The honest risks
This is the guide that costs us something to publish. If you’re carrying a balance, optimising which card you tap is close to pointless, and we’d rather say so here than let you find out later.
What borrowing on a card costs
Credit card interest sits far above almost every other kind of consumer borrowing in Canada. Here’s the average across the chartered banks.
The average interest rate charged on outstanding credit card balances at Canadian chartered banks.
Scope. A market-wide average across chartered banks for loans booked in Canada in Canadian dollars, published every month: the observation is July 2026, and the earlier months of that year ran between 20.97 and 21.50. The month is printed with the number for that reason. It is not the rate on any particular card, and the Bank of Canada does not document whether the series covers only interest-bearing balances. Federally regulated credit unions and non-bank lenders are excluded.
That’s an average across the whole market, not your card’s rate. Yours is in your cardholder agreement. The point is the order of magnitude, because it’s what any reward has to be measured against.
Carrying a balance is common
If you’re carrying one, you’re in a very large group. This isn’t a fringe behaviour and it isn’t a character flaw.
About forty-six percent of Canadian credit card holders carried a balance for at least two consecutive months on at least one card.
Scope. An average since January 2016, ranging between forty-two and forty-nine percent over that period. Based on TransUnion credit bureau microdata covering more than thirty million Canadian residents to February 2024. The definition is carrying an outstanding balance for at least two consecutive months on at least one card, which includes people who clear it soon after.
The typical amount carried is smaller than people expect, which cuts both ways. It’s less frightening than the headlines suggest, and it’s also enough to cost real money at card rates.
Half of the Canadian cardholders who carry a balance carry less than this amount, and half carry more.
Scope. A median for 2023, so half of balance carriers were above it and half below. Canadian dollars. Drawn from TransUnion credit bureau microdata. A median says nothing about the spread, and the note reports that a substantial minority of carriers were using most of their available limit.
The finding that matters most
A Bank of Canada staff note, which speaks for its author rather than officially for the Bank, worked from credit bureau records covering the same borrowers over time. It compared people who carried a balance two or more months in a row with people who paid in full. The association with later trouble is strong.
Canadians without a mortgage who carried a credit card balance two or more months in a row were about five times as likely as those who paid in full to fall behind on a debt payment within six months.
Scope. A staff analytical note, so the views are its author’s and may differ from the Bank of Canada’s. Carrying means a balance on at least one card for two or more consecutive months, compared with people who pay in full each month, with other characteristics held equal. Falling behind means at least sixty days late on any credit product within the following six months. For scale, in any given month fewer than two in a hundred people who had paid in full over the prior six months fell behind. This is an association in credit bureau data, not evidence that carrying a balance causes the later trouble, since a common shock such as lost income can produce both. Separate and larger multiples in the same note apply to people using most of their limit or carrying for many months, and should not be quoted as if they were this figure.
Read the scope note on that one carefully. It’s an association in credit bureau data, not proof that the balance caused the trouble. Someone whose hours got cut may start carrying a balance and then miss a payment, and the lost hours caused both. Still, if you’re carrying a balance month after month, that pattern is a signal worth taking seriously about your own situation.
Does a rewards program change what you spend?
Here the evidence is thinner than the confident version you’ve probably heard, and you deserve what there actually is rather than the half that suits us.
A study of one bank’s own cardholder records found that people who joined a cashback program spent more and borrowed more afterward, by amounts far larger than the rewards themselves.
Cardholders who enrolled in a cash back program went on to spend and borrow substantially more than the reward was worth.
Scope. United States, US dollars, administrative account records from one large financial institution in the 2000s. Measured over the first quarter after enrolment, comparing the same cardholders before and after, so it does not isolate the reward from anything else changing at the same time. Effects differed sharply by prior usage and by whether the cardholder was credit constrained. A working paper, not peer reviewed.
Read what that measures. It’s joining a rewards program, not the act of paying by card, and it compares the same people before and after rather than against a control group. We found no Canadian equivalent and no direct replication of it, so it’s a strong hint rather than a settled result.
Does paying by card change what you spend?
That’s a different question, and on this one the evidence really does point both ways. Older auction work suggests people will bid more for the same item when they expect to pay by card. It’s frequently quoted and the size of the effect is frequently exaggerated, so here it is with its limits attached, next to a field experiment that pushed people to pay by card instead of cash and found no overall increase, though different groups moved in opposite directions.
In sealed-bid auctions for real items, people bid more when they were told they would pay by credit card than when they were told they would pay cash.
Scope. United States, 1990s, small samples of business school and executive education participants bidding on scarce Boston sports tickets and on a low-value item. The bids were binding rather than hypothetical, which is a strength, but the population and the items are narrow, which is not. The published abstract describes the effect as possibly large, up to one hundred percent, and that is an upper bound from one study rather than a typical result. The full text is paywalled, so the item-by-item breakdown could not be independently verified here. Replications of card-versus-cash spending effects have been mixed, and the field experiment cited alongside this entry found no overall effect.
A field experiment that induced people to pay by credit card instead of cash found no overall increase in spending.
Scope. United States, lunch purchases by insurance company employees in a workplace cafeteria, so the amounts are small and the setting is routine rather than discretionary. The effect ran in opposite directions by group: people who carried a balance spent less on a card, people who paid in full spent more. A working paper, not peer reviewed. It does not settle the question for large or unplanned purchases.
Where that leaves us: the claim that cards reliably make everyone spend more is not settled science. The auction study has the cleaner contrast, the field experiment has the more realistic setting, and they disagree. What survives all of it is that a rewards program can change behaviour by more than the reward itself is worth, which is worth being suspicious of, including with the programs this app helps you use.
The wider picture in Canada
Canadian households carry a lot of debt relative to income, and that’s the environment any of this happens in.
Canadian households owed about one dollar and seventy-six cents in credit market debt for every dollar of disposable income.
Scope. A seasonally adjusted national aggregate covering all household credit market debt, which is mostly mortgages rather than credit cards. It describes the household sector as a whole and is not a description of any individual household. Published quarterly, so this figure is superseded on a regular schedule.
That figure includes mortgages and it’s a national aggregate, so it isn’t a description of your household. It’s context for why a small card balance can be harder to shift than it looks.
The part that argues against us
If you carry a balance from month to month, this app’s advice is worth very little to you.
The maths isn’t subtle. Rewards on Canadian cards are counted in cents on the dollar. Card interest is counted in tens of percent per year. Optimising which card you tap moves the small number while the large number keeps running. You could use Verdict perfectly, every purchase, for a year, and be worse off than someone who used one plain card and cleared it every month.
So if you’re carrying a balance, the highest-value thing you can do has nothing to do with choosing a card. It’s clearing the balance, or getting help to clear it. That is a better use of your attention than anything on this site, and we’d rather you spend it there.
If that’s where you are, what to do when it goes wrong has the concrete steps and the Canadian places to call.