Scotiabank Clears Its ROE Target a Year Early, and Global Banking and Markets Paid For It

Results vs. Consensus

Scotiabank reported fiscal third-quarter results before the open on Tuesday, August 25, 2026, for the three months ended July 31. All figures are in Canadian dollars under IFRS. Where management quotes a growth rate on a basis that differs from the filed income statement, both are given.

Three independent consensus providers published the same C$2.10 adjusted EPS estimate, so the 8.6% beat is not an artifact of one provider's panel. The revenue consensus of C$9.99B was cleared by 5.5% on the reported income statement.

Year-over-year comparison

On an adjusted basis, which strips only C$24M of pre-tax acquisition-intangible amortization this quarter, revenue was C$10,543M against C$9,494M (+11.0%), expenses C$5,540M against C$5,095M (+8.7%), pre-tax pre-provision profit C$5,003M against C$4,399M (+13.7%), and adjusted diluted EPS C$2.28 against C$1.88 (+21.3%). Adjusted return on equity was 14.2% against 12.4%, and adjusted return on tangible common equity 17.2% against 15.1%. In-quarter adjusted operating leverage was 2.3 points.

Sequential comparison

The sequential comparison carries a calendar distortion worth naming. The fiscal third quarter ran 92 days against 89 in the second, a 3.4% difference that management flagged for Canadian Banking and Global Wealth Management but not at the all-bank level. Net interest income grew 6.2% sequentially; on a per-day basis it grew 2.7%. The same arithmetic applies to the revenue line. Sequential adjusted operating leverage was effectively zero, and the adjusted productivity ratio was flat at 52.5%.

Nine-month view

The reported nine-month comparison is close to useless and should be discarded: it flatters this year by the absence of the C$1,365M held-for-sale impairment booked in the first quarter of fiscal 2025, and penalizes it by the C$423M loss recognized on completing the Colombia, Costa Rica and Panama sale in the first quarter of fiscal 2026. On the adjusted line the nine-month picture is clean: revenue up 8.9%, expenses up 4.7%, and 4.3 points of operating leverage.

Revenue

Reported revenue of C$10,535M grew 11.1% year over year, and the composition is the interesting part: net interest income contributed 6.8% growth while non-interest income grew 16.9%. On the constant-currency, ex-divestiture basis management used for its all-bank commentary, those become 12% and 21% respectively against 16% total. The gap between the filed 11% and the spoken 16% is the divested Colombia, Costa Rica and Panama businesses, which closed December 1, 2025 and are therefore in the prior-year base but not this one.

Net interest margin of 2.49% was 13 basis points wider year over year and unchanged sequentially, which is a cleaner outcome than it looks. The sequential stall is the product of Canadian Banking and Global Banking and Markets both widening against an International Banking step-down that management had told the market to expect. Non-interest income growth was broad: higher banking and wealth revenues, underwriting and advisory fees, and higher income from associated corporations, the last of which includes the Davivienda stake received as consideration for the divested businesses.

Margins and expenses

Adjusted expenses grew 8.7% year over year against 11.0% adjusted revenue growth, delivering 2.3 points of in-quarter operating leverage and the tenth consecutive quarter of positive all-bank operating leverage. On the ex-divestiture basis the same lines read 14% and 16%, which is a far tighter 2-point spread and the honest way to look at the underlying cost trajectory. Technology spend grew 16% to C$1.5B in the quarter, and the balance of the increase is performance and share-based compensation moving with capital-markets revenue.

That last point deserves emphasis because it cuts both ways in a mean-reversion scenario. We assumed 6% to 7% ex-divestiture expense growth for the second half; the actual was 14%. The miss is entirely explained by the same variable that drove the revenue beat, so it is not a cost-control failure. It does mean that a normalization in capital-markets revenue carries a partial expense offset, which limits how much of the C$190M sequential Global Banking and Markets step-up would actually reverse out of earnings.

Earnings per share

Adjusted diluted EPS of C$2.28 grew 21.3% year over year against 18.1% adjusted net income growth. Roughly 220 basis points of that 320 basis point wedge is the buyback: weighted average diluted shares fell to 1,226M from 1,249M on the adjusted basis, a 1.8% reduction, with the balance coming from the preferred and non-controlling-interest line, with 8.6 million shares repurchased in the quarter and 19.9 million year to date. On a trailing twelve-month basis adjusted diluted EPS is C$8.28, against C$7.09 for full-year fiscal 2025.

The effective tax rate rose 100 basis points sequentially to 24.3%, which is the normalization we assumed in May when we flagged that the April quarter's 23.3% had been flattered by jurisdiction mix and one-time items. The bank absorbed that headwind and still beat by 8.6%. That is a genuine positive and should be scored as one.

Segment Performance

A note on definitions that matters when reading Scotiabank against its own press release. The headline paragraph of the earnings release quotes total segment net income, which for International Banking is C$766M and for Global Wealth Management C$518M. The segment review, the return-on-equity tables and the investor deck all use net income attributable to equity holders, which is C$725M and C$515M. Segment return on equity is struck on the latter. This report uses net income attributable to equity holders throughout.

The Other line above is derived as the residual between the four business-line revenues and the income-statement total; the segment revenues are as disclosed. International Banking revenue is shown on the reported basis, which carries the divested operations in the prior-year comparative. On a constant-dollar, ex-divestiture basis it grew 7%.

Canadian Banking

Net income of C$1,071M grew 11.8% year over year and 14.5% sequentially, on revenue of C$3,646M and pre-tax pre-provision profit of C$1,972M. Return on equity reached 19.4%, up 160 basis points sequentially from 17.8% and 100 basis points year over year from 18.4%. The productivity ratio improved 140 basis points to 45.9%, the best of any segment, and this was the fourth consecutive quarter of positive operating leverage. Provisions fell 13% sequentially to C$498M, a 42 basis point ratio.

Net interest margin expanded for a fifth consecutive quarter, up 2 basis points to 2.38% from 2.29% a year ago, driven by both loan and deposit margin. Risk-adjusted margin, which nets provisions out of the margin and is the better read on whether the mix shift is actually earning anything, jumped 11 basis points sequentially to 1.96% from 1.85%, and is now above its year-ago 1.89% for the first time in this sequence. That is the single most encouraging number in the Canadian segment.

The composition of loan growth finally moved. Average loans grew 3% year over year, with mortgages up 4%, commercial and small business up 3%, and personal loans up 1%. Sequentially, commercial loans grew 3% after 2% in the April quarter, small business loans grew 3% and 10% year over year, and credit card balances grew 3%.

"On the business mix, we've talked about it many times, non-mortgage lending now is accelerating and actually pass mortgage growth in the quarter for the first time in 2 years, and we see that in the card book, the business banking book, ULOC and commercial, and that should continue."
— Aris Bogdaneris, Group Head, Canadian Banking

On deposits the pattern from the April quarter persisted. Personal day-to-day and savings balances grew 1% year over year while total segment deposits fell 2%, all of it in term. Management retained more than 90% of retail GIC maturities year to date, with the outflow moving either into day-to-day accounts inside the segment or into retail mutual funds, where net sales of C$4B year to date are running near 2.5 times last year.

