The Guidance Moved Sideways and Got Better: Initiating ING at Outperform

Results vs. Consensus

1Q2026 Scorecard

ING is a euro-reporting IFRS-EU bank and the consensus that matters is the euro-denominated poll the company itself compiles from contributing brokers before each print. Sixteen brokers submitted for this round. That is the basis the Amsterdam line trades against and the basis used below. Dollar-converted vendor estimates circulate widely for the New York depositary receipt and are not comparable to a euro income statement; they are excluded here. The bank reports before the European open, so the print landed on the morning of 30 April and the analyst call followed at 09:00 CET the same morning.

Two mechanical points before the interpretation. First, the consensus is a set of line-item medians, so the individual lines do not add to the subtotals; a component-by-component bridge will not reconcile exactly to the €136M profit-before-tax gap, and roughly €21M of that gap is non-additivity rather than performance. Second, earnings per share cleared the top of the poll range while net result did not: the €1,556M result sat inside a range topping out at €1,576M, but €0.54 matched the €0.54 maximum exactly. The difference is the share count. Shares outstanding fell to 2,876.6M from 3,050.2M a year ago, a 5.7% reduction, and the buyback has been running faster than the poll modelled.

Year-Over-Year Comparison

The income statement below is as filed. Note the two net-interest-income lines: commercial net interest income is the company's own measure covering lending and liability products, while the IFRS line adds the funding costs and interest of Financial Markets and Treasury positions whose revenue is booked in other income. The two grew at very different rates this quarter, and reading only the IFRS line overstates what happened.

The single most revealing pair in that table is the two interest-income lines. Commercial net interest income grew 7.0%. Reported net interest income grew 12.0%. The 5-point gap is not performance; it is other net interest income improving from negative €172M to negative €5M, which reflects Treasury funding costs whose offsetting revenue sits in other income. Other income fell 41.1% over the same period. Anyone marking ING on reported net interest income growth of 12% is double-counting a reclassification of where Treasury economics land in the P&L.

Quarter-Over-Quarter Comparison

Revenue. Total income of €5,823M was 0.3% ahead of the poll and 3.3% ahead of the year-ago quarter. That is the least interesting fact about it. The composition changed materially: commercial net interest income of €4,060M was 2.3% ahead of the median and 7.0% ahead of last year, fee income of €1,236M was 2.6% ahead and 13.0% higher, and all other income of €528M was 18.1% below the median and 29.5% below the year-ago quarter. The recurring, balance-sheet-driven lines beat. The market-sensitive line missed. Within all other income, Financial Markets fell to €286M from €406M a year ago and hedge ineffectiveness swung to negative €81M from negative €10M. On the company's own volatile-items reconciliation, income carried a negative €71M of volatile items this quarter and expenses a further €30M, so profit before tax excluding volatile items was €2,359M against €2,258M reported. The revenue quality is better than the revenue level suggests.

Margins. The commercial net interest margin was 2.26%, flat year on year and 3bp better sequentially, and the whole of the sequential improvement came from the liability side. The lending margin was unchanged at 1.26% while the liability margin rose 5bp to 1.04%. Management was explicit that this rate of improvement will not repeat, and gave two reasons: the deposit-replication hedging tailwind arrives gradually rather than linearly, and the first quarter carried unusually low campaign-related acquisition costs because no large savings campaign was run. The cost/income ratio of 55.3% was 150bp better than a year ago and 230bp better sequentially, but the sequential comparison flatters: the fourth quarter carries the annual Dutch bank tax and this quarter carries the Belgian equivalent, and the fourth quarter also absorbed €104M of incidental restructuring against €30M here.

Earnings per share. The €0.54 print was up 14.9% year on year against a net result up 6.9%. Eight of those fourteen points are the share count. This is the mechanism that makes ING work as an equity even in a quarter where operating profit grows in the mid single digits: the bank has retired 5.7% of its shares in twelve months, and on a four-quarter rolling basis earnings per share is €2.18 against €1.97 a year ago, up about 11%. Whether that mechanism keeps running is the capital question addressed later in this note, and it is the single most important thing to grade next quarter.

Segment Performance

ING runs two operating segments plus a Corporate Line. Retail Banking is the larger and the higher-returning; Wholesale Banking is the capital-intensive, more cyclical half. The Corporate Line carries foreign currency hedging, the minority stakes, and from this quarter the run-off of Luxembourg activities for business-banking clients and private individuals, which were transferred out of Retail Belgium.

