http://decision.tcc-cci.gc.ca/tcc-cci/decisions/en/item/73273/index.do 
Michael H. Lubetsky of Davies has been kind enough to provide the following commentary on the McKesson recusal decision (I am indebted to Guy Du Pont, Ad. E., for initiating the collaboration). As I have already indicated Justice Boyle is a former colleague of mine and a good friend. Therefore I will not opine on these issues. However I cannot refrain from observing that Mr Lubetsky’s obvious erudition and balance have substantially elevated the tone of the debate around this decision:
…
On September 4, 2014, Justice Boyle rocked the tax bar with his ex parte, post-trial order and reasons (the “Order”) recusing himself from any further involvement in the case due to perceived attacks upon his “honour and integrity” by the taxpayer in the related appellate proceedings. In addition to the recusal itself, the Order includes a detailed refutation of the taxpayer’s factum before the Federal Court of Appeal (“FCA”) as well as point-blank accusations that the taxpayer’s counsel intentionally sought to mislead the appellate court.
The Order is apparently without precedent and has generated much discussion about whether the taxpayer’s counsel took zealous advocacy too far and/or the judge overreacted. It also remains to be seen what impact, if any, the Order may have upon the taxpayer’s appeal. However, on a broader level, the Order raises some important questions about the appeal process and how such difficulties might be prevented in the future.
Specifically, the Order invites a serious reflection of whether the Tax Court of Canada (and other Canadian courts) should adopt the British practice in which judges circulate draft judgements to counsel before finalising them. Had such a practice been followed in this case, many of the taxpayer’s complaints may well have been avoided and the trial judge would not have been placed in a situation where he felt it necessary to publicly defend his integrity.
Background
On December 13, 2013, Justice Boyle issued his 105-page decision in the transfer-pricing case of McKesson Canada Corporation v HMQ, 2013 TCC 404, dismissing the taxpayer’s appeal (the “Trial Decision”). The taxpayer launched an appeal in the FCA, eventually filing its memorandum in support on June 11, 2014 (the “Factum”). At the same time, Justice Boyle kept carriage of several post-trial matters (i.e., costs and various confidentiality measures) that remained outstanding.
The prevailing theme in the Factum, as appeared in its opening paragraph, was that “the trial judge discarded the case pleaded and argued by the parties and decided the appeal on grounds that were not raised in the pleadings or argued at trial, but made their first appearance in the trial judge’s reasons well after the trial was over.” The Factum went on to document various findings and theories on which the trial judge relied that, supposedly, were (a) not in evidence, (b) never put to counsel for argument and/or (c) in conflict with other statements in the Trial Decision. The Factum also alleged, inter alia, that the trial judge demonstrated “palpable antipathy towards the taxpayer” (paragraph 9) and that his analysis was “infected by his pejorative and unfair comments” (paragraph 70).
As Justice Boyle explained in the Order, he does not habitually review facta filed in respect of his decisions. However, the Factum was drawn to his attention or sent to him “by several prominent Canadian tax lawyers as well as by a colleague on the Court.” (paragraph 7) Following his review of the Factum, he decided of his own motion to recuse himself from dealing with post-trial matters, on the grounds that the Factum essentially called into question his impartiality and integrity.
[4] As detailed below, I have, of my own motion, decided that I am compelled to consider whether I need to recuse myself from the two remaining issues before this Court. A consideration of this issue is required because I became aware that the Appellant and Appellant’s counsel, together with its co-counsel in the Federal Court of Appeal in respect of the appeal of the trial decision, had made certain public written statements about me in its factum in the Federal Court of Appeal (the “Factum”) which, upon reflection, appear to me to clearly include:
(i) allegations that I was untruthful and deceitful in my Reasons;
(ii) clear untruths about me, what I said and heard in the course of the trial, as well as the existence of evidentiary foundations supporting what I wrote in my Reasons; and
(iii) allegations of impartiality on my part.
[5] This requires me to consider whether:
(i) I believe that a reasonable person reading the Factum, my Reasons, and the relevant portions of the transcript would believe that the trial judge so strongly complained of by McKesson Canada might not be able to remain impartial in his consideration of costs and confidential information;
(ii) I believe I can impartially consider, weigh and decide the costs and confidential information issues before me; and
(iii) whether the public challenge of my impartiality expressed by McKesson Canada and its co-counsel in the Factum is itself sufficient to warrant recusing myself at this stage.
[…]
[133] I view these as public allegations by a party to the costs and confidential information matters remaining before this Court that, regardless of the merits of their reasoning or their thoughts, I am unable to decide the remaining matters impartially. I believe that a reasonable person reading only these phrases from the Factum, without reviewing my Reasons or the trial Transcript, would believe that such strong complaints by McKesson Canada and its counsel may give rise to a serious doubt that I will be seen to be able to dispose of the two remaining issues and discharge my duties on an impartial basis.
Justice Boyle devoted 46 pages to reviewing in detail the various criticisms levelled against him in the Factum, pointing to various extracts from the trial transcripts to address the taxpayer’s claims that his findings lacked evidentiary foundation or had never been the object of argument by counsel.
