Three of Canada's six largest banks reported this week on a four-quarter beat streak, and Thursday's results kept it alive. Royal Bank of Canada posted record net income of $6.0 billion for the quarter ended July 31, up 11% from a year earlier, with diluted earnings per share of $4.23, comfortably ahead of the roughly $4.08 analysts had modelled. CIBC's adjusted net income rose 26% to $2.65 billion, with adjusted earnings per share of $2.73, extending a streak of consecutive quarterly beats that analysts had already priced as the base case.
Both prints look like clean wins. The market's memory of how the same story played out two days earlier is the more useful guide to how today's numbers will actually trade.
Same Beat, Opposite Verdict
On Tuesday, Scotiabank and BMO reported ahead of the open. Scotiabank's adjusted profit rose 18% year over year on a record quarter, and its stock surged 7% to close at $128.73, the bank's largest single-day gain in more than six years. National Bank's profit rose by a similar magnitude, roughly 23%, and beat analyst estimates on the headline number. Its stock fell 4.2% the same session.
Two banks, comparable profit growth, opposite market verdicts. The gap is not a data error. It is the signature of what Amos Tversky and Daniel Kahneman named the representativeness heuristic in 1974: people judge how likely or how good an outcome is by how closely it resembles a mental template, not by weighing the full set of evidence behind it. A quarter matching the template of "clean beat, nothing to explain" gets rewarded in full. A quarter that beats on the headline but contains a detail that breaks the template, in National Bank's case a provisioning or outlook note investors read as inconsistent with a record quarter, gets marked down even though the primary number was just as strong.
Scotiabank and National Bank posted comparable profit growth this week and received opposite verdicts from the market, a split visible in the paired comparison below.
National Bank's decline came despite provisions and capital ratios in line with peers; the reaction traces to guidance language flagged by analysts on the post-earnings call. Source: Company filings, Aug. 25-26, 2026.
What the Template Actually Contains
For an advisor's client holding Canadian bank stocks, the intuitive read of this week is simple: banks are beating, so bank exposure is working. That read is representativeness doing its job quietly. It substitutes a category judgment, "banks are having a good quarter," for the harder task of pricing each institution's specific mix of revenue quality, provisions, and forward guidance. The category judgment is usually a reasonable shortcut. This week it produced a 27-percentage-point gap between two banks with similar underlying profit growth.
The same mechanism cuts the other way for RBC and CIBC's results this morning. Both cleared the visible beat threshold that markets currently associate with "good quarter." Whether either stock trades in line with Scotiabank's reception or National Bank's will depend on whether something inside today's releases breaks the template, a capital markets detail, a credit provision line, a remediation cost, the kind of specific fact that a headline profit number does not carry but that determines how the quarter is actually read.
Why the Distinction Matters for Portfolio Conversations
Tversky and Kahneman's original point was that representativeness is a useful heuristic most of the time and a systematic source of error the rest of the time, and the two cases are not obviously distinguishable in the moment. National Bank's investors were not irrational to expect the market to reward a 23% profit gain. They were applying a pattern that has worked in the majority of recent quarters. This week it did not hold, and the four-quarter beat streak across the sector means expectations are now high enough that beating consensus and having the stock react favourably are two separate events rather than one.