A loss pool priced into an asset purchase

The situation

An Ontario precision-parts manufacturer operated two leased plants supplying automotive and industrial customers under long-term contracts. A larger contract manufacturer agreed to buy the operating business for $6,400,000, structured as an asset purchase rather than a share purchase: one leased site carried a pending environmental remediation order the buyer wanted no part of, and the buyer wanted a stepped-up cost base on $2,850,000 of equipment nearing the end of its capital cost allowance class. The agreement identified equipment, inventory, the customer contract book and goodwill as the assets sold. The vendor corporation itself — its shares, its history, everything it did not sell — stayed exactly where it was, owned by the same shareholders after closing as before.

The problem

The vendor corporation’s balance sheet carried $1,850,000 of non-capital losses from a 2020–2022 downturn. The buyer’s corporate development associate built the offer model by adapting a template from the firm’s previous acquisition — a share purchase, where a target’s loss pool is at least a restricted tax attribute the buyer inherits with the shares. The shelter-value line carried straight across: $1,850,000 at a 26% combined rate is $481,000 of notional value, discounted for timing to a $260,000 addition to price, pushing the signed letter of intent to $6,660,000. Nobody asked, before that number reached the LOI, whether this transaction could acquire the thing being priced.

The numbers

$6,660,000 at LOI — $6,400,000 for the identified assets (equipment $2,850,000; inventory $780,000; customer contracts, backlog and goodwill $2,770,000) plus a $260,000 loss-pool premium. The vendor corporation’s $1,850,000 non-capital loss pool, modelled at $481,000 of shelter value and discounted to $260,000, was the only line in the LOI price with no asset behind it that the Asset Purchase Agreement was ever going to transfer.

The rule that decided it

Both governing subsections key the deduction to one taxpayer, not to whoever later buys the business it ran. Section 111(1) provides that “For the purpose of computing the taxable income of a taxpayer for a taxation year, there may be deducted such portion as the taxpayer may claim of the taxpayer’s … non-capital losses for the 20 taxation years immediately preceding and the 3 taxation years immediately following the year”. Even where a corporation’s own shares change hands, s.111(5) — the streaming rule this file’s sibling case works through, see non-capital losses after an acquisition of control — provides that “no amount in respect of the taxpayer’s non-capital loss, restricted interest and financing expense or farm loss for a taxation year that ended before that time is deductible by the taxpayer for a taxation year that ends after that time”. The deduction runs to the taxpayer in both subsections; neither contemplates a second, different taxpayer stepping into it.

Read the unit. An asset purchase does not acquire the vendor corporation. Its shareholders keep their shares; the operating assets simply move to a different corporation the buyer controls. Whether the vendor corporation is itself ever “subject to a loss restriction event” turns on ITA s.251.2(2)(a): “the taxpayer is a corporation and at that time control of the corporation is acquired by a person or group of persons”. That asks who controls the vendor corporation after the transaction — not who bought its assets. Here the vendor’s own control never changed, so it was never subject to a loss restriction event, and remained free to use its own losses against its own future income if it went on to earn any. What it could never do is lend that entitlement to the buyer’s acquisition company, which was never the taxpayer that incurred the loss.

The absence that matters. The Act lets one corporation inherit another’s losses in exactly two places, and a purchase of assets is neither. On an amalgamation, s.87(2.1), marginal-noted Non-capital losses, etc., of predecessor corporations, provides that for the purpose of “determining the extent to which subsections 111(3) to (5.4) and paragraph 149(10)(c) apply to restrict the deductibility by the new corporation of any non-capital loss, net capital loss, restricted farm loss, farm loss, limited partnership loss or restricted interest and financing expense, as the case may be, the new corporation shall be deemed to be the same corporation as, and a continuation of, each predecessor corporation”. A parallel deeming for a wound-up subsidiary sits in s.88(1.1)–(1.2). Both provisions do real work: they make one corporation a continuation of another, by name. Nothing performs that deeming for a purchase of assets — which is the whole reason the $260,000 line item never had anywhere to attach.

