership, the effect is “the disposition of part of a property”. As such, a portion of the total adjusted cost base (ACB) of the one partnership interest must be allocated to the part sold, being the portion “that can reasonably be regarded as attributable to that part” (under s. 43). This likewise reduces the continuing ACB of the one partnership interest at that time (under s. 53(2)(d)). Finally, because the partnership interest is one property, 100% of any income allocated to the partner at the end of the partnership’s fiscal year increases the ACB of the (one) partnership interest immediately after the end of the fiscal year (under s. 53(1)(e)(i)).
The Supreme Court of Canada has, in a unanimous judgment rendered by Justice LeBel, ruled in favour of the taxpayers in two tax-related contractual rectification cases arising under the Civil Code of Quebec (Services Environnementaux AES and Riopel).
Both cases involved efforts to remedy contracts implementing tax-motivated reorganizations where the taxpayers argued that the documentation did not accurately reflect the parties’ intentions. Revenue Quebec raised reassessments against the taxpayers based on the erroneous documentation. The taxpayers, in response, sought to rectify the contracts to accord with what they said was their original intention.
The Supreme Court reaffirmed the principle that the Tax Court of Canada has exclusive jurisdiction to determine the correctness of a tax assessment. However, it also noted that tax law applies to transactions governed by, and the nature and legal consequences of which are determined by reference to, the common law or the civil law. Therefore, one must first determine the common law or civil law relationships before applying the tax law. Here, the taxpayers were able (with the assistance of a Court order) to recharacterize the civil law relationships as being different from those resulting from the erroneous documentation. Now, armed with the rectified contracts, the taxpayers will return to the tax authorities and endeavor to overturn the tax reassessments.
Interestingly, the Court did not appear to view the revenue authorities as having acquired any rights by virtue of the tax reassessments issued in reliance on the initial documentation. In paragraphs 49 and 50, the Court states: “If there had been a dispute between the parties about the nature of their intention or if third parties had acquired rights in relation to the legal situation created by the acts, the law of evidence in civil matters would have placed certain obstacles in the way of this interpretation exercise … There is no such obstacle in the cases at bar.”
The Supreme Court of Canada expressly refrained from (despite an invitation to do so by the revenue authorities) “criticizing, approving or commenting on” rectifications in the common law context as that remedy has developed since the Ontario Court of Appeal decision in A.G. Canada v. Juliar. Thus, while theoretically leaving the door open to reconsidering Juliar in the common law provinces, there is no reason to think that the Juliar line of authority is weakened by this decision.
In Auto Parts Inc. Lecavalier v. The Queen, 2013 CCI 310 (English version not yet available), the Tax Court of Canada (TCC) applied the general anti-avoidance rule (GAAR) to effectively re-characterize a cash share subscription followed by cash debt repayment as a direct equity-for-debt exchange – resulting in debt forgiveness. Very briefly:
- A US parent company (US Parent) had capitalized its Canadian subsidiary (Canco) with both equity and debt.
- The value of Canco’s assets subsequently declined well below the debt: the asset value was approximately $10m; the debt (held by US Parent) was approximately $24m.
- Prior to the sale of Canco’s shares to a Canadian buyer (Buyer), US Parent injected the approximate $14m shortfall in value as a further cash subscription for additional shares of Canco. Canco immediately thereafter used this cash to pay down $14m of the debt owed to US Parent. (The newly issued shares effectively had no value after this $14m debt repayment.)
- US Parent then sold to Buyer both its shares of Canco (for $1) and its $10m debt of Canco (for approximately $10m).
The CRA reassessed Canco on the basis that the $14m share subscription and $14m debt repayment prior to the sale to Buyer were avoidance transactions under the GAAR (s. 245). Furthermore, these avoidance transactions abusively circumvented both the “debt parking” rule (in s. 80.01(7)) and the “shares-for-debt” rule (in s. 80(2)(g)). The TCC agreed. Accordingly, the $14m share subscription and $14m debt repayment were effectively ignored – giving rise to $14m of debt forgiveness in Canco under s. 80. This forgiven debt reduced Canco’s available tax attributes to zero and resulted in a 50% income inclusion for the balance (under s. 80(13)).
