Anonymised, illustrative composite. The confidentiality clause the parties had spent three weeks negotiating did not reach the recipient that worried the vendor most, because that recipient was the Crown.
At a glance
A family-owned distributor of specialty industrial chemicals, with four depots in Ontario and Quebec, agreed to sell its shares to a strategic buyer for $178,000,000. Combined revenues cleared the party-size test and the target's Canadian assets cleared the transaction-size amount, so a filing was required; see the pre-merger notification glossary entry.
The prescribed information for a notifiable transaction is not a summary. It reaches into product lines, customers and financial detail, and on this business the most sensitive item was a schedule of gross margin by product family and by customer tier. The vendor's view was that this schedule was the business. Two competitors had spent years trying to work out how a distributor of that size sustained the margins it reported, and the schedule answered the question directly.
The vendor's instinct was to solve the problem with contract. The share purchase agreement already carried a heavily negotiated confidentiality covenant, and the parties had run a clean-team protocol during diligence — the mechanics are set out in the non-disclosure agreement glossary entry and what a buyer should refuse to sign in an NDA.
None of that reaches a filing. Once the schedule is handed to the Commissioner it is in the hands of a party to no contract at all. The vendor asked the right question and asked it late: who at the Bureau may pass this on, and to whom? Its counsel's first answer — that the material was “confidential to the review” — was a paraphrase of nothing. The governing rule is a statutory prohibition with a defined scope and two express exceptions, and the exceptions were where this vendor's actual exposure lay.
The margin schedule covered 41 product families and four customer tiers. On the vendor's own modelling, if the two competitors it worried about could see the schedule, the realistic effect was a 90 to 140 basis point compression of blended gross margin over three years as they repriced into the accounts where the distributor made its money. On $186,000,000 of revenue at a 21.4% blended gross margin, 115 basis points is about $2,140,000 a year of gross profit — against a purchase price of $178,000,000, and against a rollover stake the vendor was keeping.
The vendor was not being precious. It was quantifying a real risk to a business it would still partly own after closing — see the rollover equity glossary entry. What it had not done was work out which of the several routes by which that schedule could escape were actually open.
s.29(1) is a prohibition on a defined class of people. It provides that “No person who performs or has performed duties or functions in the administration or enforcement of this Act shall communicate or allow to be communicated to any other person except to a Canadian law enforcement agency or for the purposes of the administration or enforcement of this Act” five listed categories, including “the identity of any person from whom information was obtained pursuant to this Act”, “any information obtained pursuant to section 11, 15, 16 or 114”, “whether notice has been given or information supplied in respect of a particular proposed transaction under section 114”, and “any information obtained from a person requesting a certificate under section 102”.
Three features are worth pinning. The unit is the person administering the Act, not the information: s.29 binds Bureau officers, not the buyer, not its counsel, and not a customer told something in a market contact. The protection extends to the existence of the filing itself under paragraph (c). And the carve-out for communication “for the purposes of the administration or enforcement of this Act” is what permits the Bureau's market contacts at all — which is how a review is done.
Then s.29(2): “This section does not apply in respect of any information that has been made public or any information the communication of which was authorized by the person who provided the information.” That is not a technicality. Anything the vendor had itself put into a trade-association benchmarking submission, a lender presentation that circulated, or a confidential information memorandum sent to eleven bidders is outside the prohibition, because the prohibition is on communication of protected information and s.29(2) removes the protection.
The analysis reordered the vendor's risks. The Bureau was the least of them: bound by a statutory prohibition, unable even to confirm a filing exists, and permitted to use the schedule only for the review. The genuine exposures were the buyer's own team — contractually bound but commercially adverse if the deal broke — and the vendor's own history of disclosure, which s.29(2) had already stripped of protection.
So the schedule went to the Bureau in full and on time; access on the buyer's side stayed inside a clean team of four named people with no operating role; and the vendor's counsel audited what had already gone out to bidders, on the footing that anything in that pile was no longer protectable by anyone. Compare a data room leak that reached a competitor, where the escape route was the private one this vendor had correctly identified, and three customers served with a production order, where s.29 binds the Bureau but says nothing about what the recipient of an order may say.
Refusing to supply the schedule was not a live option. Supplying incomplete prescribed information does not start the s.123(1) waiting period, because that period runs from the day “the information required under subsection 114(1) has been received by the Commissioner”. A filing missing a required schedule is a filing that has not started the clock, and the parties would have discovered that at the point they expected to be free to close.
The more expensive error would have been the opposite one: assuming that because the Bureau is bound, the information is safe generally, and relaxing the private-side controls. s.29 does nothing for a vendor whose margin data reaches a competitor through a bidder that saw it in diligence. The statute protects one channel. The contract has to protect the rest, and neither covers what the vendor has already published.
The tell was a phrase in a memo: the material was “confidential to the review”. That describes a feeling, not a rule. Where the concern is disclosure by a regulator, the question is always which provision binds whom and what its exceptions are — because the answer usually reallocates the worry rather than removing it.
The second tell was that nobody had asked what the vendor had already given away. s.29(2) makes prior publication decisive, so an audit of the vendor's own past disclosures is a cheaper and more useful exercise than another round of drafting on a confidentiality clause that was already fine.
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.
