Malone, 2026 FCA 166: GLGI donation credits denied again, and the Tax Court's reasons were sufficient

2026 FCA 166 · Federal Court of Appeal

Author: Sas Tullo

This is an appeal by two participants in the Global Learning and Gifting Initiative, the leveraged donation program that has produced roughly 1,500 Tax Court appeals and, so far, no taxpayer win. Barry and Brandon Malone had their donation tax credits denied and asked the Federal Court of Appeal to send the case back because, they said, the trial judge’s reasons skipped the step between the evidence and the conclusions. The Court of Appeal disagreed on every point and dismissed both appeals with costs. For anyone still holding a GLGI reassessment, the decision confirms that the Mariano principles remain intact and that an insufficiency-of-reasons argument does not get around the deferential standard of review.

Facts

Barry and Brandon Malone each appealed from a Tax Court judgment dated March 27, 2025, decided on common evidence with common reasons cited as 2025 TCC 43, which found that their purported donations to the Global Learning and Gifting Initiative (GLGI) were not eligible for credits under section 118.1 of the Income Tax Act (para 1). Under the GLGI program, a participant makes a cash contribution, purportedly receives software licences said to be worth many times the cash, and donates the licences to a participating charity; two receipts issue, one for the cash and one for the claimed value of the licences (para 2).

The court noted that GLGI has produced approximately 1,500 Tax Court appeals, with the principles of Mariano v. The Queen, 2015 TCC 244, upheld repeatedly since, including in Aslam v. The King, 2024 FCA 193 (para 3), and that no taxpayer has ever succeeded (para 4).

Issues

The appellants’ principal argument was that the Tax Court’s reasons were insufficient under R v. Sheppard, 2002 SCC 26, because of a missing step between the judge’s conclusions and the evidence (para 5). They also submitted that the Tax Court erred in finding that they lacked donative intent, that the software had nominal value, and that they did not own the licences, contending that their uncontradicted testimony had been ignored and that the judge had improperly relied on Mariano (para 6).

Holding

Both appeals were dismissed with costs in the all-inclusive amount of $500 against each appellant (para 15). The Tax Court did not err in its findings and its reasons were sufficient (para 8).

The court’s reasoning

Biringer J.A., for the court, began with the standard of review. Under Housen v. Nikolaisen, 2002 SCC 33, questions of law are reviewed for correctness and questions of fact or mixed fact and law, absent an extricable legal question, for palpable and overriding error (para 7). Because the appeals came from the informal procedure, paragraph 27(1.3)(d) of the Federal Courts Act added a further hurdle: any factual error had to have been made “in a perverse or capricious manner or without regard for the material before it” (para 7). The court characterized the substantive complaints as attacks on factually suffused findings, as in Aslam at paragraph 2, and declined to let the appellants escape that deference by recasting them as an insufficiency of reasons (para 8).

On donative intent, the appellants’ general history of charitable giving was irrelevant to their intent regarding the GLGI transactions. While subjective intent may be relevant, the test is ultimately objective: whether the donor intended to impoverish themselves, having regard to all the circumstances, citing Symes v. Canada, [1993] 4 S.C.R. 695 at page 736 and Walby v. Canada, 2025 FCA 94 at paragraph 50 (para 9). The Tax Court had applied that test, had considered the appellants’ “win-win” testimony, and had concluded that an expected refund far exceeding the cash contributed negated donative intent (para 9).

On valuation, the court rejected the claim that the Tax Court had simply borrowed findings from Mariano. Although the appraisals were identical to those in Mariano, the trial judge had dealt with them directly, finding them misaligned with the program version years and the value of the licences “infinitesimally small” relative to the receipted amounts; the court added that the appraisals concerned a different set of transactions and software from the GLGI (para 10). Exhibit E, an assignment of licence dated October 8, 2010, was not a valuation but a document stating certain appraised values whose author the appellants could not identify, so there was no reliable evidence that the software was worth more than a fraction of its receipted value (para 11).

On ownership, the Tax Court had found the testimony vague and had concluded there was no evidence of actual possession, dominion, or control over the software. Having viewed the software was not evidence of ownership, and the Court of Appeal does not reweigh evidence, as Aslam at paragraph 3 had reminded. The initial onus in a tax appeal is on the taxpayer to disprove the Minister’s assumptions of fact, citing Hickman Motors Ltd. v. R, 1997 CanLII 357 (S.C.C.) at paragraph 92 and House v. The Queen, 2011 FCA 234 at paragraph 30, and the assumption that the appellants did not own the licences was unassailed (para 12).

Finally, the reasons were sufficient: the issues and analysis were clearly laid out, providing a meaningful basis for appellate review, citing R. v. Sheppard, 2025 SCC 29 at paragraph 45 and R. v. R.E.M., 2008 SCC 51 at paragraph 17 (para 13). Reasons need not describe every piece of evidence or every argument, and are read with the record, the submissions, and the history of the case, citing Hennessey v. Canada, 2016 FCA 180 at paragraph 10 and Dnow Canada ULC v. Grenke Estate, 2020 FCA 61 at paragraph 21 (para 14).

Practical implications

The decision adds little new doctrine, and that is its point. The three findings that defeat GLGI claims, the absence of donative intent, the nominal value of the licences, and the failure to prove ownership, are findings of fact or mixed fact and law, and on an informal procedure appeal they are protected by both Housen and paragraph 27(1.3)(d) of the Federal Courts Act. The appellants tried to convert a factual disagreement into a legal error about the reasons; the court answered that reasons are read with the record and need not catalogue every argument (paras 13-14). In my view that answer will apply to most of the remaining GLGI appeals, many of which were decided on oral reasons (para 3).

The restatement of the objective test for donative intent (para 9) is also useful beyond GLGI. A donor’s broader charitable record does not establish intent for a particular transaction, and an expected tax benefit exceeding the cash outlay is, on this line of authority, inconsistent with an intention to impoverish oneself.

Section 118.1 of the Income Tax Act has been amended several times since the taxation years in issue, most recently in 2026 according to the Justice Laws website; this comment does not describe those amendments. Whether an appeal was taken is not verified here.

The takeaway is simple: a GLGI reassessment will not be reversed on appeal by re-arguing the facts, and dressing the argument as a complaint about the reasons does not change the standard of review. The $500 costs award against each appellant shows that the Court of Appeal will impose a modest cost on appeals of this kind.

This comment is general information, not legal advice. The law may have changed since the decision; check the current text in the tax wiki.

Citation and links

Malone v. Canada, 2026 FCA 166, judgment delivered at Ottawa on October 2, 2026, Gleason, Biringer, and Rochester JJ.A., reasons for judgment by Biringer J.A. Reasons on the Federal Court of Appeal’s site. Also on CanLII. The decision under appeal is the judgment of the Tax Court of Canada dated March 27, 2025, reasons cited as 2025 TCC 43 (para 1).