Two quick additional observations:
- Canco argued (in part) that the transactions were structured by US Parent primarily for US tax purposes, and as such, were not avoidance transactions from a Canadian tax perspective. However, Canco failed to introduce direct evidence from either US Parent or a US tax expert in support of this contention.
- The avoidance transactions did not involve the stop-loss rule in s. 40(2)(e.1) from the creditor’s perspective. In that context, an important question arises: Are pre-sale debt restructure transactions that rely on the CRA’s Advance Income Tax Ruling ATR-66 (and similar transactions) affected by this decision? Such transactions generally involve a transfer of debt internally within the existing group before the share sale to an arm’s length buyer. No debt forgiveness arises and the CRA sees no abuse as a policy matter. Arguably such transactions should not be affected by the decision in Auto Parts Inc. Lecavalier as a policy matter. That is, the CRA has previously accepted such transactions because the creditor does not access the inherent loss in the system (directly or indirectly). In Auto Parts Inc. Lecavalier, the creditor (i.e., US Parent) appears to have accessed the inherent loss in the system – albeit on the shares of Canco sold to Buyer. Accordingly, the decision in Auto Parts Inc. Lecavalier may well be distinguishable on that basis.
The Federal Court of Appeal (FCA) recently released its reasons in MNR et al v. JP Morgan Asset Management (Canada) Inc.¸ 2013 FCA 250. In lengthy reasons, the FCA takes the opportunity to express its exasperation with the many applicants who “time and time again” come to the Federal Court in a steady “flow of unmeritorious applications for judicial review in the area of tax.” The decision sounds the death knell for taxpayers seeking recourse for alleged Canada Revenue Agency (CRA) improprieties by means of judicial review in the Federal Court.
In general terms, the legislative scheme for tax matters in Canada bifurcates jurisdiction between the Tax Court of Canada, which has exclusive jurisdiction to deal with appeals of assessments, and the Federal Court of Canada, which has jurisdiction over matters relating to decisions and orders made by federal tribunals except to the extent that those matters are within the jurisdiction of the Tax Court.
In JP Morgan Asset Management, the FCA clearly states that the Federal Court does not have jurisdiction to address CRA misbehavior in respect of the issuance of an assessment. The reasons are two-fold: Firstly, the decision to issue an assessment and the content of the assessment (namely, the amount payable or determined) are not matters of ministerial discretion (see paragraphs 77-80). Secondly, a judicial review is a method of last resort that may only be invoked where no other avenue for relief is available. In this regard, the FCA states that the Tax Court appeal process provides effective relief where an assessment is incorrect, where evidence should not be admitted to support an assessment, where there is an abuse of the Tax Court’s process, or where there was a procedural defect in the process of assessing. If a taxpayer cannot or does not obtain relief through the Tax Court appeal process, a taxpayer may obtain relief through the provincial civil courts for such improprieties as negligence, fraud or misfeasance (see paragraphs 81-90).
The FCA’s reasoning will be unsatisfactory to tax litigants and practitioners, many of whom will strongly disagree with the FCA’s view that adequate alternatives are available to taxpayers and, in particular, with the assertions that the Tax Court’s process is adequate to remedy procedural defects in the raising of assessments and that civil litigation in the provincial courts is readily available to taxpayers. The steady stream of judicial review applications arguably demonstrates that tax litigants, on the advice of well-seasoned practitioners, are not able to obtain relief through the Tax Court’s or the provincial civil courts’ processes.
In tax disputes, the Crown is privileged to rely on assumptions in lieu of evidence in the raising and defending of an assessment. It is also able to prioritize its considerable resources as it sees fit at a time when the cost of litigation poses an insurmountable barrier to access to justice for most ordinary Canadians. These factors, among others and coupled with the well-established legal principle that the process of assessing is distinct from the assessment itself and that CRA (mis)conduct is irrelevant to the correctness of the assessment in dispute, generally operate to foreclose relief for CRA impropriety via the Tax Court’s and provincial courts’ process.
The issuance of an assessment that is subject to the Tax Court appeal process is the primary means by which CRA conduct has legal effect. As a result of the decision in JP Morgan Asset Management, the last remaining door to judicial review has been firmly closed in the vast majority of tax disputes.