"Day-to-day and savings deposits grew 1% year-over-year, in line with our strategy. However, deposits declined 2% year-over-year, mostly in term."
— Raj Viswanathan, Group Head and Chief Financial Officer

Assessment: This is the first quarter in which the Canadian turnaround stopped being a margin story and became a franchise story. Margin expansion continued, but the incremental evidence is elsewhere: risk-adjusted margin up 11 basis points, non-mortgage growth overtaking mortgage growth for the first time in two years, premium cards at 45% of new acquisitions against 34%, digital at 40% of retail sales, and a direct cost base that has not grown in twelve months while 500 salespeople were added. At 19.4% return on equity on C$21.9B of attributed equity, this segment now earns roughly 36% of the bank's business-line profit at the highest return in the house. The bull pillar is intact and strengthening.

International Banking

Net income of C$725M grew 8% year over year on the reported basis, fell 1% on a constant-dollar basis, and grew 6% on a constant-dollar, ex-divestiture basis. Revenue of C$2,948M was down 2% reported, down 9% constant dollar and up 7% constant-dollar ex-divestiture. This is the segment where the four-bases problem is most acute, and every figure below is labelled.

Net interest margin of 4.69% came in exactly inside the 4.65% to 4.70% range management guided in May, down 7 basis points sequentially from the 4.76% high-water mark and up 15 basis points year over year. That is a guide given and a guide met, and it removes the concern we raised last quarter that the margin was flattered by seasonality management had disclosed but the market had not fully discounted. Risk-adjusted margin improved 15 basis points sequentially to 3.55%, the best in the sequence.

The loan book still tells two stories. On a constant-dollar ex-divestiture basis average loans fell 1%, with retail up 5% and non-retail down 7%. Deposits rose 6% on the same basis, personal up 4% and non-personal up 7%. Provisions fell 28 basis points sequentially to 138 basis points, the largest single improvement in the bank.

"This is an important quarter in the sense that it marks the effectiveness of the pivot to growth effort we've been leading for the last 4 quarters. We're now seeing the business growing at 6% year-on-year. And when you look at the underlying business lines are growing substantially higher revenues than what we saw in 2025. And that positions us to the target in 2027 and beyond of growth within the 6% to 8% level on the revenue front."
— Francisco Aristeguieta, Group Head, International Banking

What the segment disclosure does not make obvious is where the earnings came from. The chief financial officer disclosed, in a single sentence, that the Global Banking and Markets business booked inside International Banking earned C$321M in the quarter. That is 44% of the segment's C$725M.

Assessment: The margin guide was delivered, revenue grew 7% ex-divestiture, and management gave the first quantified multi-year targets for the segment since Investor Day. Those are all improvements on the April quarter. But the composition has moved in a direction that complicates the story: a segment that carries a 15.7% return on equity does so with 44% of its earnings coming from a capital-markets business whose returns management declined to disclose separately. The retail and commercial franchise that the "pivot to growth" is supposed to be fixing is still not the thing generating the growth.

Global Wealth Management

Net income of C$515M grew 23.5% year over year and 8.6% sequentially, with Canadian earnings up 27% and international up 4%. Revenue of C$1,897M grew 18% against 16% expense growth, delivering 2 points of in-quarter operating leverage and a productivity ratio 130 basis points better at 62.9%. Return on equity reached 18.6%, up 290 basis points year over year, on C$11.0B of attributed equity, which makes this the most capital-efficient earnings stream in the bank.

Spot assets under management grew 16% to C$474B and assets under administration 13% to C$856B. Net sales were C$3.0B, a record third quarter and up 14% year over year, marking the eighth consecutive quarter of positive net flows, with year-to-date net sales already above the full fiscal 2025 total. The franchise now ranks third among bank-owned peers in long-term retail mutual fund sales, up from fifth a year ago and sixth at Investor Day.

"I think we've dramatically improved our penetration of the retail client base since Investor Day. We're currently sitting at around 11.6%. We think 15% penetration is absolutely achievable in this business."
— Jacqui Allard, Group Head, Global Wealth Management

One number moved the other way and was not framed as such. Net sales of C$3.0B compare to C$4.7B in the April quarter, a 36% sequential decline. Management presented the figure only as a record third quarter and a year-over-year increase, both of which are accurate, and did not address the sequential path.

Assessment: Wealth remains the cleanest part of the story and the pillar needs no defence. The 290 basis point return-on-equity improvement is the largest in the bank on the smallest capital base, the flow franchise is demonstrably taking share, and the referral engine with Canadian Banking produced C$14B of closed referrals year to date. The 11.6% penetration figure is the number to track: it converts an abstract cross-sell narrative into a measurable gap against an 15% stated target.

Global Banking and Markets

Net income of C$647M was a record, up 36.8% year over year and 41.6% sequentially. Revenue of C$2,014M grew 32% year over year and 26% sequentially, with capital markets revenue up 33%, business banking up 30%, and underwriting and advisory revenue a record at up 88%. Net interest income grew 34% and non-interest income 31%. Expenses grew 26%, producing 6 points of in-quarter operating leverage and a productivity ratio 270 basis points better at 55.8%. Return on equity of 15.9% was up 330 basis points year over year, though on C$16.1B of attributed equity, up from C$15.2B, the third-largest capital allocation in the bank behind Canadian Banking and International Banking.

Net interest margin widened 32 basis points year over year to 2.09%, which management attributed mostly to deposits. Average business and government loans grew 5% year over year and 7% sequentially, with Canadian loans up 9% year over year, and average deposits grew 12% year over year and 9% sequentially. The chief executive itemized the quarter's marquee mandates: joint lead and bookrunner on the two largest debt capital markets deals ever done in Canada, the largest asset-backed securities deal since the structured credit platform was established, bookrunner on the largest Canadian initial public offering since 2021, and a first lead-left leveraged finance mandate.

"I think loan growth is really just a reflection of the economic output and the focus that we have on our clients, it's not a KPI that we're trying to drive. I mean we're not out there just trying to grow loans to grow loans."
— Travis Machen, Group Head, Global Banking and Markets

Provisions rose to C$53M, an 18 basis point ratio and up 11 basis points year over year, split between higher performing provisions on forward-looking indicators and impaired provisions on new Canadian formations. In absolute terms this is immaterial; as a direction of travel in the only segment adding loan balances quickly, it is worth a note.

Assessment: There are two defensible readings. The constructive one is that three years of deliberate portfolio pruning and product build-out produced a broad-based record, not a single trading windfall, and that a Canadian bank with genuine cross-border capital-markets capability at a moment of heavy Canadian capital formation is a structurally better business than it was. The skeptical one is that Bank of Montreal reported the same morning with capital-markets net income up 45%, which means the condition was sector-wide rather than franchise-specific, and that a segment now producing 22% of business-line earnings against 19% a year ago is a mix shift toward the least persistent earnings in a bank. Both readings survive this quarter's evidence. The tiebreaker is what the October quarter prints.