The outcome

The buyer’s own tax counsel flagged the $260,000 line while reviewing the asset allocation schedule, asking the one question the model had skipped: whose corporation are we buying the losses from? The vendor’s counsel pushed back once, arguing the buyer was acquiring the full economic benefit of a going concern that had generated those losses — a commercial argument, not a tax one, and it does not engage s.111 at all. Economic benefit is not the test; taxpayer identity is. The parties repriced to $6,400,000 and closed on the same structure. Nobody proposed a share purchase instead — the environmental carve-out and the stepped-up cost base were worth more than $260,000, and buying the shares would only have put the pool through the restricted, streamed test in s.111(5) rather than handed it over intact.

What it would have cost otherwise

Had the $260,000 stayed in with nobody asking the structure question, the buyer would have paid full value for a tax attribute its own acquisition company had no lawful route to deduct. That is a different failure from the risk in a share purchase, where a real loss pool survives in restricted form — here there was no pool on the buyer’s side to begin with. No post-closing quality-of-earnings work would have surfaced it, because the loss was never going to appear on the buyer’s own tax return; it would simply have been $260,000 paid for nothing, discovered only once that return was filed with no loss to deduct.

The tell

The tell sat upstream of the numbers, in the model itself: a loss-carryforward shelter-value line surviving, unchanged, out of a share-deal template and into a workbook headed Asset Purchase — Sources & Uses. A financial model built for one deal structure carries its assumptions with it when it is reused for another. The fix is not a bigger diligence team; it is one question asked before a number reaches a letter of intent — are we buying this corporation, or only what it owns?

Takeaways

• • ITA s.111(1) and s.111(5) both confine the loss deduction to “the taxpayer” that incurred it — not to whoever later buys the business the loss-making corporation ran.
• • An asset purchase does not acquire the vendor corporation. Its shares, and everything the Act attaches to those shares, stay exactly where they were.
• • The Act lets one corporation inherit another’s losses in exactly two places — amalgamation under s.87(2.1) and a subsidiary’s wind-up into its parent under s.88(1.1)–(1.2) — and both work by expressly deeming one corporation to be a continuation of the other. No equivalent deeming applies to a purchase of assets.
• • Contrast directly with the share-purchase case: buying the shares at least puts the loss pool through the restricted survival test in s.111(5); buying the assets skips that test only because there is nothing left to test.
• • A financial model reused from a different deal structure is a live source of pricing error. Check that every line item survives the structure actually being used before it reaches a letter of intent.

Sources

• • Income Tax Act, R.S.C. 1985, c. 1 (5th Supp.), s.111 — s.111(1), marginal note Losses deductible, paragraph (a): the general non-capital-loss carryforward, confined to the taxpayer that incurred the loss; s.111(5), marginal note Loss restriction event — certain losses and expenses: the streaming rule quoted above, which likewise runs only “by the taxpayer”.
• • Income Tax Act s.251.2(2)(a) — a corporation is subject to a loss restriction event only when control of that corporation is acquired; an asset purchase does not acquire the vendor corporation, so this test was never in play on the buyer’s side.
• • Income Tax Act s.87(2.1), marginal note Non-capital losses, etc., of predecessor corporations — the amalgamation continuity deeming quoted above, cited to show what a purchase of assets lacks.
• • Income Tax Act s.88(1.1)–(1.2) — the parallel continuity mechanism where a subsidiary is wound up into its parent (s.88(1) itself, Winding-up, is the general wind-up provision). Named for contrast; not quoted here.
• • Treadstone Law — what happens to loss carryforwards if I buy shares instead of assets — directly on point: its own plain-language answer states that “in an asset purchase…the seller’s loss carryforwards generally stay behind with the selling corporation”, consistent with the statutory analysis above, which is the operative source for the rule itself.