He also offered his view that the “untruths” in the Factum were stated “intentionally”:
[20] For these Reasons, it is my view that the Appellant has wrongly accused me of being untruthful, dishonest and deceitful. I am simply unable to read their Factum or the Reasons any other way on this point.
[21] I believe they have wrongly written these things in the Appellant’s Factum about me intentionally under the guise of fearlessly advancing and representing the interests of McKesson Canada. I believe this clearly crosses the line as to what is appropriate.
[…]
[40] It appears very clear to me that, while the Appellant may have every right to seek to challenge the evidentiary foundation of my conclusions and findings, they have simply told clear untruths about me and what I did or did not say when they state that McKesson’s tax motivation was not ever put to them during the trial and that they were therefore deprived of any opportunity to address it.
[41] I certainly believe I clearly put it to Appellant’s counsel during his first witness’ testimony, and raised it again with counsel at the start of his oral argument. The Appellant made written submissions on the issue of tax planning as he acknowledges it arose in argument. One can read what they will into the Appellant’s decision not to argue the point or conduct redirect examination, but it appears to me to be patently untrue that I did not raise it with the Appellant early, at times when they could respond with additional evidence, with a summary of the evidence to change my impression, or with whatever legal argument they chose.
[…]
[103] It appears to me that my concerns with credibility and weight issues were communicated clearly on several occasions during the trial, as noted above with both Mr. Schabas and Mr. Gilliland, and as noted above with Mr. Reifsnyder at the end of his testimony. It appears to me that the Appellant writing that “nor did the trial Judge indicate any such concern during the trial”, and that paragraph 245(r) of my Reasons is “flatly inconsistent” with my quoted comment are inherently and demonstrably untrue. That is, I believe the Appellant was telling untruths about me that go beyond the appellate advocacy craft of colour, spin and innuendo.
[…]
[137] It may be that some of the perceived untruths about the trial judge described above under heading II might individually not warrant recusal, and may be within an appellate advocate’s licence to overstate through the use of absolutes like ‘never’, ‘only’ and ‘any’.
[138] However, I am satisfied that a reasonable fair-minded Canadian, informed and aware of all the issues addressed above, would entertain doubt that I could remain able to reach impartial decisions. I believe that such a reasonable fair-minded and informed person, viewing this realistically and practically would, after appropriate reflection, be left with a reasoned suspicion or apprehension of bias, actual or perceived. Canadians should rightly expect their trial judges to have broad shoulders and thick skins when a losing party appeals their decision, but I do not believe Canadians think that should extend to accusations of dishonesty by the judge, nor to untruths about the judge. Trial judges should not have to defend their honour and integrity from such inappropriate attacks. English is a very rich language; the Appellant and its counsel could have forcefully advanced their chosen grounds for appeal without the use of unqualified extreme statements which attack the personal or professional integrity of the trial judge.
It bears note, however, that the Order did acknowledge that one of the complaints made in the Factum (regarding one particular fact) was well-founded:
[134] Before concluding, it appears appropriate for me to acknowledge an error in my Reasons that the Appellant has correctly identified in its Factum. In paragraph 56 of the Factum, the Appellant refers to the phrase in the sentence in paragraph 13 of my Reasons that “a portion was loaned for a period to another Canadian corporation to permit its tax losses to be used”. I can confirm that they are correct in saying that there was no support in the evidence that the purpose of that loan was for the Vancouver affiliate to use its tax losses. I sincerely apologize for using the words “to permit its tax losses to be used”. It will be for others to decide, of course, the relevance of my mistaken phrase to my overall Reasons and decision.
[135] By way of explanation and not excuse, I can see that I mistook a two-part question I had intended to ask Mr. Brennan by way of clarification at the end of his testimony, as a question and an answer.
[…]
Issues of Fairness
The FCA now faces unenviable task of sorting out the repercussions of the Order on the taxpayer’s dispute with the CRA, while members of the legal and tax communities debate whether the Factum overstepped the bounds of fair advocacy and, if so, whether the Order constituted an appropriate response. However, in addition to these immediate issues, the Order raises more fundamental questions about the entire appellate process and especially how such problems can be avoided in the future.
As noted in the Order, a trial’s judge job “ends with the rendering of reasons and judgment”:
[8] A trial judge’s job on the merits ends with the rendering of reasons and judgment. There is rightly no role for the trial judge in the appeal of the trial decision. Counsel on each side in the appellate court is free to make whatever arguments they wish, including claiming or denying support in the record, the use of emphasis and spin, or even trying to argue a case it thinks it can win instead of the case it has. That is all of counsel’s choosing and to be ultimately considered and decided by the appellate court.
There are, of course, sound reasons for having the trial judge playing no role in the appeal of his or her judgement; inter alia that it would essentially turn the trial judge into a party to the appeal. On the other hand, the systematic exclusion of the trial judge from the appellate proceedings comes at a cost, particularly in fact-driven cases (such as those involving transfer pricing). The trial judge knows best his or her own thought process and is thus the person best able to explain that reasoning underlying the judgement under appeal. In the absence of the trial judge, litigants can interpret (if not deconstruct) a written judgement so as to attribute to the judge thought processes that may not, in fact, have played any role in the decision.
