Anonymised, illustrative composite. Nanaimo, BC. The seller backstopped $350,000 of the buyer's acquisition loan instead of taking a bigger note. Fourteen months later the lender rewrote the facility without telling her, and the guarantee she gave stopped existing.
At a glance
A buyer was acquiring a specialty machine shop on Vancouver Island for $2,300,000. The buyer’s asset-based lender would advance $1,500,000 against equipment and receivables — short of what the deal needed. Rather than carry the whole gap as a subordinated vendor take-back note, the seller agreed to personally guarantee $350,000 of the lender’s facility, on an 18-month, amortizing basis, in exchange for a modestly higher headline price. She had confidence in the buyer’s numbers and did not expect to be called on.
The guarantee itself was one page, drafted by the lender’s counsel for the purpose: it obligated the seller to pay on demand if the buyer defaulted, and it said nothing else. No consent-to-dealings clause, no indulgence clause, no acknowledgment that the lender and buyer could vary the facility without going back to her. Nobody drafted it to look like an accommodation guarantee outside an ordinary lending relationship, because functionally that is exactly what it was.
Fourteen months in, the buyer lost a customer representing about a fifth of revenue. The lender and the buyer negotiated a response: a $200,000 increase to the facility as a bridge over-advance, a revised borrowing-base formula, a nine-month extension of the maturity date, and a new priority security position on specific equipment that effectively postponed the seller’s existing guarantee behind it. None of this was put to the seller. Her guarantee was treated as a standing instrument that did not need her involvement to keep working — a reasonable-sounding assumption that the law does not share.
Revenue down 22% at the customer loss; the facility increased by $200,000 to bridge it; nine months added to the maturity. Seven months later the buyer defaulted outright, leaving the lender with a shortfall of roughly $410,000 once the collateral was realized. The seller’s guarantee, by then amortized down to $260,000, would have covered 63% of that shortfall — on paper.
No statute governs this. The rule is common law, and the Court of Appeal for Ontario has stated it plainly: “A guarantor will be released from liability where the creditor and the principal debtor agree to a material alteration of the loan agreement without the consent of the guarantor”, citing Manulife Bank of Canada v. Conlin, [1996] 3 S.C.R. 415, at para. 2. The court traced the rule to Cotton L.J. in the English case Holme v. Brunskill (1878), 3 Q.B.D. 495 (C.A.): “the surety ought to be consulted, and that if he has not consented to the alteration…the surety himself must be the sole judge whether or not he will consent to remain liable notwithstanding the alteration, and that if he has not so consented he will be discharged.”
The rationale is fairness, not technicality: “The basis for the rule relating to material alterations arises from the guarantor’s agreement to guarantee the risk arising from the contract between the creditor and the principal debtor. Fairness dictates that this risk not be altered unilaterally by the parties to that contract.” And the court was explicit about the default position: “alterations to the principal contract will be held to be material unless they are plainly unsubstantial or necessarily beneficial to the guarantor.” A $200,000 facility increase, a changed borrowing base and a nine-month maturity extension are none of those things — each expands or extends the risk the guarantor originally agreed to stand behind, not narrows it.
Four things can save a lender in this position, and none of them was present here: the alteration is “plainly unsubstantial”; the alteration is “necessarily beneficial to the guarantor”; the guarantor contracted out of the protection in advance — “typically, an institutional lender’s standard form guarantee will contain provisions allowing for time extensions for repayment, renewal, and forbearance”, which this one-page document did not; or the guarantor actually consented to the specific change, which she was never asked to do. Turfpro’s own guarantee had the same defect, for the same reason — a private, purpose-built document rather than an institutional form — and the Court of Appeal noted it directly: the language was “very sparse and did not contain any provisions that expressly allowed for time extensions for repayment, renewal or forbearance.”
When the lender demanded the $260,000 outstanding under the guarantee, the seller’s counsel set out the unconsented facility increase, the borrowing-base change and the maturity extension, and put the lender’s own counsel on notice of Turfpro and the line of authority behind it. The lender’s litigation counsel advised against pursuing a claim it was likely to lose on a documents-only record, and the demand was withdrawn. The lender absorbed the $260,000 as an uncollectible balance.
A single sentence would have changed the outcome: an express consent-to-dealings clause, of the kind institutional guarantees carry as a matter of course, letting the lender and borrower vary the facility — extend it, increase it, change its terms — without further reference to the guarantor. Had that clause existed, the same $200,000 increase and nine-month extension would have changed nothing about the seller’s exposure, and the lender would have recovered the full $260,000 it eventually wrote off.
The amendment documents were never copied to the seller or her counsel — not because anyone was hiding anything, but because nobody on either side of the amendment thought of her as a party whose consent mattered. An accommodation guarantor who is not otherwise a party to the credit agreement has no visibility into what the lender and borrower do to it later, unless the guarantee itself creates a duty to tell her. That silence is not a red flag of bad faith; it is the ordinary and entirely foreseeable way this kind of guarantee gets lost.
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.
