Anonymised, illustrative composite. The freeze itself worked exactly as designed. It was just twenty months old, and the Act asks for twenty-four.
At a glance
A founder of a specialty commercial printing company had been told by his accountant, more than once, to do an estate freeze while the company was still small — roughly $700,000 in value a decade before he eventually sold. He kept putting it off. When a strategic buyer made an unsolicited offer for the whole company at $5,000,000, the freeze finally happened, twenty months before the closing date, structured as a straightforward reorganization of capital.
The mechanics themselves were sound. Under ITA s.86(1) — marginal note Exchange of shares by a shareholder in course of reorganization of capital — a taxpayer who disposes of all the shares of a particular class in a capital reorganization is deemed by paragraph (c) to have disposed of the old shares “for proceeds of disposition equal to the cost to the taxpayer of all new shares and other property receivable by the taxpayer for the old shares.” Done properly, no gain is triggered on the exchange itself. Done carelessly it can be: s.86(2) claws the relief back where the old shares were worth more than the consideration received and the excess is reasonably regarded as a benefit conferred on a related person, which is precisely the shape of a freeze that under-values the founder’s preferred shares in order to push value at the children. The founder exchanged his common shares for fixed-value preferred shares redeemable at $4,300,000, the company’s value at the freeze date, and new common growth shares were issued directly to his two adult children, both active in the business, for a nominal subscription price. Family agreement allocated $700,000 of the eventual $5,000,000 sale price — the growth since the freeze — to those children’s shares, split evenly.
At sale, each child held shares worth $350,000, against a nominal cost base. The taxable half of each child’s gain, under the ordinary ITA s.38(a) one-half inclusion, was $174,950 — comfortably inside either child’s own $625,000 lifetime capital gains exemption under ITA s.110.6, if the shares qualified. They did not. Paragraph (b) of the Act’s definition of a qualified small business corporation share, in s.110.6(1), requires that, “throughout the 24 months immediately preceding the determination time,” the share “was not owned by anyone other than the individual or a person or partnership related to the individual.” The children’s shares had existed for twenty months. Twenty is four short of twenty-four.
It is worth being precise about why a brand-new share fails a test phrased as a test about ownership by unrelated people, because the reason is a separate provision and it is the one that actually decided this file. s.110.6(14)(f) — marginal note Related persons, etc. — deems shares issued by a corporation to a person to “have been owned immediately before their issue by a person who was not related” to that person, with three exceptions: shares issued “as consideration for other shares,” shares issued on a transfer of all or substantially all the assets of an active business, and shares issued “as payment of a stock dividend.” The children subscribed for their growth shares in cash, so none of the exceptions applied and their clock started on the day of issue. The founder’s preferred shares were issued at the same moment, twenty months before the same closing — but they were issued as consideration for his old common shares, squarely within the first exception, so his clock was never reset. Two share classes, one issue date, opposite outcomes, entirely on account of paragraph (14)(f).
Failing the 24-month test is not a discount or a haircut — it removes the exemption entirely for that share. Neither child could claim any part of their own $625,000 cap (the figure in s.110.6(2)(a), indexed for taxation years beginning after 2025 under s.117.1(2)(c)) against their $174,950 taxable gain. Combined, the two children paid tax on $349,900 of taxable capital gain that a freeze completed on schedule would have sheltered completely. The founder’s own preferred shares still carried his own $625,000 exemption against his much larger share of the gain — not because he had “held them for decades,” since the preferred shares themselves were minted at the freeze, but because s.110.6(14)(f)(i) exempts shares issued as consideration for other shares from the deeming rule that would otherwise have restarted his clock too.
Run the same $700,000 growth allocation through the freeze the accountant originally recommended, done ten years earlier instead of twenty months before closing. The children’s shares would have been held for well over 24 months by the time of any eventual sale, satisfying s.110.6(1)(b) with years to spare. Each child’s $174,950 taxable share sits well under their own $625,000 cap, so both gains would have been fully sheltered: $0 taxable, against $349,900 taxable in the twenty-months-too-late scenario actually run. The entire cost of waiting sat in a four-month gap on the calendar.
The tell is procrastination itself — but the checkable version of it is the share register. The moment a sale process starts, or even a serious inbound offer arrives, the issue date of every class of shares relative to any plausible closing date needs checking immediately, because the 24-month clock cannot be fixed retroactively once a transaction is in motion.
The accountant’s original advice was not really about tax efficiency at all — it was about optionality. A freeze done when the company was worth $700,000 costs almost nothing to unwind if a sale never materializes, and it buys years of runway against exactly this kind of test. A freeze done at $4,300,000, twenty months out, has no such margin: it has to work on the first try, against a clock that was already running before anyone decided to start it.
The founder’s own advisor put it plainly once the numbers were laid out side by side: a freeze is not a single decision made once, at whatever value the company happens to sit at when someone finally gets around to it. Its value depends entirely on how much runway sits between the freeze date and whatever comes next — a sale, an offer, a liquidity event nobody scheduled in advance. Ten years of runway costs nothing extra to build in. Twenty months, discovered only because a buyer showed up first, is what a missed decade actually looks like on a family’s tax return.
A 30-minute call can tell you whether a succession plan is running against a deadline nobody has counted yet.