Many appeals involve suggestions that a trial judge may have overlooked a particular piece of evidence or relied on facts not in evidence, or even relied on points that had not been the object of full briefing and argument. Such allegations, when they reach a certain point, ultimately call into question the basic competence of the trial judge while denying him or her the opportunity to straighten out the record himself or herself.
Of course, judges can significantly reduce the risk of having their findings challenged on appeal by producing thorough, convincing, and well-written judgements. However, not every written judgement meets this standard, and in very complex and fact intensive cases, it is not realistic to expect judges, given their already heavy work load, to produce, in every case, written reasons that exhaustively review every fact and issue free from error or omission.
The British Solution: Circulating Drafts of Reasons
The British practice, instituted in a 1998 Practice Statement [Practice Statement (Sup. Ct. Judgments) (No. 1), [1998] 1 WLR 825, [1998] All ER 667. The practice statement was amended the following year (Practice Statement (Sup Ct. Judgements) (No. 2), [1999] 1 WLR 1, [1999] 1 All ER 125), and today appears as Direction 40E of the Rules of Civil Procedure (
http://www.justice.gov.uk/courts/procedure-rules/civil/rules/part40/pd_part40e )%5D, where judges circulate advance drafts of their reasons confidentially to the parties has the potential not only to significantly improve the quality of judgments (and thus prevent appeals in the first place), but also to help remedy the problems caused by the absence of the trial judge from any appellate proceedings.
In the very recent judgement of the UK Supreme Court in
Marley (Appellant) v Rawlings and another (Respondents) (Costs), [2014] UKSC 51, President Neuberger of Abbotsbury (Chief Justice in a Canadian setting) for the Court described this practice as customary:
28. In the usual way, a copy of this judgment was sent in draft to counsel for the parties and for the insurers of the Solicitor, with an invitation to make comments. Save for some helpful typographical corrections and the like, the only response of substance came from the respondents’ counsel, who formally confirmed that they disclaimed any entitlement which they may have had under their CFAs to uplift or success fees “for all purposes”.[…]
While the primary purpose of circulating drafts is to allow the correction of typographical and clerical errors, the British courts have recognised that litigants can, in appropriate circumstances, also invite the trial judge:
a) to provide additional reasons on particular points [G, Aerospace Publishing Ltd v Thames Water Utilities Ltd, [2006] EWCA 717];
b) to allow parties to make supplemental submissions on points that appear in the draft judgment but which were not previously argued [Director of Public Prosecutions v P (No 2) (Note), [2007] EWHC 1144 (Admin); [2007] 4 All ER 648];
c) to edit “detrimental observations about an individual or his lawyers, which on the fact of it are not necessary to the judgment and appear to be based on a misunderstanding of the evidence, or a concession or a submission [R (Mohammed) v Secretary of State for Foreign and Commonwealth Affairs, [2010] EWCA Civ 158]” .
Indeed, the courts in England have remarked that counsel have an obligation to call the trial judge’s attention to a perceived defect in the draft judgement, otherwise they may potentially lose the right to invoke the perceived deficiency on appeal. [Ini re M (A Child) (Non accidental Injury: Burden of Proof), [2008] EWCA Civ 1261; In re A&L (Children), [2011] EWCA Civ 1611.] On the other hand, the courts have also held that the circulation of draft judgement is not intended to lead to a relitigation of the case and that judges can summarily refuse attempts by counsel to reopen a case in response to draft reasons [Egan v Motor Services (Bath) Ltd, [2007] EWCA Civ 1002, [2008] All ER 1156; Browning v Brachers, [2004] EWHC 16 (QB).].
The practice of circulating draft judgements, and considering representations made in response thereto, had led to the a number to celebrated cases where a trial judge changed his or her mind on the disposition of a case. As noted by the Lord Justice Peter Gibson in his concurring opinion in one such case:
Such judicial tergiversation is in general not to be encouraged, but circumstances may arise in which it is necessary for a judge to have the courage to recall his order. If, as in Millensted and Pittalis, the judge realises that he has made an error, how can he be true to his judicial oath other than by correcting that error so long as it lies within his power to do so? [
Robinson v Fernsby, [2003] EWCA Civ 1820]
Could the Order Have Been Avoided?
Back here in Canada, had the taxpayer received an advance draft of the Trial Decision, virtually all of the complaints made in the Factum would have had to be made (perhaps more tactfully) first to the trial judge, who would have had an opportunity to expand upon, clarify and/or correct his reasons accordingly, le cas échéant. Had an appeal then ensued (by either party), it would have been very difficult to then allege that the trial judge had decided the case based on “grounds not argued at trial” and there would have presumably been few if any lingering questions about the evidentiary foundation for the facts on which he relied. In addition, the error properly identified by the taxpayer would, in all likelihood, have been corrected. All in all, the parties would have received a judgement that would have addressed their alleged errors and ambiguities thereby allowing them, on appeal, to focus on the established facts of the case and applicable legal principles — which, before an appellate court, is how it should be.