Smith Manoeuvre risks, and when the strategy does not make sense

Most of what is written about this strategy is written by people who would like you to adopt it. This page is the other half: what it costs, what can go wrong, and the circumstances in which the arithmetic does not support it. It is the page worth reading first, and it sits under the same hub as everything else about borrowing to invest in Canada.

The case for it, stated plainly

Interest on money borrowed to earn income from a business or property can be claimed as a carrying charge, and interest on a mortgage against your own home cannot. A household with a large mortgage and a long amortisation is therefore paying a great deal of interest that does nothing on its return, and the Smith Manoeuvre is a mechanical way of moving that interest from the first category into the second while a portfolio is built alongside it. That is real, it is not exotic, and at a high marginal rate the amount of interest moved from the first category to the second is substantial.

Everything below is what that sentence leaves out.

Smith Maneuver pros and cons, side by side

For	Against
Interest on the re-borrowed money can be claimed, where mortgage interest cannot	Total debt does not fall; it changes character and then grows as it is re-borrowed
A portfolio is built decades earlier than it otherwise would be	The portfolio is bought with borrowed money, so a drawdown is felt against a fixed obligation
The mechanism runs off payments you are already making	The credit line’s interest is a new monthly cost, in cash, from month one
The refund loop compounds the conversion	The refund is not guaranteed; it depends on the rest of your return
The paperwork is the same every month	The paperwork is the whole risk, and it runs for twenty-five years

Leverage: the debt does not shrink, it changes name

This is the point most readers miss on the first pass. At the end of a full conversion the mortgage is gone and a credit-line balance of roughly the same size stands in its place. The household is exactly as indebted as it was, and now owns a portfolio bought with that debt.

That is the trade, and it is a defensible one — but it is a trade, not a saving. A household that could not comfortably carry the original mortgage through a bad two years cannot carry this either, and the version with a portfolio attached fails in a more uncomfortable way.

Rate risk and sequence risk

Two different things, often collapsed into one.

Rate risk is that the credit line is variable. Its rate moves with prime, usually immediately, while the mortgage it replaces may have been fixed for years. A conversion begun in a low-rate year can be carrying a materially higher cost three years later with the same portfolio behind it.

Sequence risk is that the order of returns matters when you are investing steadily into a leveraged position. The average return over twenty-five years can be exactly what the projection assumed, and the outcome still be worse than projected, because the poor years landed while the portfolio was small and the good years landed after the borrowing was done. No projection built on a single expected return can show this. The projection published on this site prints the assumption set behind its figures for exactly that reason: the limit stays visible next to the number.

The risk that has nothing to do with markets

Contamination is what happens when borrowed money that was tracked to an income-producing use becomes mixed with personal funds or personal spending. The borrowed dollars can no longer be traced to the use that made their interest deductible, so the deduction is at risk for the contaminated portion of the balance.

Contamination is the failure this segment underestimates most, because it does not feel like a risk while it is happening. A line used once for a vehicle, a draw that sat in a chequing account for six weeks before it was invested, interest capitalised onto the line during a tight year, a portfolio sold and partly spent — each one breaks the connection between a borrowed dollar and the use that supported claiming its interest.

Nothing announces it. The consequence appears years later, when the connection has to be demonstrated and cannot be. What deductibility turns on is the tax side of this, what a mixed line does to the arithmetic is what it costs, and keeping the trail intact is the practical side.

Why does the net advantage here look smaller than on other calculators?

Because it is net of the two things most calculators leave out: the line of credit you still owe at the end, and what the same out-of-pocket cash would have grown to unlevered. Subtract both and the remaining advantage is smaller, later, and far easier to defend.

When it does not make sense

Several situations where the arithmetic or the circumstances argue against it, and a page that only listed the first five paragraphs would be leaving them out:

• A short remaining amortisation. The conversion happens at the rate principal is repaid, and it needs years of payments to reach a meaningful balance. With eight years left there is little to convert, and the same leverage can be taken in one draw.
• A low marginal rate. The value of a carrying charge scales with the rate it is claimed against. At the bottom of the schedule the deduction is thin and the leverage is unchanged.
• Income that is unstable. The line’s interest is payable monthly regardless. A variable income is precisely the case where it gets capitalised, which is where the tracing starts to fail.
• No appetite for the record-keeping. Twenty-five years of draws, purchases and distributions, each tied to the other. If that will not happen, the deduction is being claimed against a trail that will not exist.
• A plan to sell the home soon. A sale forces the credit line to be repaid, usually by selling the portfolio, and the whole arrangement unwinds with tax consequences at the worst possible moment.

Three of those five are about the leverage rather than the conversion, and a reader with rental or business income can have the conversion without them: the version with no portfolio behind it re-routes expenses that already exist, so a drawdown has nothing to fall.

None of that makes the strategy wrong. It makes it a strategy with conditions, and the mechanism itself is easier to judge once the conditions are on the table.