Other

The Other segment recorded a reported net loss of C$50M and an adjusted net loss of C$42M, against C$28M of reported income and C$35M adjusted in the April quarter. The C$77M adjusted swing is the reversal of last quarter's elevated investment gains, which is precisely what the chief financial officer told the market would happen when he described the April quarter's corporate result as returning to "a modest loss." That commitment was made and kept, and it is worth roughly C$0.06 of the sequential EPS bridge working against the bank.

Key Performance Indicators

Credit Quality

This is the quarter the credit trend inflected, and it deserves to be graded against what management told the market in May rather than against a general impression. In the April quarter the impaired provision guide was revised up to the mid-50 basis point range for the balance of fiscal 2026, above the outlook given in December. This quarter the impaired ratio printed 52 basis points, inside that guide, with no further revision.

Impaired provisions had risen for four consecutive quarters, from C$975M to C$1,129M. This quarter they fell to C$1,018M. Gross impaired loan formations fell to C$2,210M from C$2,662M, the lowest of the five-quarter sequence. Both moves are the shape of a turn rather than a single soft print.

"Our credit performance improved this quarter, with PCLs beginning to decline in line with our outlook for the second half of the year."
— Shannon McGinnis, Chief Risk Officer

The stock of impairment has not turned. Gross impaired loans rose to C$7,801M from C$7,608M and the gross impaired ratio ticked up a basis point to 100 basis points, its highest in the sequence, though management attributed the International Banking portion largely to foreign currency translation. Total allowance for credit losses rose to C$7,551M, a 97 basis point ratio, with performing allowances up C$104M and impaired allowances up C$103M sequentially. Excluding currency, those builds were C$39M and C$46M. Net impaired loans as a percentage of loans and acceptances were 0.68%, unchanged.

Canadian retail: everything improved except the mortgage book

Canadian Banking retail provisions were C$369M, down C$66M sequentially, with performing down C$10M to C$24M and impaired down C$56M to C$345M. The improvement came from lower net write-offs in unsecured lines of credit and lower auto impairments, both attributed to collections initiatives that came online during the year. Every unsecured product improved sequentially. Mortgages did not, and the total retail delinquency rate rose a basis point in spite of the unsecured improvement, because mortgages are the larger balance.

"While we are encouraged by the improving trends in impaired provisions, and 90-day delinquency across most retail products, we continue to monitor some pockets of weakness, including elevated mortgage delinquencies. That being said, mortgage clients remain resilient and our overall retail portfolio quality remains strong with an average FICO score of 798."
— Shannon McGinnis, Chief Risk Officer

The chief risk officer named the mortgage exception in prepared remarks, which is the right disclosure posture. The Greater Toronto uninsured number was not named on the call and sits in the appendix of the investor deck. It has now risen 17 basis points in four quarters against an 11 basis point rise for the uninsured book nationally, on a sub-portfolio with an average FICO of 807 and an uninsured loan-to-value near 60%. This is a payment-shock problem in a specific vintage and a specific geography, not a credit-quality problem in the underwriting, but it is compounding rather than flattening.

International retail and the Brazilian corporate account

International retail 90-plus-day delinquency was 3.24%, down a basis point sequentially, with personal loans down 18 basis points and cards down a basis point offsetting mortgages up 4 basis points to 3.22%. The deck attributes the mortgage drift to Chile, where the portfolio is indexed to persistently high inflation, and notes marginal improvement in Mexican mortgages and continued improvement in Peru.

On the single-name corporate exposure in Brazil that cost roughly 7 basis points of all-bank impaired provisions in the April quarter, the bank took an incremental C$57M provision this quarter and reclassified a further C$14M from credit valuation adjustment to provisions on a derivative exposure, with no change in the underlying exposure. Management said only that it continues to work through the account and did not disclose the remaining exposure. International commercial provisions still fell sequentially despite this.

Assessment. We wrote in May that a second consecutive quarter of single-name corporate impairments in International Banking would reframe the credit story from idiosyncratic to systemic. A second quarter of provisions did arrive, but on the same account rather than a new one, and against a segment impaired ratio that fell 27 basis points. That is the continuation of a known problem, not the emergence of a pattern, and we do not treat the downgrade trigger as having fired. If the October quarter produces a different name, that judgment changes.

Capital and Shareholder Returns

Risk-weighted assets rose to C$492.9B from C$474.4B, of which C$7.1B was currency translation and C$7.4B was book size, with a further C$2.4B from recalling a synthetic risk transfer that had been put in place roughly three years ago under capital-floor constraints and is no longer economic. Tier 1 capital fell 30 basis points to 15.1% and total capital 10 basis points to 16.9%; the leverage ratio held at 4.3%, and TLAC ratios of 28.6% and 8.2% remain well above minimums.

The quarterly dividend was held at C$1.14, unchanged from the 4% increase declared with the April-quarter print. The bank repurchased 8.6 million shares in the quarter and 19.9 million year to date, returned C$6.3B of capital year to date, and by the chief executive's count C$8.3B over the trailing twelve months, up from roughly C$7.5B on the same measure a quarter ago.

Key Topics & Management Commentary

Overall Management Tone: The most confident call of this coverage window, and unusually, the confidence was earned rather than asserted: management led with a target it had beaten a year early and then spent most of the call being asked why it should not raise the target further. Segment heads were unusually specific, offering quantified multi-year objectives for International Banking revenue growth, a 20%-plus return-on-equity ambition for Canadian business banking and a 15% wealth penetration goal, none of which were volunteered a quarter ago. The one place management would not engage was profitability disclosure below the segment level, where a direct question about the return on equity of the capital-markets business embedded in International Banking was declined outright.

1. The 14% return-on-equity target was cleared, and about a point of it came from one segment

Adjusted return on equity of 14.2% cleared the bank's 14%-plus medium-term objective in the July quarter of fiscal 2026, against a schedule that as recently as May pointed to fiscal 2027. Reported return on equity was 14.1% and return on tangible common equity 17.2%, both up more than 200 basis points year over year on the adjusted basis.

"Q3 was a record quarter for the bank as we reported strong earnings across all business lines and exceeded all of our medium-term objectives. We are particularly proud of the fact that we demonstrated our ability to hit our 14% plus return on equity target sooner than we had projected. This achievement was aided by strong markets, but is also the product of strategic repositioning and improved capital allocation that have led to sustainable improvements across the bank."
— Scott Thomson, President and Chief Executive Officer

The arithmetic behind the qualifier is straightforward. Global Banking and Markets earned C$647M against C$457M in the April quarter. Hold it flat and adjusted net income to common shareholders falls from C$2,798M to C$2,608M, adjusted return on equity falls to 13.2% on the same C$78.2B of average common equity, and the target is not cleared. Roughly 100 basis points of the 14.2% is the sequential step-up in the least predictable segment in the bank.

Assessment: The milestone matters, and management was straight about the assist. But a target cleared with the help of a record quarter in a business whose own head declined to forecast next quarter's revenue is a target reached, not a target held. What would make it structural is the October quarter printing 14% with a normalized capital-markets contribution. That is a testable proposition three months away, and it is the single most important number in the next print.