Retail Banking

Retail Banking produced €1,620M of profit before tax on €4,039M of income, a 12.2% year-on-year increase in profit against a 6.5% increase in revenue, and a 20.1% return on equity measured against 13% of risk-weighted assets. Commercial net interest income of €3,021M grew 8.4% year on year and 3.7% sequentially, faster than the group, and that is where the deposit franchise earns its keep: €9.4bn of net core lending growth of which €5.9bn was mortgages, and €4.3bn of net core deposit growth into savings and term products despite the seasonal current-account outflow the first quarter always brings.

Fee income grew 13% year on year across every market, with investment products the standout. Active investment-product customers reached 5.2 million, up 8%, assets under management and e-brokerage reached €281bn, up 15% with roughly half of that from net inflows rather than markets, and the total number of trades rose 13%. Two composition caveats belong with that number. The fourth quarter of 2025 carried a €66M positive one-off in Retail lending fees from a retroactive reclassification of brokerage expenses in Germany, which is why lending fees fell to €297M from €349M sequentially. And from this quarter daily banking fee income benefits from a structural shift of revenue out of other income, which is part of why that line rose to €441M from €414M. Neither is disclosed in euros.

"Investment products, in particular, performed very well, a record quarter even benefiting from 8% growth in customers with an investment account and 15% growth in assets under management and administration, of which roughly half comes from net inflows, while also benefiting from 13% more trades, which besides a higher customer base, was supported by the increased market volatility towards the end of the quarter."
— Ida Lerner, Chief Financial Officer

The blemish is credit. Retail risk costs rose to €275M from €175M a year ago and €177M in the fourth quarter, taking the division to 21bp of average customer lending against 14bp in both comparable quarters. A €45M management overlay for higher energy prices and the secondary effects of the Middle East conflict accounts for less than half of the increase. The rest was net additions in business and consumer lending, with mortgage-related risk costs described as remaining low.

Assessment: This is a division compounding customers, balances and fees at double-digit rates with a cost line growing 1.1%, which is the definition of operating leverage, and it is delivering a 20% return on allocated capital while doing it. The credit line is the one to watch: a 7bp year-on-year increase in a division this size is €100M of pre-tax profit in the quarter, close to €400M annualised if it persists, and only part of it is a discretionary overlay that management says should diminish.

Retail Netherlands

The best franchise in the group and it is not close. A 36.5% cost/income ratio, a 33.1% return on allocated equity, 8bp of risk costs, and €800M of profit before tax on €1,319M of income. Commercial net interest income of €1,000M was up from €959M in the prior quarter and €881M a year ago. Regulatory costs were nil in the quarter because the annual Dutch bank tax is booked in the fourth quarter, which is worth remembering when the sequential comparison against €594M of fourth-quarter operating expenses looks flattering. Net core lending grew €4.1bn while core deposits were essentially flat, with savings inflows from private individuals offset by business-banking withdrawals tied to first-quarter tax payments.

Assessment: The Dutch book is the engine and the mortgage floor expiring in December frees €4bn of risk-weighted assets against it. There is nothing to fix here; the question is only how much of the group's growth can be reallocated toward markets earning less than a third of this return.

Retail Belgium

Profit before tax of €38M on €676M of income, a 2.4% return on allocated equity and an 81.3% cost/income ratio. Two things drive that. The Belgian deposit guarantee scheme contribution and the Belgian bank tax are both booked in full in the first quarter, putting €182M of regulatory costs into a single three-month period. And risk costs of €88M at 36bp, described as primarily business lending, more than doubled the 16bp run-rate of both comparable quarters. Underlying expenses excluding regulatory costs and €7M of restructuring did decline both year on year and sequentially, helped by the Luxembourg run-off transferring to the Corporate Line.

Assessment: Even normalising the regulatory seasonality, Belgium is the structurally weakest retail franchise in the portfolio, and the 36bp credit print is the first quarter in the disclosed run where that weakness showed up in provisions rather than just in costs. Restructuring provisions taken this quarter for headcount reduction in Belgium suggest management agrees. It got no attention on the call.

Retail Germany

A 25.0% return on allocated equity on 46.3% costs, with profit before tax of €350M roughly flat year on year and down from €435M sequentially. The sequential decline is almost entirely fee composition: the €66M brokerage reclassification one-off landed in the fourth quarter, and excluding it fee income rose sequentially on investment products. The German deposit base is enormous relative to the lending book, at €159.2bn of deposits against €117.4bn of loans, which makes Germany the single largest beneficiary of the liability-margin improvement and the single most exposed to deposit competition.

