Anonymised, illustrative composite. Most records problems are about what was thrown away. This one is about what was never made — which is a different obligation, a different offence, and a different loss for the buyer.
At a glance
A strategic buyer in the same sector agreed to acquire the shares of a Saskatchewan logistics operator, eleven years old and thirty-one employees. The $3,750,000 price was struck on a multiple of adjusted earnings and included value for $412,000 of non-capital losses carried forward.
The vendor had run it with a part-time bookkeeper, a spreadsheet and a filing cabinet. Returns were filed, in most years, from a year-end package the external accountant built out of bank statements. None of that is unusual at this size.
The buyer’s accountants asked for the general ledger. One existed for the two most recent years, from the 2022 move to accounting software. For the nine years before that there was no ledger, no trial balance and no sub-ledgers. Sales invoices had been kept because customers wanted copies; purchase invoices had not. Fuel and repairs, the largest expense line in this business, existed only as bank debits and a supplier statement the supplier no longer held.
This is not a retention failure — nothing had aged out and been destroyed. The records were never created in a form from which the tax could be worked out, which is what the statute requires. Two taxation years had also never been filed.
For the buyer the loss is not the vendor’s exposure. After closing the buyer is the one who has to substantiate the corporation’s positions, with nothing to substantiate them from.
Eleven years of trading, two years of usable ledger. $412,000 of non-capital losses claimed on the corporation’s own returns, generated in years with no supporting records. $186,000 of input tax credits across two of those years. Two taxation years unfiled.
The unfiled years matter more than they look: the retention clock does not start until the return is filed, so nothing in those years had aged out. The corporation was still required to have the records, and did not.
The buyer valued the loss pool and the credit position at nil and cut the price by $310,000. A pre-closing reconstruction of the most recent twenty-four months from bank and supplier statements cost $47,000 and produced usable records for two of the six years still open. The other four were priced, not fixed.
Two Acts impose the obligation and they are not worded the same.
Income Tax Act s.230(1) requires every person carrying on business to keep records and books of account “in such form and containing such information as will enable the taxes payable under this Act or the taxes or other amounts that should have been deducted, withheld or collected to be determined”. That is a functional standard. A filing cabinet of sales invoices is records; it is not records in a form from which the tax can be determined, and the section is not satisfied by volume.
Retention runs under s.230(4)(b), which requires records to be kept, “together with every account and voucher necessary to verify the information contained therein”, until six years from the end of the last taxation year to which they relate. Where a return was never filed, s.230(5) restarts the clock: the records must be retained “until the expiration of six years from the day the return for that taxation year is filed”. And under s.230(8) early disposal is permitted only “if written permission for their disposal is given by the Minister”. There is no self-service exemption anywhere in the section.
Section 230(3) is often read as the consequence. It is not a cure: where a person has failed to keep adequate records the Minister may require records the Minister specifies, and the person “shall thereafter keep records and books of account as so required”. Forward-looking, and no help on the nine years already missing.
The Excise Tax Act runs in parallel but not identically. Section 286(1) requires a person to “keep all records that are necessary to enable the determination of the person’s liabilities and obligations under this Part or the amount of any rebate or refund to which the person is entitled”, s.286(3) sets six years after the end of the year to which they relate, and s.286(2), headed Inadequate records, provides that “Where a person fails to keep adequate records for the purposes of this Part, the Minister may require the person to keep such records as the Minister may specify and the person shall thereafter keep the records so specified.”
The enforcement architecture then diverges, and it is worth knowing which way. Income Tax Act s.238(1) makes failure to comply with any of sections 230 to 232 a summary conviction offence carrying “a fine of not less than $1,000 and not more than $25,000”, or that fine and imprisonment for up to 12 months. The Excise Tax Act s.326(1) offence is narrower: it catches a person “who fails to comply with subsection 286(2) or 291(2) or with an order made under subsection (2)”. Under the ETA the offence attaches once the Minister has specified records and the person still fails — not to the bare failure to keep them in the first place.
Three things went into the agreement.
The price came down $310,000, being the full value the vendor had attributed to the loss pool and the credit position. The buyer’s position was not that the losses were wrong; it was that they could not be defended, which for a buyer is the same thing.
A condition precedent required the vendor, at its own cost, to reconstruct twenty-four months of purchases and payroll from bank records and third-party statements before closing.
The two unfiled years were made a specific, uncapped indemnity with its own escrow, because the buyer had no way to size them and no records with which to argue about them later. Compare reconciling tax filings to the internal accounts, which is what a buyer does when there are internal accounts to reconcile to.
The visible cost is the $412,000 loss pool and the $186,000 of credits, both of which the buyer would have paid for. The real cost is quieter and arrives later.
Every buyer protection in a share purchase agreement is proved with documents created before closing. With no ledger for nine years the buyer cannot prove a breach, cannot quantify one, and cannot resist a reassessment on the merits either. It carries the burden without the evidence, for the six years s.230(4) and s.286(3) keep open.
A long survival period is no substitute. It gives you time to bring a claim, not anything to bring one with.
A data room where every folder contains a document and no folder contains a ledger.
The test is cheap: ask for a trial balance for the oldest year you intend to rely on, and for the source documents behind one expense line on it. A business that produces both in a day has records within s.230(1). A business that offers to pull something together is telling you they do not exist in the form the section requires, and the answer is a price adjustment, not a longer request list. See what buyers ask for first and how to prepare it.
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.