2. Canadian Banking's mix shift finally showed up in the loan book

For four quarters the Canadian story has been margin and cost. This quarter it became volume and mix. Commercial loans grew 3% sequentially after 2%, small business 3% sequentially and 10% year over year, credit cards 3% sequentially, and non-mortgage lending grew faster than mortgage lending for the first time in two years. Risk-adjusted margin, the measure that nets provisions out of the spread, rose 11 basis points sequentially to 1.96% and is now above its year-ago level.

"In Canadian Banking, commercial loans grew 3% sequentially in Q3 after growing 2% in Q2. Looking ahead, we expect growth to continue to improve supported by investments we are making in verticals where we've been historically underpenetrated, including the mid-market and small business lending where loan growth was up 3% quarter-over-quarter and 10% year-over-year."
— Scott Thomson, President and Chief Executive Officer

The mid-market build is now quantified: nearly 700 mid-market clients added year to date, up almost 85% year over year, in a segment management describes as deposit-rich and higher-margin, against a historical over-weight to real estate. Premium cards were 45% of new card acquisitions against 34% a year ago.

Assessment: This is the commitment we asked to be graded, and it was partly kept. The commercial pipeline converted, small business is compounding at double digits, and the mix rotation is visible in the risk-adjusted margin rather than only in the narrative. What has not yet happened is total Canadian loan growth catching the market: average loans still grew 3% year over year, the same rate as the April quarter. The promise was that this closes by the end of fiscal 2026, which leaves one quarter.

3. A fifth consecutive quarter of Canadian margin expansion, and the source has changed

Canadian Banking net interest margin rose 2 basis points sequentially to 2.38%, the fifth consecutive increase, up 9 basis points year over year from 2.29%. Management attributed the increase to expansion in both loan and deposit margins, and specifically to mortgage renewals repricing upward.

"On the business mix, we've talked about it many times, non-mortgage lending now is accelerating and actually pass mortgage growth in the quarter for the first time in 2 years... The other big component of our capital heavier businesses in auto and mortgage is the improvement in RAM, and we're seeing that. Also, as we renew the mortgages, you're going to see the RAM lifting, and we saw that in the quarter as well, a big increase in RAM."
— Aris Bogdaneris, Group Head, Canadian Banking

Assessment: Two basis points a quarter is exactly the pace management guided in May, delivered for a fifth consecutive quarter, and it is now sourced from asset-side repricing rather than deposit-cost relief as management said it would be. The renewal tailwind has a defined runway as the 2021 and 2022 vintages roll, which makes the next several quarters of Canadian margin the most forecastable line in the bank. We model 2.40% by the October quarter.

4. Global Banking and Markets: a record quarter with two honest readings

Revenue grew 32% year over year to C$2,014M, with capital markets up 33%, business banking up 30% and underwriting and advisory revenue a record at up 88%. Net interest margin widened 32 basis points to 2.09%, mostly on deposits, while average deposits grew 12% year over year and 9% sequentially. Loans grew 7% sequentially after what management described as a period of deliberate optimization.

"We've been deemphasizing some businesses or regions, and we've been doubling down on building new products and services and focus in our core footprint, both Canada and U.S. and rest of the world. And so we're on the journey of trying to build a very durable franchise that's very broad-based."
— Travis Machen, Group Head, Global Banking and Markets

The bank's mandate list for the quarter supports the breadth claim: the two largest Canadian debt capital markets transactions on record, the largest domestic initial public offering since 2021, the largest asset-backed securities deal since the structured credit platform was built, and a first lead-left leveraged finance mandate. That is a franchise widening, not a trading book getting lucky.

Against that, Bank of Montreal reported the same morning with capital markets net income up between 45% and 46%. Whatever Scotiabank did well, the conditions were available to everyone.

Assessment: We set positive Global Banking and Markets operating leverage as one of three upgrade conditions in May, and 6 points of it arrived. But the mechanism matters: expenses still grew 26% year over year, and the leverage came from a record revenue line rather than from cost discipline. Year-to-date operating leverage in the segment is 0.9%. The condition was met on the letter and only barely on the spirit.

5. The capital-markets business hiding inside International Banking

The chief financial officer disclosed, in one sentence of prepared remarks, that the Global Banking and Markets business booked inside the International Banking segment earned C$321M this quarter. Combined with the standalone segment's C$647M, that puts capital-markets-linked earnings at C$968M, or roughly 33% of the C$2,958M of business-line net income attributable to equity holders. The segment head described the strategy candidly.

"We decided to build market-leading capital markets capabilities supporting the international footprint... So what you saw in this quarter and sequentially year-on-year, you're seeing growth on the revenue front of 40% without necessarily absorbing material capital."
— Francisco Aristeguieta, Group Head, International Banking

Assessment: Capital-light revenue growth is genuinely good, and building sovereign and domestic capital-markets coverage across a footprint the bank already banks is a sensible use of an existing distribution network. The problem is disclosure symmetry. International Banking's 15.7% return on equity is the number that supports the "pivot to growth" narrative, and 44% of the earnings behind it come from a business whose standalone returns the bank declines to publish. An investor cannot presently tell whether the retail and commercial franchise in Latin America is earning its cost of capital.

6. International Banking's pivot gets quantified targets for the first time

Management put numbers on the international outlook that did not exist a quarter ago: revenue growth of 6% to 8% in 2027 and beyond, expense growth held near 4%, pre-tax pre-provision profit growth of 8% or better, and double-digit earnings growth from 2027. Net interest margin of 4.69% landed inside the 4.65% to 4.70% guide given in May, retail loans grew 5% year over year ex-divestiture, and deposits grew 6%.

"We are very encouraged with the quality of the new vintages that we're onboarding and the effectiveness and penetration on our GTB business across corporate and commercial. So the combination of those 2 should allow us to see a more stable credit performance going forward that should allow us to deliver double-digit earnings in '27 and beyond."
— Francisco Aristeguieta, Group Head, International Banking

The non-retail book still shrank 7% year over year on a constant-dollar ex-divestiture basis, which management continues to describe as restraint by design in service of primary relationships rather than balance growth.

Assessment: The margin commitment from May was met precisely, which materially raises our confidence in the segment's guidance quality. The growth targets are new, dated and falsifiable, which is an improvement on the qualitative framing of the April call. The bear point is narrower than it was: this is no longer purely a shrinking business, it is a business where the growing parts are retail, deposits and capital markets while the non-retail corporate book continues to be run down.

7. Credit turned, and the guide held without revision

Impaired provisions fell to C$1,018M, or 52 basis points, after three consecutive quarterly increases, and formations fell C$452M sequentially. The mid-50 basis point guide set in May was met rather than revised again. Performing provisions fell to C$61M, or 4 basis points.

Q: "I think you expected or you talked about impaired PCL rate to be mid-50 basis points in the second half was 52 this quarter, and that included a decent drag from the Brazil loan this quarter again. So I think it's safe to say things seem to be progressing better than expected."
— Doug Young, Desjardins Capital Markets

A: "When I go back to our original outlook, our performance is in line with what we said at the time, which was that impaired PCLs were going to trend down in the latter half of the year, which is certainly what we are seeing."
— Shannon McGinnis, Chief Risk Officer

Assessment: Note what the chief risk officer did not do: she declined the invitation to call the result better than expected, and anchored back to the original outlook instead. That is the right posture from a risk function and it is a meaningful signal about the October quarter, in which no reduction to the guide was offered. We take the credit turn as real and the guidance as unchanged rather than improving.