Normally, the CRA cannot collect any amount of reassessed income tax, interest or penalties while the reassessment is under objection or appeal (although CRA can collect half the reassessed amount, if the taxpayer is a large corporation). However, if there are reasonable grounds to believe that tax collection is at risk, the CRA can move for a jeopardy collection order and, if successful, then take immediate steps to collect the full reassessed amount. Jeopardy collection orders can be made on an ex parte basis, i.e. by the CRA to a judge of the Federal Court, without notice to the taxpayer.
In the Federal Court’s October 28, 2013 decision in Tassone v. Canada (National Revenue), a husband and wife sought to overturn jeopardy collection orders made against them on the basis, among others, that CRA had not fully disclosed all material facts in obtaining the orders. Material non-disclosure is a stand-alone ground for overturning a jeopardy collection order even if the facts otherwise support the CRA’s position. In addition, the CRA had doubled up on the reassessments, whereby the same income was assessed to both husband and wife because the CRA was not certain who should report it.
The Court rejected the taxpayers’ position, finding that the CRA had fully disclosed all material facts. There was some evidence that the taxpayers had funnelled assets through an offshore jurisdiction, using the same company that the husband’s uncle had utilized for the same purpose a few years earlier.
However, the Court does not focus on the impact of a jeopardy collection order in the context of the relatively recent, and all-too-frequent, practice of the CRA in raising alternative and mutually exclusive reassessments. Here, both husband and wife were reassessed to tax on the same income inclusions, each totalling about $3.6 million in tax, interest and penalties. The CRA conceded that the same income had been assessed to both husband and wife, such that the total owing under the two sets of reassessments was twice what would ultimately be owed, even on a worst-case scenario for the taxpayers. The CRA was effectively seeking to immediately collect twice the tax owing.
Unfortunately, for the purposes of overturning a jeopardy collection order, the current legal test presumes that the reassessed tax is properly owed even when, as in this case, at least half the amount of the reassessments clearly cannot be upheld on any possible version of events.
In the Canadian domestic tax context, an accrued gain on shares held by one Canadian company in another Canadian company may be safely reduced by the payment of an inter-corporate dividend provided the gain so reduced is attributable to underlying previously-taxed income (“safe income”, s. 112 and s. 55(2)). The reduction of a gain on shares by an inter-corporate dividend is sometimes referred to as “capital gains stripping” (or a “safe-income strip”, where the dividend is limited to the underlying safe income). The basic tax policy is that underlying safe income has already been taxed once at the corporate level; such income should therefore pass tax free from one Canadian company to another without any further Canadian corporate tax. Interesting safe income results can arise in a chain of companies. For example, the accrued gain on shares that a particular corporate shareholder (Company A) owns in another company (Company B) may be attributable to both (i) the underlying safe income in Company B and (ii) the underlying safe income in a subsidiary of Company B (Company C) – provided that both of these (safe income) amounts can reasonably be considered to contribute to the gain on Company A’s shares of Company B. At the very same time, the accrued gain on shares that Company B owns in Company C may also be attributable to the underlying safe income in Company C.
In D&D Livestock Ltd. v. The Queen, 2013 TCC 318 the corporate group exploited this concept of safe income within a corporate chain. The equivalent of Company A above managed to co-ordinate a complex series of stock dividends and share rollover transactions (within the group) which allowed it to, in effect, take advantage of Company C’s safe income twice, and thereby reduce a capital gain on a subsequent sale of shares to an arm’s length buyer. The Tax Court recognized that the use of this safe income twice defeated the general purpose of the rule that permits the use safe income to reduce the gain on a share (s. 55(2), see paragraph 33). However, based on the precise steps undertaken by the group – which resulted in safe income dividends on shares of both Company B and Company C – no ambiguity arose when it came to applying the actual words of s. 55(2). In the absence of such ambiguity, the Tax Court said it had no authority “to simply re-write the subsection to give effect to its purpose”. The Tax Court further observed that if the government finds such transactions to be abusive, it can always attack them using the general anti-avoidance rule (s. 245) or recommend that Parliament amend the Act.
