Anonymised, illustrative composite. A book allowance is not a tax write-off. Six years of carrying one had used up nothing, so a real deduction was still sitting there when the shares changed hands — and a deemed year-end decided which side of the table got it.
At a glance
A second-generation commercial printer with a folding-carton line, selling to regional food and cosmetics brands. The vendor was selling shares. Diligence was well advanced and the receivables had already been through the usual tests: ageing, concentration, subsequent receipts. Nothing on that front was contentious.
The item that had not been looked at was a static line on the balance sheet. The allowance for doubtful accounts had read $180,000 in every set of statements for six years. Not roughly $180,000 — exactly $180,000, in each of six consecutive years, against a receivable balance that had moved between $940,000 and $1,410,000 over the same period.
Both sides had been treating the allowance as a loss already taken. The vendor's price expectation was built on net receivables. The buyer's peg was built on net receivables. Everybody had agreed to write $180,000 off, once, six years ago, and move on.
For tax, nothing had happened. The buyer's accountant pulled the reconciliation of book income to taxable income for all six years and found no reserve claimed and no reserve added back. The corporation had computed its income as though the allowance did not exist — which, for tax purposes, it did not. Every dollar of those receivables was still in the corporation's income and had already been taxed. The deduction was unspent.
Which raised a question nobody had put on a checklist: it is worth real money, so who gets it? Not the corporation in the abstract — the corporation gets it either way. The question is which party's economic account it lands in, and on a share deal the answer is decided by a date.
Receivables $1,310,000 gross, allowance $180,000, net $1,130,000 in the peg. Of the $180,000, the buyer's accountant could support $124,000 as bad rather than doubtful: three debtors dissolved, two in bankruptcy with the trustee's final statement received, and six accounts with no contact and no payment for more than three years. The remaining $56,000 was live disputes and slow payers — doubtful, not bad.
So $124,000 of deduction was available immediately and $56,000 was not. On a closing date of 12 August, that deduction falls into one of two taxation years that did not exist a week earlier: the stub from 1 January to 11 August, and the stub from 12 August to the following year end.
The share purchase agreement defined pre-closing taxes as the vendor's account and indemnified the buyer for them. A $124,000 deduction taken in the first stub reduces pre-closing taxes, which reduces the vendor's indemnity exposure and increases the vendor's net. Taken in the second stub, it reduces the buyer's own tax. Same deduction, same corporation, two different beneficiaries, and the vendor's accountant had already scheduled the write-off for July.
Three provisions set the timing, and the distinction between the first two is the one most often collapsed. Income Tax Act s.20(1)(l) allows a deduction for a reserve, computed as “a reasonable amount in respect of doubtful debts” — but only for debts “that have been included in computing the taxpayer’s income for the year or a preceding taxation year”. It is elective and it is annual.
Section 12(1)(d) is the other half of the same machine: there must be included in income “any amount deducted under paragraph 20(1)(l) as a reserve in computing the taxpayer’s income for the immediately preceding taxation year”. Every reserve claimed is brought straight back the following year and must be claimed again from scratch. A s.20(1)(l) reserve is a one-year deferral, re-earned annually. It is never a write-off.
The write-off is s.20(1)(p)(i), which allows a deduction for “all debts owing to the taxpayer that are established by the taxpayer to have become bad debts in the year and that have been included in computing the taxpayer’s income for the year or a preceding taxation year”. Two conditions: the debt must have been in income, and it must be established to have become bad in the year. The first was satisfied years ago. The second is the one the parties were about to decide by default.
Then the deal mechanic. Section 251.2(2)(a) provides that a taxpayer is subject to a loss restriction event if “the taxpayer is a corporation and at that time control of the corporation is acquired by a person or group of persons”. Section 249(4)(a) then supplies the consequence: “the taxpayer’s taxation year that would, but for this paragraph, have included that time is deemed to end immediately before that time, a new taxation year of the taxpayer is deemed to begin at that time”. Closing cuts the year in two, without anyone electing anything. See the deemed year-end nobody budgets for.
Note the units carefully, because this is where the reasoning usually goes wrong. s.20(1)(l) governs a year — it is claimed or not claimed, annually, and lapses. s.20(1)(p) governs a debt, in the year that debt becomes bad. s.249(4) governs the taxation year, not the deduction. Put together, they mean an unclaimed allowance is not a stored-up asset that transfers with the shares; it is an unexercised opportunity whose value falls to whoever controls the timing of the write-off.
One boundary. On an asset sale the question is answered by a different provision, s.22, under which vendor and purchaser elect jointly and “the debts so sold shall be deemed, for the purposes of paragraphs 20(1)(l) and 20(1)(p), to have been included in computing the purchaser’s income for the taxation year or a previous year”. Without that deeming an asset buyer has no s.20(1)(p) deduction at all, because the debts were never in its income. This was a share deal, so s.22 had no application. See the section 22 election on accounts receivable.
The write-off was moved. The agreement was amended so that the vendor covenanted not to establish any debt as bad, or to claim any reserve, in the pre-closing stub without the buyer's consent, and the $124,000 was written off in the post-closing period on the buyer's instruction and supported by the trustee statements and dissolution records already collected in diligence.
A $124,000 deduction claimed in the vendor's stub, reducing pre-closing taxes the vendor was indemnifying anyway. In cash terms the buyer would have paid the same price for a corporation whose unused deduction had been spent for the other side's benefit in the last six weeks before it took over — and it would never have appeared as a line item anywhere, because a lower pre-closing tax liability looks like good news.
An allowance for doubtful accounts that is identical across several years. Estimates move; a number that repeats to the dollar is a plug, and a plug that has never been tested has usually never been claimed for tax either. Pull the book-to-tax reconciliation and look for the reserve line. If it is absent in every year, the deduction is intact and unpriced.
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.
