Skip to content

Objective — shift investment income from a higher-income spouse to a lower-income spouse legally, so the same portfolio is taxed at a lower rate.

Risk level — 1 — Low risk. Risk justification — no borrowing from a third party and no increase in household market exposure; capital that was already invested is invested by a different family member. The risk is structural and legal — the attribution exception fails if a condition is missed. Benefit justification — unverified — needs Talbot. Jurisdiction — Canada. Depends entirely on ITA §74.5(2); no U.S. read-across.

  • A genuine and material spread between the spouses’ marginal tax rates.
  • Capital the higher-income spouse can lend — own capital, not borrowed. See Risks.
  • A written loan agreement bearing interest at the CRA prescribed rate in effect when the loan is made.

The attribution rules (ITA §74.1–§74.5) normally attribute income from property transferred or loaned to a spouse back to the transferor, defeating income splitting. §74.5(2) is the exception: attribution does not apply if the property is loaned at the CRA prescribed rate and the borrower actually pays that interest by January 30 of the following year.

The lower-income spouse invests the loan proceeds and keeps the spread between the investment return and the prescribed rate, taxed at their own lower rate. The prescribed rate is set quarterly from the 90-day T-bill rate; the rate locked at inception applies for the life of the loan.

  • Permanent reduction in household tax on investment income, with no change in the investments held.
  • Most valuable when the prescribed rate is low and the spousal rate gap is wide.
  • Missing the January 30 interest payment, in any year, kills the exception and attribution applies. This is the whole strategy in one deadline.
  • Combining this with borrowing to invest is a documented trap. If a taxpayer borrows and lends the proceeds to a spouse below the prescribed rate, the investment income is still attributed back — creating a “potential double disadvantage.” The Lipson strategy attempted to exploit this structure with an interest deduction and GAAR denied it.
  • §74.5(11) disapplies the attribution rules where a main reason for the transfer was to reduce the transferor’s income — an anti-avoidance provision to be checked, not assumed away.
  • Late or unpaid January 30 interest.
  • No written agreement, or a rate other than the prescribed rate at inception.
  • Layering the loan on top of third-party borrowing — the Lipson shape.
  • Assuming a later drop in the prescribed rate applies retroactively; it does not.

Legal drafting of the loan; the interest itself is taxable to the lending spouse (which is the point — it is taxed at their rate, but only on the prescribed-rate amount, not on the whole return).

ITA §74.1–§74.5 with the §74.5(2) exception; ITA §245 (GAAR) as applied in Lipson; §74.5(11) anti-avoidance. TOSI (§120.4) does not apply directly here but shapes the wider family-splitting landscape.

Households with a large, durable marginal-rate gap and non-registered capital to lend.

  • Households with similar marginal rates — the machinery is not worth it.
  • Anyone who would fund the loan with borrowed money. See Risks.
  • Anyone who will not reliably make a payment by January 30 every year.
  1. Confirm the current CRA prescribed rate — it is set quarterly.
  2. Execute a written loan agreement at that rate.
  3. Transfer the funds; the borrowing spouse invests in their own name.
  4. Pay the interest by January 30 each year, without exception, and keep evidence.
  5. The lending spouse reports the interest received; the borrowing spouse deducts it and reports the investment income.

external-sourced — ITA §74.5(2) and the Lipson/GAAR interaction, via Core/_WorkingOn/Research/canada-investment-taxation.md §13.2. unverified — needs Talbot: the current prescribed rate. The source records 5% for 2024 Q1, “trended lower through 2025”, and instructs checking CRA quarterly announcements — so no rate is stated here.

  • This is a tax strategy, not a debt strategy, and arguably outside a $MART DEBT library. It is included because it is a loan between parties whose terms are the entire strategy — but the scope question is legitimate and routes to the CEO.
  • Loan to a family trust for minor children rather than a spouse. Mechanics not sourced here — unverified — needs Talbot.

corporate-borrowing-to-invest · claim-investment-interest-deduction

  • Core/_WorkingOn/Research/canada-investment-taxation.md §13.2 — attribution rules, the §74.5(2) prescribed-rate exception, the January 30 condition, prescribed-rate mechanics, the leverage interaction and Lipson
  • Same file §13.3 (TOSI), §13.1 (GAAR)
  • The current prescribed rate — must be read from CRA at the time of use, never from this record.
  • Whether income-splitting strategies belong in a debt-strategy library at all. CEO scope call.