Anonymised, illustrative composite. Two unrelated business partners, one corporately-owned life insurance policy on each, and a buy-sell agreement that worked exactly as designed when the first partner died. The design itself is what left the survivor holding twice the company at the same cost base as before.
At a glance
Two unrelated business partners had run OpCo together for years, 50/50, and had funded their buy-sell agreement the common way: the corporation owned a life insurance policy on each of them, was the named beneficiary, and would use the proceeds to redeem the deceased partner's shares from the estate. When Shareholder A died, the structure worked exactly as designed — OpCo received the $3,000,000 death benefit and redeemed A's shares from the Estate for the same amount, funded entirely by the payout.
The redemption closed cleanly and the Estate was paid promptly. What the buy-sell agreement's design had not addressed was what happened to Shareholder B's own cost base once the dust settled. B now owned 100% of a company that had just spent $3,000,000 buying out the other half — and B's adjusted cost base in the shares B had always held did not move by a dollar. The corporate redemption paid the Estate efficiently. It did nothing for the survivor's own tax position.
Death benefit received by OpCo: $3,000,000. Adjusted cost basis of the policy immediately before death: $180,000. Redemption price for A's 50% stake: $3,000,000, matched exactly to the insurance proceeds. Capital dividend account credit: $2,820,000. B's adjusted cost base in OpCo's shares before and after the redemption: unchanged. If B were to sell the now-wholly-owned company for its current value, the capital gain would be computed against B's original, low cost base — with no credit for the $3,000,000 the corporation itself paid to acquire the other half.
The tax-free funding side of this worked because of two specific provisions, not because life insurance proceeds are generally untaxed. ITA s.89(1)'s definition of capital dividend account credits a corporation, on the death of an insured person, with “the proceeds of a life insurance policy… received by the corporation… in consequence of the death of any person… exceed[ing]… the adjusted cost basis… immediately before the death… of… a policy… to the corporation”. That is the $2,820,000: proceeds minus adjusted cost basis, not the full death benefit. Section 83(2) is what let OpCo actually pay that amount out tax-free: where a private corporation elects on a dividend, “the dividend shall be deemed to be a capital dividend to the extent of the corporation's capital dividend account immediately before the particular time”, and “no part of the dividend shall be included in computing the income of any shareholder”.
s.70(5.3) is the reason the Estate's own deemed disposition of A's shares was not inflated by the same death benefit. On death, s.70(5) deems A to have disposed of the OpCo shares at fair market value immediately before death — a moment at which OpCo's own assets included a policy about to pay out $3,000,000. Section 70(5.3) provides that for valuing property deemed disposed of on death, “the fair market value at that time of any life insurance policy, under which the particular individual… was a person whose life was insured, were the cash surrender value … of the policy immediately before the particular individual died”. Pinning the policy's contribution to OpCo's value at cash surrender value, rather than the imminent death benefit, keeps the Estate's own deemed-disposition gain on the shares from double-counting a payout that shows up instead through the corporate dividend mechanics above. The two provisions divide where the value is taxed — not whether it is.
None of that reaches B's cost base, because it was never the question those provisions answer. The redemption is a transaction between OpCo and the Estate. B's shares were never sold, transferred, or redeemed — so nothing in s.89(1), s.83(2) or s.70(5.3) has any occasion to touch B's own adjusted cost base, and it stays exactly where it was.
The Estate was paid quickly and almost entirely tax-free, and OpCo's ownership consolidated in B without a dispute. B's adviser flagged the cost-base position going forward — not to unwind anything, since the redemption had already closed and worked as intended for its actual purpose, but so that B understood the shape of the eventual tax bill on a future sale and could plan the exit, or a further reorganisation, with that fact priced in from the start rather than discovered at closing.
The alternative structure — a criss-cross arrangement where B personally owned and was beneficiary of the policy on A's life — would have delivered the same $3,000,000 to fund the same purchase, but through B personally. Life insurance proceeds paid to a named individual beneficiary are not included in the recipient's income, so B would have received the $3,000,000 personally with no tax cost, and then used it to buy A's shares directly from the Estate for $3,000,000 — a purchase, with an ordinary cost base equal to what B paid. B's total adjusted cost base in the now-wholly-owned company would have been the original nominal amount plus $3,000,000, sharply reducing the gain on a future sale. The cost of that better outcome is that the premiums funding a personally owned policy are paid with after-tax personal dollars, rather than by the corporation at its own, typically lower, tax cost — a real trade-off the original buy-sell agreement never weighed against the cost-base consequence it was locking in.
The buy-sell agreement's insurance section was drafted entirely around funding the redemption — who is insured, how much, who is beneficiary — and said nothing about what either partner's cost base would look like afterward. The funding mechanism and the cost-base outcome were treated as the same decision. They are not.
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.
