Anonymised, illustrative composite. The buyer and seller of a specialty manufacturer had been competitors for a decade. The non-compete protecting the sale was drafted, and cleared, as an ordinary restrictive covenant — before anyone asked whether it was also a federal offence.
At a glance
A private manufacturer of specialty industrial fasteners agreed to sell substantially all its assets to a larger competitor for $30,000,000. The two businesses had bid against each other for the same accounts for roughly ten years. The purchase agreement carried a standard, seller-side non-compete: the vendor's principal agreed not to compete anywhere in Canada for five years, with $2,000,000 of the purchase price allocated to the covenant. See the glossary entry on the non-competition covenant for how that allocation is normally structured.
Buyer's counsel reviewed the clause the way sale non-competes are usually reviewed: duration, geographic scope, and the business interest it protects, against the common-law test for an enforceable restraint of trade. On that test the clause was comfortable — five years and a national scope were defensible given the target sold nationally, and the business being protected was the goodwill the buyer was paying $2,000,000 for.
What the common-law review never asked is the question a competitor's antitrust counsel asked when the deal was later disclosed in an unrelated proceeding: is an agreement between two competitors, one of whom promises to stop competing entirely, itself the offence Competition Act s.45 exists to catch? s.45(1) provides that “every person commits an offence who, with a competitor of that person with respect to a product, conspires, agrees or arranges…to fix, maintain, control, prevent, lessen or eliminate the production or supply of the product”. A vendor who agrees not to compete at all is agreeing, for the covenant's term, to eliminate its own supply — the paradigm case the paragraph describes.
The penalty is not a contract remedy. s.45(2): a person convicted under s.45(1) “is guilty of an indictable offence and liable on conviction to imprisonment for a term not exceeding 14 years or to a fine in the discretion of the court, or to both”. Common-law unenforceability just means a court will not order specific performance of an unreasonable clause. Criminal liability under s.45 is a separate question entirely, and it does not go away because the clause happens to also be enforceable.
Deal value $30,000,000; non-compete allocation $2,000,000; duration five years; scope national, justified by the target's actual national customer base rather than by habit. None of those figures change under the criminal analysis — what changes is which test decides whether the clause survives.
The escape route exists, and it has a burden the defendant carries. s.45(4): “No person shall be convicted of an offence under subsection (1) or (1.1) in respect of a conspiracy, agreement or arrangement that would otherwise contravene that subsection if (a) that person establishes, on a balance of probabilities, that (i) it is ancillary to a broader or separate agreement or arrangement that includes the same parties, and (ii) it is directly related to, and reasonably necessary for giving effect to, the objective of that broader or separate agreement or arrangement; and (b) the broader or separate agreement or arrangement, considered alone, does not contravene that subsection.”
Read the unit. The defence does not ask whether the non-compete itself is reasonable in isolation — that is the common-law question, and it is a different question with a different burden allocation. It asks whether the non-compete is ancillary to something else: here, the asset purchase agreement, which is the “broader or separate agreement” the two limbs of s.45(4)(a) test against. A genuine, arm's-length sale of a business is not itself an agreement that contravenes s.45 — satisfying s.45(4)(b) — so the question narrows to whether the covenant is directly related to, and reasonably necessary for, that sale's objective.
“Reasonably necessary” is doing real work in that sentence, and it is not automatically satisfied by a covenant that would also pass common-law scrutiny. A national, five-year non-compete tied to a business that only ever sold nationally is a defensible fit between restraint and objective. The same clause copied onto the sale of a business that only ever operated in one province would be harder to justify as reasonably necessary for that sale, even though nothing about the words of the clause changed.
Nothing in s.45(4) requires the parties to build a contemporaneous record of that reasoning — the defence can be established at trial on a balance of probabilities from whatever evidence exists. But a defendant carrying that burden is in a materially better position with a recital explaining why the scope matches the business sold than with a template clause and no explanation at all.
The parties did not change the substance of the deal. They narrowed the recital language to tie the five-year, national scope explicitly to the target's actual national customer footprint and the buyer's stated integration timeline, and confirmed in a closing certificate that the covenant's geographic and temporal scope had been considered against that footprint rather than adopted as boilerplate. The asset purchase agreement itself — the broader agreement s.45(4)(b) requires to be clean — was reviewed and confirmed to raise no separate s.45 concern of its own, such as a customer-allocation term of the kind discussed in the seller kept a division and the parties split the customers.
None of this changed the covenant's common-law enforceability, which had already been satisfied. What it created was the record needed to discharge a burden the common-law analysis had never engaged with at all.
A vendor non-compete found not to satisfy s.45(4) is not merely unenforceable — it exposes the individuals who negotiated and signed it to the s.45(2) penalty, up to 14 years' imprisonment or a fine at the court's discretion, or both. A common-law review that stops at reasonableness answers the wrong question for that exposure.
The tell was the parties' prior relationship. A non-compete between a seller exiting the market and a buyer that never competed with it barely raises the s.45(1) question, because there is no ongoing rivalry the restraint could be seen as suppressing. A non-compete between two businesses that spent a decade bidding against each other for the same customers is, on its face, exactly the fact pattern the ancillary-restraints defence exists for — and exactly the fact pattern a purely common-law review has no reason to flag.
A 30-minute call is enough to tell you whether AI pays for itself here.
Canadian small-business transaction data is not published anywhere, so most valuations in this country quote an American benchmark. The Deavo–Treadstone Acquisition Index is a daily record of Canadian listings built to replace that: asking-price distributions by province and city are published now, and days on market, departure rates and asking-to-sale spreads follow as the series lengthens. Leave an email and we will tell you as each measure lands.
No pitch, no listings. One email when the first report lands.
