Anonymised, illustrative composite. One provision in the whole programme is written for a change of ownership, and the first draft of the asset schedule would have taken this deal outside it.
At a glance
The vendor had run a six-bay repair and tyre shop on a provincial highway for eighteen years through a corporation, and was selling to a long-serving licensed technician backed by a family investor at $585,000 for the assets: shop equipment, hoists, diagnostic gear and tyre inventory, the leasehold improvements, the customer database and the trade name. In 2022 the vendor company had borrowed $310,000 under the Canada Small Business Financing Programme to re-equip the bays, fixed at 6.35% on a fifteen-year amortisation. Ninety-two months and $214,000 remained — the single cheapest money anywhere in the transaction.
Two obstacles. The buyer’s adviser had assumed that a loan made to one company cannot follow the business into another, so the model refinanced it; that is wrong, and the provision that makes it wrong is easy to miss because it is filed under transfers between borrowers rather than under anything to do with acquisitions. The second was self-inflicted: the vendor wanted to keep the alignment rack and the bulk-oil system for a mobile-servicing business he intended to run in retirement, so the first draft of the asset schedule excluded them. Nobody connected that to the financing, because on any ordinary view a couple of retained assets is a price adjustment rather than a legal problem.
The closing structure that worked: $160,000 of buyer equity, a $120,000 vendor take-back over five years, the assumed programme loan at $214,000, and a new $91,000 programme advance for two replacement hoists and a wheel balancer.
The assumed loan and the new advance had to fit the prescribed maxima together, because after the assumption both belong to the same borrower: $305,000, inside the $500,000 maximum for purposes other than real property. Had the buyer carried programme debt in another company, as in the second newco that shared the first one’s ceiling, that would have been aggregated too and the assumption could have failed on arithmetic alone.
The saving was real money. Refinancing the whole $305,000 would have attracted a 2% registration fee of $6,100; assuming the existing facility and registering only the new advance cost $1,820. The rate difference on $214,000 between 2022 and 2026 was worth several thousand dollars a year on its own, and the buyer inherited an amortisation already three and a half years in.
Section 33(1) of the Regulations, marginal note Transfer of Loans Between Borrowers, is the only provision in the programme written for a change of ownership: “On the sale of all assets of a small business whose purchase or improvement is being financed by a loan, the borrower may be released by the lender from, and the purchaser may assume, liability in respect of the loan” on the conditions that follow.
Read the trigger first, because that is what the asset schedule nearly broke. It is the sale of all assets of the small business. A carve-out takes the transaction outside the subsection, and there is no partial-assumption route anywhere else in the Regulations. Both retained items came back into the sale at $22,000, which the new advance covered, and the vendor leased equivalent equipment for his retirement business.
The conditions are usually described as three. They are better read as four, because paragraph (a) carries two independent requirements: “the purchaser is approved by the lender as a borrower in accordance with the due diligence requirements referred to in section 8 and the outstanding loan amount is not greater than the applicable limit”. The first limb is a credit test; the second is the ceiling test, applied to the purchaser and, through the Act’s aggregation rule, to every borrower related to it. Advisers routinely list the three headline conditions and drop the second limb of the first one.
Paragraph (b) requires that “security of the same rank is maintained or taken by the lender on the assets that were used to secure the loan” — which on an asset sale means fresh registrations against a new debtor and the sequencing to make sure nothing intervenes. Paragraph (c) requires that a section 19 guarantee be “replaced with another guarantee or suretyship in accordance with that section of an equal or greater value”. That produced the only real surprise: the vendor’s guarantee had been given for the original loan amount of $310,000, not the current balance, so the buyer’s principals gave $310,000 of unsecured personal guarantee against a $214,000 balance. That is consistent with section 19 itself, where the post-2014 ceiling is measured on the original amount of the loan rather than what is left of it.
The provision has close relatives that are not the same thing. Sections 33(2) and (3) cover partnership changes — same conditions, no asset transfer to re-secure. Section 29(1) covers assignment between lenders, conditioned on the Minister’s remaining exposure to the transferor, and section 29(2) requires the Minister to determine whether those requirements are met, so it is not something two banks can simply agree. Section 30 lets a new lender make a loan to repay another’s. Four transfers, four tests.
The assumption completed at closing. The vendor was released from the loan and his personal guarantee released under section 21, which permits it only where the loan is in good standing. The buyer’s company became the borrower on the existing facility at the existing rate and took a separate $91,000 advance registered on its own.
A share purchase would have avoided the exercise entirely, and it is worth saying why: on a share deal the borrower does not change, so section 33(1) is never engaged and there is nothing to assume or release. The trade-offs run the other way, of course — pricing the difference between a share and asset deal sets them out.
The risk in a failed assumption is not that the buyer pays more. It is that the vendor stays on the hook: without a release the vendor company remains the borrower on a loan secured by assets it has just sold, and its principal remains liable on a guarantee for a business he no longer controls.
That is uncomfortable as a credit matter and worse as a compliance matter, because section 16(1)(b) of the Act makes it an offence for a person, “being a borrower, without the consent of the lender and with fraudulent intent, disposes of any assets taken as security for the loan where the disposition is not permitted by the regulations”. Nobody here was near that, because the lender was involved throughout. But it is why a quiet closing that moves the secured assets and sorts the paperwork out afterwards is not an option on a programme loan.
The tell was in the model: a line item refinancing $214,000 of debt at 6.35% with new debt at a higher rate, for no reason anybody could articulate. When a model replaces cheap money with expensive money, ask whether it has to.
The second tell was the retained-assets list. On most asset purchases a short list of excluded assets is a commercial detail; where the target carries a programme loan it is a trigger condition.
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.
