Anonymised, illustrative composite. The freeze was executed correctly. Nothing was filed late and no election was missed — because section 86 needs none. The only variable was the number, and the number was wrong.
At a glance
A founder aged 61 owned all 1,000 common shares of an Ontario specialty food manufacturer, with an adjusted cost base of $1,000 in total. He wanted to cap his own estate at the current value of the business and let his two adult children take the growth from there. His accountant executed a standard estate freeze: the founder exchanged all of his common shares for redeemable preferred shares with a fixed redemption amount, and a discretionary family trust for the children subscribed for new common shares for $100.
That transaction sits squarely inside subsection 86(1) of the Income Tax Act, which applies where, “in the course of a reorganization of the capital of a corporation”, a taxpayer disposes of capital property that was “all the shares of any particular class of the capital stock of the corporation that were owned by the taxpayer at the particular time” and receives other shares of the same corporation. There is no election, no prescribed form, and no filing deadline. Section 86 simply applies.
The redemption amount on the preferred shares was set at $4,200,000. That figure came from a one-page internal calculation: normalised EBITDA of $1,050,000 multiplied by four. There was no valuation report, no comparable-transaction evidence, and no price-adjustment clause in the share provisions or in the reorganisation resolutions.
Twenty-two months later the business was sold to a strategic buyer for $6,100,000 on materially the same earnings base. The CRA opened the freeze year and took the position that the fair market value of the common shares at the freeze date had been $5,600,000. The difference was not a disagreement about the future. It was a disagreement about the day of the freeze.
Because the founder took back only $4,200,000 of value for shares worth $5,600,000, and the beneficiaries of the gap were his own children, the CRA applied subsection 86(2) rather than 86(1).
Subsection 86(2) engages where the fair market value of the old shares immediately before the disposition exceeds the total of the boot and the fair market value of the new shares, and it is reasonable to regard the excess “as a benefit that the taxpayer desired to have conferred on a person related to the taxpayer”. Here: $5,600,000 less nil boot less $4,200,000 of preferred shares = a gift portion of $1,400,000.
Paragraph 86(2)(c) then provides that “the taxpayer shall be deemed to have disposed of the old shares for proceeds of disposition equal to the lesser of” the boot plus the gift portion, and the fair market value of the old shares immediately before the disposition. That is the lesser of $1,400,000 and $5,600,000 — so $1,400,000. Against a $1,000 adjusted cost base, the founder had a capital gain of $1,399,000 in a year in which he received no cash at all.
He could not shelter it. He had crystallised his lifetime capital gains exemption in a 2019 reorganisation and had none left. Paragraph 86(2)(d) removed the other obvious answer in advance: “the taxpayer’s capital loss from the disposition of the old shares shall be deemed to be nil”, so a valuation error in the opposite direction produces no deduction to offset one in this direction.
The part that costs the money is paragraph 86(2)(e), and it is the part nobody reads. Under the ordinary rule in 86(1)(b), the cost of the new shares picks up the adjusted cost base of the old ones. Under 86(2)(e), the cost of the new shares is the amount, if any, by which the adjusted cost base of the old shares exceeds the total already deemed to be proceeds under 86(2)(c)(i). Here that is $1,000 less $1,400,000 — a negative number, so nil.
So the founder was taxed on $1,399,000 and received nothing for it. His $4,200,000 of preferred shares carry a cost base of zero. When those shares are redeemed, subsection 84(3) deems a dividend equal to the redemption proceeds over their paid-up capital; if they are instead sold, the gain runs from nil. The $1,400,000 he has already paid tax on is not added back anywhere in the system.
Be precise about who loses. Nothing in section 86 takes cost base from the children or the trust; their new common shares cost what they paid, as in any freeze. The grind falls on the transferor. A freeze that fails on the calendar costs growth that is already gone; a freeze that fails on the valuation costs the founder tax and cost base at once, on value not yet realised. Compare a freeze done ten years too late.
There was nothing to restructure. The freeze had happened, the trust owned the common shares, and the business had been sold. The founder settled the freeze-year reassessment on a valuation of $5,300,000 rather than $5,600,000, which reduced the gift portion to $1,100,000 and the deemed gain to $1,099,000. Arrears interest ran from the balance-due date for the freeze year, two years earlier.
The corrective work was forward-looking: the family's remaining holdings were re-papered with a price-adjustment clause before any further reorganisation was attempted. On the sale itself the trust's QSBC share status held and the children's gains were sheltered — the whole cost of the error sat with the founder.
If the redemption amount had been fixed at fair market value, subsection 86(1) would have applied instead of 86(2): no gain, and the preferred shares would have carried the $1,000 of cost base rolled over from the common shares. The entire exposure came from a single number in the share conditions.
A price-adjustment clause — a term that adjusts the redemption amount if the fair market value is later determined to be different — is the standard protection. No statute governs such clauses; their effect rests on the CRA's administrative position, and no paragraph number is cited here because the current statement of that position could not be retrieved from a primary source. The point that does not depend on it is in the Act: subsection 86(2) measures the gift portion at the freeze date.
Two things were visible on the file before the reassessment. The freeze multiple was four times EBITDA; the broker's teaser for the same business, prepared fourteen months later, opened at 5.5 to 6 times on the same earnings base, and the sale cleared at 5.8. Nothing in the business had changed. That is not a market movement, it is a valuation that was low at the outset.
The second tell was the absence of a price-adjustment clause in a reorganisation whose whole tax result turns on one estimated number.
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.