Is aggressive tax avoidance immoral? Some people think so, but I don’t. (I’ve been on this topic before – see my blog post, June 5, 2013, “Tax Planning, Morality and Cowboys”). The bottom line for me is this: if the rules as written allow for the reduction of taxes through legally permissible planning, then no one should be heard to say that doing so is somehow immoral.
Most critics of aggressive tax planning argue that a literal reading of the Act is somehow not “right” if it results in what they think is an unintended tax benefit. Instead of relying on the words as written, they would have us go behind the words to surmise what Parliament really intended when enacting the provision in question. My response is that if we believe in the rule of law, we should be very wary of buying into a system in which the obligation to pay a particular amount of tax depends on what someone feels should be right as opposed to one in which the obligation is determined by the words Parliament actually uses.
I got to thinking about this again when for some reason The Merchant of Venice came to mind. Shakespeare’s play tells the story of a kind business man (Antonio) who guaranteed a loan to a friend from a local moneylender (Shylock). The terms of the loan called for no interest, but included a tough forfeiture clause – a pound of Antonio’s flesh – if the principal was not paid on time. Antonio was unable to pay the loan when it came due and Shylock, relying on a strict interpretation of the loan agreement, went to court demanding his pound of flesh. Antonio was defended by what today we might call a clever tax lawyer (Portia). She made the case that Shylock was indeed entitled to a strict interpretation of the agreement, but pointed out that the agreement referred to a pound of flesh. No mention of blood here, so Shylock could only realize on his security if he could shave off a pound of flesh without drawing blood. The court agreed with Portia’s interpretation and, as I remember my reaction the first time I read the play, it was “awesome!” I doubt that I was alone in thinking so.
Fast forward to the assertion that it is immoral to rely on the actual words of the Income Tax Act in assessing the consequences of tax avoidance planning. I can’t imagine anyone suggesting that Portia’s defense was immoral because it relied strictly on the actual words used in the agreement. At least in the context of the play, we applaud her cleverness in saving Antonio this way. I’m not suggesting there is an exact parallel between the interpretation of a public statute like the Income Tax Act and the terms of a private agreement, but I do suggest that the way we react to Portia’s defense is instructive. There is a basic fairness in the proposition that where the meaning of the words is clear, they should be applied as written.
Some will nonetheless argue that it seems unfair to apply the plain meaning if the result exploits a gap in the Act. While I recognize that some may feel this way, I suggest that the emotional reaction is irrelevant in a system based on the rule of law. And I should point out that strict interpretation does not always favour the taxpayer. Any experienced tax professional knows of cases in which a court regretfully dismisses a taxpayer’s appeal saying something like: “The result in this case is patently unfair to the taxpayer but, unfortunately, this court is bound to apply the law regardless of any unfairness this may cause.” See the OK Payday Loans Inc. case, 2013 TCC 17, for a recent example.
Unfairness can (and does) work both ways when the subject matter is the application of a provision of the Act. For me, unfairness (of either kind) has nothing to do with morality, and I think we are better off not trying to conflate the two. The underlying issue here is the proper interpretation of the words used by Parliament. To be very specific, a system based on the rule of law depends on what the words actually say and not on what someone wishes the words would say. Consider the implications of accepting an “even if it’s legal, it’s not right” approach to interpretation. How would we govern our conduct if there was no certainty that the rules as written would govern? We would object strenuously to any decision in which a judge rejected the legal approach to statutory interpretation in favour of an approach based on a personal view of what was “right.” The correct application of the tax rules in any specific case is often unclear – that’s why there are so many reported tax court decisions each year. Even in cases where the tax plan is aggressive, the courts may agree that the taxpayer has correctly applied the provisions of the Act, as the 2005 decision of the Supreme Court in the Canada Trustco Mortgage Co case makes clear. Adding a requirement that the plan be morally “right” would create impossible hurdles for the taxpayer, the CRA and the court. Whose standard of “right” would govern?
That is not to say that non-legal factors are necessarily irrelevant when considering an aggressive tax plan. For the taxpayer, even if the plan is upheld there may be reputational risk involved in adopting it: consider the Starbucks experience in the United Kingdom. And for the professional advisor, the applicable rules of professional conduct may impose ethical constraints when opining on the plan. I’ll return to this ethical issue in a subsequent posting.