8. Tariffs got a smaller answer than the question implied

New tariff measures were announced in the days around the print, and management sized the exposure narrowly on both the credit and the macro side.

"In terms of our exposure, we've been monitoring industries more vulnerable to tariffs since last year. And if I look at the latest measures that were announced, just to give you some context, that represents less than 1% of our total bank loans."
— Shannon McGinnis, Chief Risk Officer

The chief executive added a macro frame: the measures cover 5% of Canadian exports and imply a 0.2% to 0.3% impact on gross domestic product, against a backdrop he described as resilient on jobs and energy. He also noted that the bank had built 18 basis points of performing provision in the second quarter of fiscal 2025 against downside scenarios modelling Canadian tariff rates of 12.5% and up to 25% with full retaliation, while the current base case assumes tariffs are implemented and negotiations continue.

Assessment: The allowance was built for a worse world than the one that arrived, which is why performing provisions could fall to 4 basis points this quarter without a reserve release headline. This is the quietest part of the credit story and the part most likely to provide a cushion if the macro deteriorates.

9. Wealth compounds, but the flow number went the other way

Net sales of C$3.0B marked an eighth consecutive quarter of positive flows and a record third quarter, up 14% year over year, with year-to-date sales already exceeding all of fiscal 2025. Assets under management grew 16% to C$474B, the franchise moved to third from fifth in long-term retail mutual fund sales among bank-owned peers, and return on equity improved 290 basis points to 18.6%.

"Rankings are nice, but I think we take our confidence really from the underlying operating improvement. It's -- the drivers are right in line with the strategy that we laid out stronger execution on our partnership with Canadian Banking, multiyear investments that we're making in advisers, investment specialists, financial planners, technology as well as better coverage in our wholesale channel."
— Jacqui Allard, Group Head, Global Wealth Management

Assessment: Every framing management chose for the flow number was a true framing that avoided the sequential comparison, where C$3.0B follows C$4.7B. Wealth flows are seasonal and one quarter does not break a trend that is now eight quarters old, but a segment this central to the bull case should be reported against its own prior quarter as well as its own prior year. The 11.6% penetration disclosure, against a stated 15% target, is the more useful number and we will track it.

10. Capital stopped self-funding, and two known drags sit in the October quarter

CET1 fell 20 basis points to 13.1% with only 5 basis points of net internal capital generation, against 32 in the April quarter. Management then disclosed two identified fourth-quarter items.

"In Q4, certain international banking portfolios are migrating from the standardized approach to the AIRB approach that will reduce our capital ratios by approximately 15 basis points. We expect to absorb this impact and maintain our CET1 ratio of around 13% next quarter."
— Raj Viswanathan, Group Head and Chief Financial Officer

"Looking ahead, Chile announced a reduction in the tax rate by 4% over the next 3 years to 23%. Although this will result in lower taxes in future years, once enacted, it will also require a onetime deferred tax asset write-down in Q4."
— Raj Viswanathan, Group Head and Chief Financial Officer

Assessment: Neither item is a surprise or a scandal; both were volunteered. But together with a capital-markets comparison that will be hard to repeat, they mean the October quarter carries three identified headwinds against a share price that has just re-rated. Management sized the capital item and not the tax item, which is the disclosure asymmetry to watch when the fourth-quarter print lands.

11. Expenses, technology, and a direct cost base that has not grown in a year

All-bank adjusted expenses grew 8.7% year over year, or 14% on the ex-divestiture basis management used. Technology spend grew 16% to C$1.5B in the quarter. Inside Canadian Banking the picture is different and better.

"We haven't grown direct costs in 12 months. So year-on-year, the direct cost base for the Canadian bank has been flat. That said, we've added over 500 salespeople and continue to invest in digital, AI and technology enablement."
— Aris Bogdaneris, Group Head, Canadian Banking

Assessment: A flat direct cost base alongside 500 net sales additions and a 140 basis point productivity improvement is the most concrete evidence in the quarter that the Canadian efficiency programme is structural rather than a deferral. At the all-bank level the expense line is a different animal, because performance compensation moves with capital-markets revenue. The two should not be read as one trend.

12. Management's growth bases multiplied again

The standing caution from our initiation applies with equal force this quarter. Revenue growth for the same three months is 11% on the filed income statement, 11% adjusted, and 16% adjusted ex-divestitures. Expenses are 9%, 9% and 14%. The productivity ratio improved 100 basis points reported, 120 adjusted, and 90 ex-divestitures. Year-to-date operating leverage is 11.3% reported, 4.3% adjusted on our calculation, and 3.9% on management's adjusted ex-divestiture basis.

Return on equity attracted a similar treatment. The chief executive said Canadian Banking's return on equity "improved 160 basis points sequentially and hit 19.4% this quarter," which is correct, and later described the same segment as "up 160 basis points year-over-year," which is not: the year-over-year improvement was 100 basis points, from 18.4%.

Assessment: The disclosure itself is clean and the reconciliations are all published; the problem is spoken precision, and it recurs quarter after quarter. The practical rule for this name has not changed: never compare a Scotiabank growth rate quoted on a call to a peer's headline without first establishing which of four bases it sits on.

Guidance & Outlook

Scotiabank does not issue quarterly earnings guidance. What follows is the set of forward commitments management made on this call, alongside what was outstanding from the April call.

Two things stand out about that table. Six of the nine commitments carried into the quarter were met, one was partly met, and two remain open with one quarter to run. And every new commitment made on this call is a multi-year one: no one asked for and no one gave a fourth-quarter earnings frame, in a quarter where three identified headwinds are already visible.

Implied fourth-quarter path: nine-month adjusted diluted EPS is C$6.35. On our assumptions, a fourth quarter in the C$2.10 to C$2.20 range produces C$8.45 to C$8.55 for fiscal 2026, against C$7.09 in fiscal 2025. That requires a Global Banking and Markets contribution well below this quarter's record but above the C$457M April trough, a stable Canadian trajectory, and no deterioration in the impaired provision ratio.

Guidance style: Scotiabank under this management team has been conservative on credit and precise on margin. The International Banking margin guide was met inside a five basis point band, the Canadian margin cadence has been delivered five quarters running, and the corporate-segment reversal was flagged a quarter ahead and arrived on schedule. The one guide that has been revised in the wrong direction in this coverage window is impaired provisions, in May, and it was met on the first test after revision.

Analyst Q&A Highlights

The path from an 18% Canadian return on equity to the low 20s

The first question of the call went straight to the gap between the segment's year-to-date return on equity and the ambition set out at Investor Day, and asked what specifically has to happen to close it. The answer came from the segment head rather than the chief executive and was almost entirely about commercial banking rather than retail: a mid-market build-out, deposit gathering, and process reengineering, with a specific return target attached.

