Anonymised, illustrative composite. A non-Canadian investor bought identical 40% economic stakes in two similar Canadian businesses the same month — one a corporation, one a limited partnership. The Investment Canada Act treated the two deals as opposites.
At a glance
A non-Canadian investor completed two acquisitions in the same month, structured by different sellers for unrelated reasons: a $30,000,000 purchase of 40% of the voting shares of a Canadian corporation (Deal A), and an $18,000,000 purchase of a 40% limited partnership interest in a similarly sized Canadian business operating in an adjacent sector (Deal B). The investor's in-house acquisition questionnaire, used for both, asked a single question relevant to the Investment Canada Act: what percentage of the target was being acquired.
On that question, the two deals looked identical — 40% economic and voting participation in each case — and the investor's team treated them as raising the same ICA analysis.
The Act does not define control by a flat percentage that applies uniformly across every kind of entity. It supplies its own definition, and that definition is asymmetric depending on whether the target is a corporation or something else — a limited partnership among them. Applying one deal's answer to the other deal, because the percentages matched, was the error.
Deal A: $30,000,000 for 40% of a corporation's voting shares. Deal B: $18,000,000 for a 40% limited partnership interest. Identical proportional stake, opposite treatment under the Act, as the next section sets out.
s.28(1) sets out the only ways a non-Canadian acquires control under the Act, including at s.28(1)(c) “the acquisition of all or substantially all of the assets used in carrying on the Canadian business” — a fourth route, distinct from any share or interest acquisition, discussed further below.
Read the unit — the definition runs opposite ways by entity type. s.28(3)(c): “the acquisition of less than a majority but one-third or more of the voting shares of a corporation or of an equivalent undivided ownership interest in the voting shares of the corporation is presumed to be acquisition of control of that corporation unless it can be established that, on the acquisition, the corporation is not controlled in fact by the acquirer through the ownership of voting shares.” Deal A's 40% cleared the one-third floor, triggering a rebuttable presumption of control — the investor would need to establish it did not in fact control the corporation to displace that presumption, and nothing about the deal suggested grounds to try.
s.28(3)(b), by contrast: “the acquisition of less than a majority of the voting interests of an entity other than a corporation is deemed not to be acquisition of control of that entity.” A limited partnership is an entity other than a corporation. Deal B's 40% interest is less than a majority, so s.28(3)(b) deems it, categorically, not an acquisition of control — not a rebuttable presumption running the other way, a flat deeming provision. There is no evidentiary inquiry into whether the investor controlled the partnership in fact; the subsection answers the question by entity type before that inquiry can begin.
The consequence is not a matter of degree. Deal A proceeded through the Act's notification-or-review analysis as an acquisition of control, exactly as a notifiable-only investment nobody filed describes for a corporate target. Deal B sits outside the Act's control-acquisition framework entirely — no notification, no review, regardless of the economic stake, because s.28(3)(b) has already answered the only question the Act asks.
The asset-deal fork sharpens the same point. Had the investor instead structured either transaction as a purchase of the underlying business's assets rather than its equity, s.28(1)(c) would apply on its own terms: control is acquired only where the purchase is of “all or substantially all of the assets used in carrying on the Canadian business.” A divisional carve-out — less than substantially all of the assets — can fall outside the Act altogether even where the equivalent full share purchase would not, which is a separate structuring lever from the entity-type question Deals A and B turned on.
Deal A was run through the Act's standard analysis: 40% of a corporation, presumptively controlling, assessed for notification and, on the numbers, below the review threshold — notifiable only, following the same s.11/s.12 path as a notifiable-only investment nobody filed, and filed on time this time.
Deal B was documented as falling outside the Act's acquisition-of-control definition, with a contemporaneous legal memorandum on file identifying s.28(3)(b) as the specific basis — not a post-hoc rationalization built after a later question, but a record made at the time so any later reviewer, lender or co-investor asking about ICA exposure on Deal B has something more than an assertion to look at.
Had the investor's team simply assumed Deal B required the same ICA notification as Deal A — the safer-seeming, more-cautious-looking error — the practical cost would have been limited to an unnecessary filing. The more consequential version of the error runs the other way: assuming a corporate acquisition at one-third or more of voting shares is exempt because a similarly sized partnership interest was, and skipping notification on a deal the Act actually reaches.
The tell was the acquisition questionnaire's only field: percentage acquired. It never asked what kind of entity was being acquired, silently assuming that variable did not matter. Under s.28 it is the variable that decides the answer before the percentage is even relevant.
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.
