The three Canadian banks that reported Q2 earnings Wednesday collectively beat adjusted profit estimates, raised dividends, and described credit conditions as stable. Scotiabank earned an adjusted $1.90 per share against a $1.85 consensus. BMO topped estimates on capital markets strength. National Bank reported $3.23 adjusted EPS, up 13% year over year, with return on equity at 16.8% and provisions for credit losses more than halved from the same quarter last year. By any fundamental measure, Canada's banking sector is reporting one of its stronger earnings seasons of the past three years.
The TSX composite fell 241 points on Wednesday.
The divergence between institutional results and index performance is not a contradiction. It is a predictable outcome of the current environment: energy sector weakness on ceasefire optimism more than offset financial sector gains. But for clients watching the composite fall while their bank just reported record profits, the divergence produces a specific kind of confusion that, if not addressed, tends to resolve itself through anxiety rather than understanding.
What Mental Accounting Does to Earnings News
The concept of mental accounting, introduced by Richard Thaler in his 1980 paper "Toward a Positive Theory of Consumer Choice," describes the tendency of individuals to categorize financial events into separate psychological "accounts" and evaluate each account independently rather than as part of a unified ledger. In investment contexts, the relevant error is the tendency to treat a single holding's issuer news as a proxy for overall portfolio health.
A client who holds Scotiabank in their non-registered account and sees the bank report better-than-expected Q2 earnings will, in the absence of an advisor's framing, conduct a rapid mental calculation: good results should mean the price goes up; the price is flat or slightly down; therefore something is wrong that the bank is not telling me. The actual explanation -- that the TSX is an energy-heavy index and falling oil prices on ceasefire optimism dominated sector rotation on Wednesday -- is not intuitively accessible to a client who is not watching the mechanism in real time.
The chart above shows the divergence between TSX financials performance and the composite index since the Hormuz scenario began in early March, illustrating the structural decoupling of bank fundamental performance from index-level moves.
TSX Financials and the TSX Composite indexed to 100 at March 4, 2026. Financials outperformed the composite by approximately 4 points over the period, as oil price volatility drove energy-sector moves that dominated composite-level performance while bank fundamentals remained stable or improving.
The Specific Trap: Conflating Issuer Health With Price Direction
Thaler's mental accounting framework identifies several categories of error. The one most relevant to the current bank earnings cycle is the tendency to assign the outcome of one "account" as evidence about the state of another. A client who holds Scotiabank in their RRSP and RBC in their TFSA is, in their own mental ledger, holding two separate bets. When both issuers report strong results and the composite still declines, the client's mental ledger has produced three accounts with contradictory outputs: Scotiabank (positive), RBC (positive), portfolio (negative). The resolution the brain reaches without guidance is not "there must be a sectoral explanation I don't understand." It is: "the strong results must not be as strong as they appear."
This is the availability heuristic operating alongside mental accounting. Daniel Kahneman and Amos Tversky's 1974 work on heuristics and biases established that people evaluate probability and significance based on how easily examples come to mind. Negative portfolio moves are highly available -- they produce visceral reactions, they appear on mobile screens, they come up in conversations. Strong bank earnings releases are much less available -- they require active reading, they involve adjusted versus reported metrics, they are easy to discount. When the two signals compete, the negative price move wins the availability contest almost every time.
What the Advisor Provides That the Earnings Release Does Not
The earnings release provides numbers. The advisor provides the mechanism that connects the numbers to the portfolio context. In the current environment, that mechanism is: oil price movement driven by ceasefire negotiations is dominating TSX index-level performance on a daily basis, and that movement has no direct relationship to the credit quality, earnings power, or dividend sustainability of Canada's major banks. Those two things are operating on different frequencies at the moment.
Werner De Bondt and Richard Thaler's 1985 research on investor overreaction documented that markets tend to overweight recent negative information and underweight structural positive data. The client who dismisses strong bank earnings because their portfolio screen is red is exhibiting exactly this pattern. The advisor who names the pattern, explains the mechanism, and connects it to the client's actual position is providing something the client cannot access independently -- not because the client is unsophisticated, but because the mental accounting and availability heuristics that produce the error are universal cognitive features, not individual weaknesses.