Q: "Just talk to us when we think about the ROE in that business today at 18% and that journey towards maybe low to even mid-20s, one, is that the right way to think about that business? And if so, what else needs to happen over the next year or 2 for us to get to that point from where we are today?"
— Ebrahim Poonawala, Bank of America

A: "Year-to-date, we've added nearly 700 mid-market clients, which is up almost 85% year-on-year. And obviously, this mid-market client is deposit-rich, higher-margin lending, and that pipeline is building in market as we also leverage the capabilities of GTB or transaction banking in that segment... And you see the ROE there almost at 25%. And when you take these 2 businesses together, we're confident as we continue and especially as our transaction banking capabilities improve that we can hit the 20% plus ROE over time."
— Aris Bogdaneris, Group Head, Canadian Banking

Assessment: The answer converted an aspiration into a mechanism, which is progress, but it also relocated the return-on-equity upside from the retail bank to commercial and small business. That is a narrower base than the Investor Day framing implied, and it makes the mid-market client-acquisition number the metric that carries the Canadian bull case from here.

Whether International Banking has genuinely inflected

A direct question about whether sequential loan and revenue improvement marks a pivot point, paired with a challenge on why expenses remained elevated, drew the most forward-looking answer of the call and the segment's first quantified multi-year targets.

Q: "Francisco, I was hoping that we could dive into the outlook for international. I mean, we're seeing, on a sequential basis, loan growth revenues pick up. Do we think that we're at an inflection point or a pivot in terms of the loan growth looking forward? And then secondarily, even though with the revenue growth, we are still seeing expenses remaining reasonably high. Any sense in terms of when that might move into positive operating leverage territory?"
— John Aiken, Jefferies

A: "This is an important quarter in the sense that it marks the effectiveness of the pivot to growth effort we've been leading for the last 4 quarters. We're now seeing the business growing at 6% year-on-year... We don't see a reason for expenses to move beyond where we've been, which is around the 4% level. That continues to be materially below inflation."
— Francisco Aristeguieta, Group Head, International Banking

Assessment: A 4% expense ceiling and 6% to 8% revenue growth imply a structurally positive operating-leverage machine, and the segment has now delivered against a margin guide, which raises the credibility of the rest. The gap in the answer is that neither the question nor the response distinguished between the retail franchise being fixed and the capital-markets business that generated 44% of the segment's earnings.

Why an internal-ratings migration costs capital rather than releasing it

The disclosure that certain international portfolios move to the advanced internal ratings-based approach in the fourth quarter at a cost of 15 basis points prompted the obvious question, since such migrations normally free capital rather than consume it.

Q: "You mentioned some conversion of portfolios from standardized to AIRB and that's going to reduce your core Tier 1 by 15 basis points. Just wondering why that is. Typically, it goes the other way. And is this a kind of a one and done? Or is there more of that type of transition taking place?"
— Gabriel Dechaine, National Bank Financial

A: "Yes, it is one and done. I think there were certain portfolios we should have converted a few years back... Some of it comes down to conservatism that is expected in Basel because the data quality in the countries are not as great, not our portfolio... Obviously, there will be some near-term impact to the ROE in the International Banking business because the denominator is going to increase next quarter, but it's pretty much done."
— Raj Viswanathan, Group Head and Chief Financial Officer

Assessment: A candid answer, and the admission that the conversion should have happened years earlier and is penalized for country-level data quality is more informative than the 15 basis points. It also means the International Banking return on equity, currently 15.7%, faces a mechanical denominator increase in the October quarter that has nothing to do with the operating performance the segment is being credited for.

What the international segment earns without its capital-markets business

The sharpest exchange of the call. Having observed that the capital-markets business inside International Banking accounts for more than 40% of the segment's earnings year to date, the questioner asked what the segment's return on equity would be without it. The segment head answered a different question about strategy, was pressed a second time on the specific number, and the chief executive closed the line of inquiry.

Q: "We're up over 40% of total segment earnings from that source so far this year. Like how intertwined is that business with your personal and commercial bank, if you will, across the region? And what does the ROE look like if that business is not there? I expect quite a bit lower."
— Gabriel Dechaine, National Bank Financial

A: "We don't break that out, Gabe. Obviously, the ROE between the separate business lines."
— Scott Thomson, President and Chief Executive Officer

Assessment: The refusal was polite, immediate and total, and it came only after the questioner had to ask twice. This is the most important thing management did not say on the call. The International Banking return on equity is the headline metric supporting the "pivot to growth" thesis, and a material and growing share of it is generated by a business with a different capital intensity, a different persistence and a different multiple. Until that split is disclosed, the segment's returns should be discounted, not taken at face value.

Whether 14% is a floor or a peak

Having cleared the medium-term objective a year early, management was asked the natural follow-up: what stops this from being a 15%-plus return-on-equity bank. The answer pointed entirely at Canada and did not offer a number or a date.

Q: "ROE is at 14% or adjusted ROE top of house 14%. I think that's your target and if you're targeting that for next year, so a little earlier than expected. I guess my question is like is there any structural reason why this bank can't be a 15% plus ROE bank? And like what takes you from where you are to that 15% plus?"
— Doug Young, Desjardins Capital Markets

A: "There is more opportunities for sure. And if you think about the Canadian bank... we see a significant opportunity to continue that progression... We are just getting started in Canada. And over the last couple of quarters, you've seen these green shoots. And this quarter, you're starting to see more than green shoots. And that is going to be the biggest driver of the ROE improvement of this bank over the next journey."
— Scott Thomson, President and Chief Executive Officer

Assessment: Management was offered the chance to reset the medium-term target upward on the day it beat the old one and declined, which is the disciplined choice. But the substantive content of the answer is that all further return-on-equity upside is expected to come from Canadian Banking, not from International, Wealth or capital markets. That is a useful concentration of the forward case, and it makes the Canadian loan-growth and margin cadence the whole argument.

What would stop the capital-markets run

A recurring line of questioning across two analysts asked what conditions would slow or reverse capital-markets revenue, and whether a fourth-quarter giveback should be expected. The response was a framework rather than a forecast.

Q: "What are the conditions that caused this to slow or even reverse? And I take you back to last week when there was a fair bit of uncertainty around U.S. treasuries, there was some intervention there. Is that the sort of condition that drives up liquidity and hurts capital markets?"
— Mario Mendonca, TD Securities

A: "You want volatility on the capital market side, but you want constructive volatility. Too much volatility. So if the VIX pops to 50 plus, like, you're going to see your ECM and DCM businesses compress. But when you have constructive volatility and which is what we have right now on a global scale is you have clients trying to navigate really complex environment right now."
— Travis Machen, Group Head, Global Banking and Markets

Assessment: The honest answer, and the one that matters for how this quarter's earnings should be capitalized. Asked directly whether revenue could contract from here, management said the business is hard to predict and declined to reverse-engineer a number. An investor should therefore not extrapolate a C$647M quarterly run rate, and should note that the same management that would not forecast the segment's revenue has just used its record output to clear a bank-level return target.

Global Banking and Markets loan growth and whether 7% is a run rate

Sequential loan growth of 7% in the wholesale book, coming after a period of deliberate shrinkage, drew a question about what a sustainable pace looks like and how it connects to deposits and fees.

