In its latest SR&ED decision, National R&D Inc v Canada, 2022 FCA 72, the Federal Court of Appeal reminded litigants of the first principles of expert evidence, the rules and requirements that govern it and of the dangers of failing to do so.


The taxpayer, National R&D Inc., had claimed scientific research and experimental development tax credits for certain "applied sciences" work. National appealed to the Tax Court after the Minister of National Revenue denied the credits on reassessment.

The Tax Court considered whether National's work fell within the definition of SR&ED in s. 248(1) of the Income Tax Act as "experimental development." Applying the five-part test from the 1998 case of Northwest Hydraulic Consultants Ltd. v The Queen, [1998] 3 CTC 2520, the Tax Court found that National had not met four of the five SR&ED criteria.

National took particular issue with the Tax Court's finding that its work did not follow the scientific method, as required by the Northwest Hydraulic test, and appealed to the FCA. The FCA dismissed the appeal.

The FCA's Decision

Common Law Tests Interpreting Statutory Provisions

National argued that the Tax Court judge had committed a legal error by relying on the Northwest Hydraulic test. The test departs from the text of the ITA, creating a more stringent standard, among other things, by requiring that projects follow the scientific method. National argued that Northwest Hydraulic gives guidance only, and is not mandatory.

National is not the first taxpayer to advance this argument—it had been considered and rejected by the FCA only a year earlier (in Kam-Press Metal Product Ltd. v Canada, 2021 FCA 88). The FCA forcefully affirmed the 24-year-old Northwest Hydraulic test, saying that National's argument "proceeds on a misunderstanding of the relationship between the courts and legislation." It is Parliament's role to draft statutory language. Defining and applying that language is not a legal error—it is "precisely what courts are required to do."

If a frontal attack on the Northwest Hydraulic test was ever a viable strategy for litigants, National R&D has now foreclosed it.

Admissibility of Expert Evidence

To counter the finding that its work did not follow the scientific method, National tried to tender an expert report at trial that distinguished the "engineering method" from the "scientific method." The report was not admitted.

The FCA dismissed the argument that the Tax Court judge had erred in rejecting the report, identifying errors of substance and procedure in National's report.

Substantive Grounds

National's report was rejected for failing to meet the core requirements that it be impartial, objective, relevant and necessary.

Experts at trial must balance mutually incompatible objectives. They must assist the court, showing objectivity, as part of an adversarial proceeding in which the reality is that they have been retained by a party whose position their opinion supports. The FCA criticized National's expert's report for its openly partisan tone and excessive advocacy. The FCA also noted that National (and not its expert) had written most of the expert's report.

National's expert also committed the cardinal error of opining on the application of the law to the facts, which only the court can do.

Procedural Grounds

The FCA also affirmed the Tax Court's finding that National's report did not comply with s. 145 of the Tax Court of Canada Rules (General Procedure), which codifies certain admissibility requirements. An expert's report must set out the expert's evidence in full, and accompany a signed certificate acknowledging that the expert has read (and agrees to be bound by) the Tax Court's code of conduct for expert witnesses. Here, National did not file the certificate.

The FCA's ruling reminds litigants to comply with the expert report requirements in the Rules of Court of their jurisdiction—a basic but essential step.

Expert Evidence & Books of Authorities

In the FCA, presumably in response to its expert report having been rejected, National included a scholarly article in its book of authorities that commented on the engineering method, in contrast to the scientific method as traditionally understood.

The FCA found this to be an impermissible attempt to establish as authority a fact which should be a matter of expert evidence at trial. Only legislative enactments from a court's own jurisdiction, as well as decisions of courts and tribunals from all jurisdictions, can be in a book of authorities (see, e.g., Brewer v Fraser Milner Casgrain LLP, 2008 ABCA 435). Anything else, including matters of "social, applied and natural sciences," must be adduced through expert evidence. The FCA noted that opinions in scholarly articles "cannot be taken on faith" and must be subject to cross-examination.

The ruling in National R&D cautions litigants against trying to pass off evidence as authority.


There are times when expert evidence is needed to establish a party's case, as when a party seeks to establish a non-legal opinion as fact. In proffering such expert evidence, National R&D serves as a cautionary reminder to litigants to ensure compliance with both substantive and procedural admissibility requirements.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.