Anonymised, illustrative composite. A manufacturing business, one operating company, one holding company above it, and a buyer who would only do an asset deal. Selling the assets and then redeeming the vendor's shares out of the proceeds let both sides get what they were asking for — at a price the redemption dividend itself set.
At a glance
The target was a specialty manufacturer worth roughly $18,000,000, operated inside a corporation (“OpCo”) that was itself wholly owned by a holding company (“Holdco”) belonging to the founder. The buyer, a strategic acquirer, would negotiate only on an asset-purchase basis — it wanted a stepped-up cost base on the equipment and no exposure to OpCo's twenty-two years of corporate history. Holdco's adviser wanted the opposite: a share sale, because an asset sale out of OpCo triggers corporate-level tax on the disposition, and the cash then has to be moved up to Holdco and eventually to the founder — a second layer of tax an ordinary share sale avoids.
Neither side was going to move off its structural preference, and reframing the disagreement as a pricing argument would have just moved the same tax cost from one side's spreadsheet to the other's. What actually resolved it was recognising that an asset sale followed by a share redemption is not two competing structures — it is one sequence that gives the buyer the asset deal it insisted on while giving Holdco's shares an exit that is not a straight sale to the buyer at all.
OpCo sold its business assets to the buyer for $18,000,000. After corporate tax on the recapture and the gain on the sale, OpCo was left holding approximately $16,550,000 in cash. OpCo then redeemed the entirety of Holdco's shareholding for that amount, against nominal paid-up capital, producing a deemed dividend to Holdco of $16,549,900. Of that, $11,200,000 fell within the shares' accumulated safe income and was sheltered; the remaining $5,349,900 was recharacterised as an ordinary capital gain to Holdco, taxed at Holdco's own corporate rate rather than deducted — a real cost, quantified before closing rather than discovered after it.
ITA s.84(3) is what turns a share redemption into a dividend at all. Where a corporation “has redeemed, acquired or cancelled in any manner whatever… any of the shares of any class of its capital stock,… the corporation shall be deemed to have paid at that time a dividend… equal to the amount, if any, by which the amount paid by the corporation on the redemption… exceeds the paid-up capital in respect of those shares immediately before that time”, and the redeemed shareholder — here, Holdco — “shall be deemed to have been received” a matching dividend.
That dividend is what made the structure work for Holdco. ITA s.112(1) lets a corporation deduct “an amount equal to the dividend… from the income of the receiving corporation for the year for the purpose of computing its taxable income” where the dividend is received from a taxable Canadian corporation. Paid correctly, the redemption dividend from OpCo to Holdco is not taxed twice — it is taxed once, at OpCo, on the asset sale, and then moves up to Holdco without a second layer.
s.55(2) is the reason it did not move up entirely clean. Section 55(2.1)(b)(i) extends the anti-stripping rule expressly to a redemption: “one of the purposes of the payment or receipt of the dividend (or, in the case of a dividend under subsection 84(3), one of the results of which) is to effect a significant reduction in the portion of the capital gain that, but for the dividend, would have been realized on a disposition at fair market value of any share… immediately before the dividend”. Extinguishing Holdco's shares by redemption plainly has that result. Section 55(2.1)(c) then caps how much of the dividend can stay a dividend at all: it applies where “the amount of the dividend exceeds the amount of the income earned or realized by any corporation… that could reasonably be considered to contribute to the capital gain that could be realized on a disposition at fair market value… of the share on which the dividend is received” — the shares' safe income. Below that line, s.112(1) applies as intended. Above it, s.55(2) deems the excess to be “a gain of the dividend recipient… from the disposition of a capital property”, which is exactly the $5,349,900 above.
This is not the same computation as a pre-sale purification dividend testing whether shares qualify for the small business deduction — there is no purification here, and the safe-income constraint is a limit on this file's redemption structure rather than the whole story of the file.
Both structural preferences were satisfied. The buyer closed an asset purchase with a stepped-up cost base and no assumed corporate history. Holdco exited through a redemption rather than a sale to the buyer, kept the s.112(1) deduction on the safe-income portion, and accepted the capital gain on the excess as a known number agreed before signing rather than a position CRA discovered on reassessment.
Had Holdco simply sold its OpCo shares to a buyer who would only do an asset deal — which was not on offer here — there would have been no redemption dividend and no s.112(1) deduction at all: the whole gain on the shares would have been an ordinary capital gain to Holdco, with none of it sheltered. Structuring the exit as a redemption converted $11,200,000 of that gain into a deductible inter-corporate dividend instead of a taxed capital gain — the safe-income cap is a real limit, not a reason to skip the structure.
Both advisers initially treated the disagreement as a pricing gap to be split, because each was assuming the other side's preferred structure was the only alternative to its own. The fix showed up only once someone asked whether the transaction had to be one sale at all, rather than an asset sale followed by a separate redemption.
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.