Q: "So very strong sequential loan growth, 7%. So obviously, that's by design, but maybe talk a little bit about that in terms of what a reasonable run rate is. Obviously, 7% is not a -- probably not a run rate."
— Paul Holden, CIBC

A: "Loan growth is really just a reflection of the economic output and the focus that we have on our clients, it's not a KPI that we're trying to drive... If you look at some of the data you will see that our return per risk unit are up significantly. You can see that our fees per loan unit are up significantly."
— Travis Machen, Group Head, Global Banking and Markets

Assessment: Redirecting from balance growth to return per risk unit is the right answer from a wholesale bank and consistent with the capital discipline the whole strategy rests on. It is also, in a quarter where book-size risk-weighted-asset growth consumed 23 basis points of CET1, a reminder that the wholesale bank is now the marginal user of the bank's capital as well as the marginal contributor to its earnings.

What the excess capital is actually for

The final question of the call linked the regulator's capital relief to the government's domestic investment agenda and asked whether the freed capital funds infrastructure and defence lending or simply more buybacks.

Q: "I guess, OSFI is giving you another 50 basis points of excess capital. So I'm wondering if part of that capital is going to be used to support that Canada agenda, so defense, infrastructure, AI? Or is that excess capital just going to be used to continue to buy back shares?"
— Stephen Boland, Raymond James

A: "The first call for our capital is organic growth... Now if there is excess capital, well, the first place that, that will go will be share repurchases. And that's because there's a valuation gap, and we still think there's great opportunity. You saw a little bit of that in last quarter, you'll see -- continue to see us renew and do more as we go forward."
— Scott Thomson, President and Chief Executive Officer

Assessment: The buyback rationale is still framed as an arbitrage against a valuation gap, which is the same framing used a quarter ago at a share price 15% lower. That framing has a natural expiry. The more consequential half of the answer is the ordering: organic growth first, buybacks second, tuck-in acquisitions third. With internal capital generation down to 5 basis points a quarter and a 15 basis point regulatory drag ahead, the ordering implies the repurchase pace is the variable that flexes.

What They're NOT Saying

1. The return on equity of the capital-markets business inside International Banking: asked twice, refused outright. It earned C$321M of the segment's C$725M, so the segment's reported 15.7% return is not a franchise return and cannot be interpreted as one.
2. A new medium-term return-on-equity target: the old one was cleared a year early and the response was "we don't see 14% return on equity as a ceiling for the bank," which is a sentence, not a target. Investor Day framing has now been beaten without being replaced.
3. Any fourth-quarter earnings frame: in a quarter carrying three identified headwinds (a 15 basis point capital migration, an unsized Chilean deferred tax write-down, and a capital-markets comparison management will not underwrite), no one offered and no one requested a near-term earnings shape.
4. The size of the Chilean deferred tax asset write-down: disclosed as certain to occur in the October quarter, with no dollar amount, no range and no indication of whether it will be treated as an adjusting item.
5. Remaining exposure to the Brazilian corporate account: a further C$57M of provisions plus a C$14M reclassification, and the only forward statement was that the bank continues to work through it. Two quarters in, the residual has never been sized.
6. The Greater Toronto uninsured mortgage number: 46 basis points, up 5 sequentially and 17 in four quarters, appears in the appendix of the investor deck and was not mentioned on the call. Prepared remarks referred to "some pockets of weakness" without geography.
7. Why internal capital generation fell from 32 basis points to 5: the CET1 walk shows it plainly and no one on the call asked about it, including in the question about what excess capital is for. The answer given assumed the excess still exists.
8. The sequential decline in wealth net sales: C$3.0B against C$4.7B, presented only as a record third quarter and a 14% year-over-year increase. Both true, neither the sequential comparison.
9. Canadian deposits down 2% year over year: framed entirely as a deliberate mix shift out of term, with the retention statistic offered as evidence. The absolute funding base of the segment shrinking for a second consecutive quarter was not addressed as such.
10. Whether the C$8.3B trailing capital-return pace survives the October quarter: the chief executive said buybacks are the first call on excess capital and that the bank would "renew and do more," without sizing a renewed repurchase programme against a CET1 ratio guided to hold at roughly 13% while absorbing a 15 basis point regulatory hit.

Market Reaction

• Pre-print setup: BNS closed at C$120.29 on the Toronto Stock Exchange and $86.86 in New York on Monday, August 24. The Toronto line was up 18.8% year to date, up 51.3% over twelve months, and down 2.0% over the trailing 30 days, sitting roughly 5% below its 52-week closing high of C$126.98. The setup entering this print was a stock that had gone sideways for a month, not one that was extended.
• Reaction session: the bank reports before the open, so the reaction is the print-day cash session. BNS.TO opened at C$122.31, traded a C$122.31 to C$129.25 range, and closed at C$128.73, up 7.0% or C$8.44. The New York line closed at $93.10, up 7.2%. Both are the largest single-session moves in this coverage window.
• Volume: 7.7 million shares in Toronto against a 30-day average of 3.3 million, 2.3 times normal. New York volume was 4.3 million against 2.1 million, 2.0 times normal.
• Benchmark: the S&P/TSX Composite closed up 0.7% and the S&P 500 up 0.3% on the same session, so essentially the entire move was idiosyncratic.
• Peer context: Bank of Montreal reported the same morning and also beat, on capital-markets strength, so Canadian bank financials were bid as a group into a heavy reporting week.

The shape of the session is as informative as its size. The stock gapped up only 1.7% at the open, which is roughly what an 8.6% earnings beat with an in-line balance sheet would justify on the headline alone, and then ground higher through the day to close within 0.4% of the intraday high. That is a market working through the disclosure rather than reacting to the press release, and the thing it found on the way through was the return-on-equity milestone, the credit inflection and the breadth of segment growth.

What the market plainly did not price is the composition question. A 7% single-session move for a bank with a 3.5% dividend yield capitalizes roughly two years of dividends against one quarter of evidence, most of which came from the segment management would not forecast. The re-rating we flagged in May as already having happened continued for another 15% between prints, and the multiple expansion outpaced the earnings revision.

Sell-side reaction over the following session was consistent with that reading: targets moved up sharply across the coverage universe, but most of the houses that raised targets left their rating letters unchanged, with several new targets landing at or barely above the post-print close. The Street raised its estimate of what the shares are worth and largely did not change its view of what to do with them.

Street Perspective

Debate: Is 14% return on equity a floor or a peak?

Bull view: the bull case on the Street is that 14.2% understates the run rate, because Canadian Banking is still improving at 100 to 160 basis points a year on the segment measure, International Banking has just quantified a multi-year revenue and expense path, and Wealth is compounding at a 18.6% return on the smallest capital base in the bank. On this view 14% is the floor of a range that ends closer to 15%.

Bear view: the bear camp contends the milestone was borrowed from a record capital-markets quarter, and that stripping the sequential Global Banking and Markets step-up returns the bank to 13.2%, exactly where it sat in April. On this view the target was touched, not held, and the October quarter reverts.

Our take: the bears have the better of the arithmetic and the bulls have the better of the trajectory. Both are answered by the same data point three months out. What tips us toward caution is that management, offered the opportunity to raise the target on the day it beat the old one, declined to do so, which is the behaviour of a team that does not yet consider the level durable.

