Anonymised, illustrative composite. The debt sat in a holding company whose only asset was shares. What put the deduction in question was not the absence of a dividend. It was a clause the equity lawyers had drafted saying there would never be one.
At a glance
A management buyer and a minority financial partner bought a commercial signage and fabrication business through the standard structure: a newly incorporated holding company borrowed $4,200,000 from a Schedule I bank and used it, with $2,600,000 of equity, to buy all the issued shares of the target. After closing the holding company's balance sheet held one asset — the shares — and one liability, the loan. The operating company traded well, but was using every dollar of free cash flow to retire its own equipment facility, so for three years it declared nothing.
Three years in, the holding company had deducted $995,400 of interest against no revenue of its own, generating losses it could not use, and the file was queried. The buyer's assumption — a common one — was that the problem was the absence of a dividend, and that the answer was to declare one.
It is not, and that misreads which unit the provision governs. The test in s.20(1)(c)(i) is about the purpose for which the borrowed money was used, assessed by reference to the borrower's own source of income. Whether a dividend was in fact paid in a given year is evidence, not the test. What creates the exposure is a documented position that dividends will never be paid — and the shareholders' agreement said exactly that, in a clause drafted for entirely sensible equity reasons: earnings would be reinvested, and holders would realise value on the year-five exit.
$4,200,000 at 7.9% is $331,800 of interest a year, or $995,400 over three years. That is the deduction in question. The tax consequence depends on the rate applicable to the corporation in the relevant years, which this file does not state and no reader should assume — what is at stake is the deduction, not a computed tax figure.
Two further amounts framed the response. The operating company's earnings before interest and tax averaged $1,940,000, so it could comfortably have declared dividends; and the group's total interest and financing expense was well under the thresholds that bring the excessive-interest rules into play, which is a separate question addressed in the EIFEL file.
Start with what interest actually is for tax purposes, because it explains why the test is so specific. Interest on account of capital is not deductible under the general rules; it is deductible only because a provision says so. ITA s.20(1) opens “Notwithstanding paragraphs 18(1)(a), 18(1)(b) and 18(1)(h)” and then permits, at paragraph (c), “an amount paid in the year or payable in respect of the year (depending on the method regularly followed by the taxpayer in computing the taxpayer's income), pursuant to a legal obligation to pay interest on (i) borrowed money used for the purpose of earning income from a business or property”.
The unit is the borrower's own source of income, traced through the use of the money. Not the group's income; not the target's income. A holding company that borrows to buy shares is borrowing to acquire property, and the only income that property can produce is dividends. That is a thinner trace than the same debt in the operating company, whose borrowed money would be linked directly to business assets — and it is precisely why deals amalgamate or wind up after closing.
The CRA's administrative position on the point is in Income Tax Folio S3-F6-C1, and it should be read as an administrative position rather than as the rule. At paragraph 1.70 it says: “Generally, the CRA considers interest costs in respect of funds borrowed to purchase common shares to be deductible on the basis that at the time the shares are acquired there is a reasonable expectation that the common shareholder will receive dividends.” It then names the exception directly: “If a corporation has asserted that it does not pay dividends and that dividends are not expected to be paid in the foreseeable future such that shareholders are required to sell their shares in order to realize their value, the purpose test will not be met.” And the safe formulation: “However, if a corporation is silent with respect to its dividend policy, or its policy is that dividends will be paid when operational circumstances permit, the purpose test will likely be met.”
Read against the shareholders' agreement, that is not a marginal call. The document asserted the no-dividend position and pointed holders at a sale to realise value — the exception, almost word for word.
The same folio supplies the structural answer at paragraph 1.44: where a corporation acquires shares with borrowed money and the acquired corporation is subsequently wound up or amalgamated with the borrower, “a link for the current use of the borrowed money is readily established between the shares that were initially acquired (and have disappeared) and the assets formerly held by the acquired corporation”. After a merger the trace no longer depends on a dividend expectation at all.
Two changes. The dividend clause was amended prospectively to the formulation the folio treats as acceptable — dividends payable when operational circumstances permit — which removed the asserted position going forward but could not rewrite the three years already filed. Then the two companies were amalgamated, so that the borrowed money, the business assets and the business income sat in one taxpayer. For the corporate mechanics of that step, and the conditions that can take it off the directors'-resolution route, see the vertical amalgamation that lost its short-form route.
The tax attributes of an amalgamation or wind-up — cost-base adjustments and the treatment of the acquired corporation's accounts — are a separate analysis and are deliberately not addressed here.
Leaving the clause in place while continuing to deduct would have compounded the exposure at $331,800 a year and left the losses stranded in a company with no income to apply them against. Declaring a single defensive dividend, which was the buyer's first instinct, would have moved cash and changed nothing: the asserted policy, not the payment history, is what the folio treats as decisive.
The tell was in the equity documents, not the tax file. An agreement promising reinvestment and an exit is good drafting for aligning investors. It is also a written assertion about dividend policy, and where a holding company has borrowed to buy shares that assertion is the only thing standing between the interest and the use test.
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.