Debate: Does a third of earnings from capital markets deserve a bank multiple?

Bull view: the argument gaining traction is that Scotiabank's capital-markets earnings are not a trading book but a client franchise, evidenced by the breadth of the quarter across debt and equity capital markets, advisory, transaction banking and structured credit, and by a 32 basis point margin expansion driven by deposits rather than by risk-taking. Capital-light, fee-based, relationship-driven revenue deserves better than a discount.

Bear view: the bear camp points out that a rival reported the same morning with capital-markets earnings up more than 45%, which makes this a cycle rather than a franchise, and that Global Banking and Markets now consumes C$16.1B of attributed capital, up from C$14.9B a year ago, for a 15.9% return that is below Canadian Banking's 19.4% and Wealth's 18.6%.

Our take: the franchise-widening evidence is real and the mandate list is not the output of a lucky trading quarter. But the capitalization question is separate from the quality question. A bank earning a third of its business-line profit from markets-sensitive revenue at a 15.2x multiple is being paid a full price for its most cyclical dollar, and the disclosure needed to underwrite that dollar, starting with the return on the international capital-markets book, has been withheld.

Debate: Is the credit turn real, or is the mortgage book a slow leak?

Bull view: the constructive read is that the turn is unambiguous: impaired provisions fell after three consecutive increases, formations dropped C$452M, every unsecured Canadian retail product improved sequentially, the mid-50 basis point guide held on its first test after being revised, and the allowance was built for a tariff scenario materially worse than the one that has arrived.

Bear view: the skeptical read is that the composition is deteriorating even as the level improves. Gross impaired loans hit their highest ratio of the sequence at 100 basis points, Canadian mortgage delinquency rose for a fourth consecutive quarter, and the Greater Toronto uninsured book is now deteriorating half again as fast as the national uninsured average.

Our take: both are true and they operate on different clocks. The provision line turns with the cycle and has turned; the mortgage delinquency line turns with payment shock in specific vintages and has not. Given mortgages are by far the largest Canadian retail balance, roughly 93% of that book is secured, and the affected geography carries an average FICO of 807 at an uninsured loan-to-value near 60%, this is a drag on provisions rather than a solvency question. It is nonetheless the one line in the bank that has moved the wrong way in every quarter we have covered.

Debate: Has the re-rating gone too far?

Bull view: the case being made on the Street is that 15.2x annualized year-to-date adjusted earnings is unremarkable for a bank that has just proven a 14% return on equity and a 17.2% return on tangible common equity, that the multiple compresses to roughly 14x on a normalized fiscal 2027, and that the discount to Canadian peers that persisted through the restructuring years should not survive its completion.

Bear view: the bear camp contends the shares have re-rated 15% in a single quarter against a 4% increase in annualized earnings, that 2.45x average tangible common equity embeds a perpetual growth rate near 5% at a 10% cost of equity, and that the dividend yield has compressed from 4.1% to 3.5% in three months.

Our take: the multiple is now a fair price rather than a discount, which is the same conclusion we reached in May at 13.7x, reached again at a higher price and a higher earnings base. That is not a bearish statement about the business. It is a statement that the improvement and the price have moved together, and that further return from here requires the fiscal 2027 return-on-equity level to be exceeded rather than merely sustained.

Model Update & Valuation Framework

First, how our May assumptions performed against the quarter they were written for.

The expense miss is the only material one, and it is a coherent miss: we assumed 6% to 7% ex-divestiture expense growth while explicitly flagging performance compensation as tracking capital-markets revenue. Capital-markets revenue grew 32% and the compensation followed. The correct lesson is not that cost control failed but that our expense line should be modelled as partly variable against Global Banking and Markets revenue rather than as a fixed growth rate.

Revised assumptions from here.

Valuation

Average tangible common equity is derived from the bank's own disclosure: adjusted net income to common shareholders of C$2,798M annualized on a 92-day quarter is C$11,101M, which against the reported 17.2% return on tangible common equity implies C$64,542M. The same annualization against the disclosed C$78,217M of average common equity reproduces the reported 14.2% return on equity, which validates the method.

Valuation impact. Between the two prints, annualized year-to-date adjusted earnings rose from C$8.14 to C$8.47, an increase of 4%, while the share price rose 15%. The entire remainder is multiple. At 15.2x forward-ish earnings and 2.45x average tangible common equity against a 17.2% return on tangible common equity, the shares embed a perpetual growth rate near 5% at a 10% cost of equity. That is a demanding assumption for a bank generating a third of its business-line profit from capital markets and roughly a quarter from International Banking, and it is not obviously wrong for a bank that has just cleared its return target a year early. It is, however, a full price.

On C$8.50 of fiscal 2026 adjusted earnings, a 14x to 16x band produces a fair-value range of roughly C$119 to C$136, with a midpoint near C$127.50. The shares closed at C$128.73, inside that band and marginally above its midpoint, which implies a total return over the next twelve months roughly in line with the market once the 3.5% dividend is included. That is the definition of a Hold.

What would change the rating. We set an upgrade test in May and it is worth being explicit about how it fared. The test was two of three: Canadian commercial and total loan growth reaching market rates in the October quarter, the impaired provision ratio printing at or below the mid-50s without a further guidance revision, and Global Banking and Markets returning to positive operating leverage. The second condition was met cleanly. The third was met on the letter, at 6 points in-quarter, though year-to-date segment operating leverage is only 0.9% and the improvement came from record revenue rather than cost discipline. The first was written for the October quarter and cannot be tested yet; Canadian total loan growth remains at 3% year over year, unchanged from April.

We are therefore not upgrading, and the reason is a judgment about the test rather than about the quarter: the condition we considered most important was deliberately scheduled for the print that has not happened, and in the interval the shares absorbed the improvement and then some. To avoid a moving target, here is the test for the October quarter, and it is final. Upgrade to Outperform requires adjusted return on equity at or above 14% with a Global Banking and Markets contribution below C$580M, plus either Canadian total loan growth at 4% or better year over year or an impaired provision ratio at or below 52 basis points. Downgrade to Underperform requires Canadian mortgage 90-plus-day delinquency above 38 basis points — a level reset from the 35 basis points named in May, since the metric printed 34 and the May threshold would now trip on a single basis point of further drift — a new single-name corporate impairment in International Banking distinct from the Brazilian account, or a CET1 ratio below 12.75% that forces the repurchase programme to pause.

Thesis Scorecard Post-Earnings

Overall: The thesis strengthened operationally and weakened on valuation, which nets to unchanged. Four of the six standing pillars moved in the bank's favour, including both of the credit and international bear points that had been escalating. The two that moved against it are the two that determine the rating: the capital story stopped self-funding, and the re-rating that kept us at Hold in May extended another 15%. A seventh pillar has been added for the earnings-mix shift, which is the genuinely new risk this quarter introduced.

Action: Hold. This is a better bank than it was three months ago and a more expensive one by a wider margin than the improvement justifies. We would add on a pullback toward the low C$120s, which is the bottom half of our fair-value band, and we would upgrade on an October quarter that holds 14% return on equity without a record capital-markets contribution.