Assessment: Germany is where the replication tailwind has the most balance sheet to work on, and where a savings-rate war would hurt most. The absence of a large German savings campaign this quarter is why the liability margin rose 5bp; management said as much and warned it will not repeat.

Retail Other

The remainder of the retail footprint, spanning Poland, Spain, Australia, Italy, Romania and Türkiye, delivered €431M of profit before tax on €1,308M of income at a 15.8% return, down from 19.1% a year ago. Net core lending grew €3.1bn, of which €2.1bn was mortgages in Italy, Australia, Spain and Poland, and deposits grew another €3.1bn. The return decline is credit: risk costs of €103M at 32bp against 26bp a year ago, with net additions primarily in Spain and Poland. Türkiye carries a cost/income ratio above 100% under hyperinflation accounting, and the IAS 29 hyperinflation indexation it drives ran €25M negative through group income this quarter; Italy also runs above 100% as Business Banking launches.

Assessment: This is the growth portfolio and it is priced as such internally, absorbing the sub-scale losses of new markets against the profits of Poland and Spain. The 6bp year-on-year rise in risk costs concentrated in exactly the two markets carrying the most consumer exposure is the thing to track. Poland also now carries a materially higher corporate income tax rate for banks from 2026, which management flagged as pushing the group effective rate up.

Wholesale Banking

Wholesale Banking is the quarter's paradox: income fell 1.3% year on year and 0.7% sequentially, yet profit before tax rose 1.3% year on year and 36.7% sequentially, and the return on allocated equity improved to 10.7% from 8.2% in the fourth quarter. The entire improvement came from provisions. Risk costs fell to €62M from €138M a year ago and €188M sequentially, at 12bp against 38bp last quarter, because a €49M management overlay for the Middle East conflict was more than offset by a provision release following a large repayment of a Stage 3 loan.

The capital story inside Wholesale Banking is better than the income story. Net core lending grew €5.6bn, driven by a €7.8bn gross inflow in Lending partly offset by the repayment of a short-term Working Capital Solutions facility, and risk-weighted assets were essentially unchanged at €149.2bn. Income over average risk-weighted assets held at 466bp, identical to a year ago. Growing customer lending 3.3% in a quarter, from €203.1bn to €209.9bn, without consuming capital is what management means by capital velocity, and it is the reason the division's return improved on falling revenue.

"Within Wholesale Banking, the risk-weighted assets remained broadly stable despite strong lending growth, reflecting the continued capital velocity measures that have been taken within Wholesale Banking."
— Ida Lerner, Chief Financial Officer

Assessment: Two of the four income lines are genuinely improving and two are not. Lending and fee income are compounding on client activity; Financial Markets and Treasury are hostage to the rate path and both were hit by the March move. The 10.7% return on allocated equity is still barely above a plausible cost of equity, and the €62M provision print is the least repeatable number in the whole release. Underwrite this division on 466bp of income over risk-weighted assets and a normalised 25–30bp of risk costs, not on this quarter's 12bp.

Corporate Line

A €81M pre-tax loss against a €30M loss a year ago, on income of €48M against €88M. Foreign currency hedging income fell to €71M from €152M, and the €39M interim dividend from the Bank of Beijing stake that landed in the first quarter of 2025 was not repeated. The Luxembourg run-off now reports here.

Assessment: Small in absolute terms but moving the wrong way, and the year-on-year deterioration accounts for roughly a fifth of the gap between group revenue growth of 3.3% and Retail Banking's 6.5%.

Key Performance Indicators

The deposit line needs its caveat stated plainly, because it looks alarming. Net core deposit growth of €7.2bn against €22.6bn a year ago is not a franchise problem; the first quarter of 2025 included a large German savings campaign that pulled in €15.3bn in a single quarter and partially ran off again in the third. Stripped of campaign effects, €7.2bn is roughly the quarterly run-rate consistent with the bank's stated ambition of around 5% annual growth on both sides of the balance sheet, and it was achieved at a 5bp better liability margin precisely because no campaign was run.

Key Topics & Management Commentary

Overall Management Tone: Confident and unusually specific on the mechanics, with the confidence concentrated in the balance sheet and the caution concentrated in everything priced by a market. Management pre-empted the obvious extrapolation on the liability margin before a single analyst asked, which is the posture of a team that expects to be held to its guidance rather than praised for it. Where it was least convincing was on the implied cost path for the rest of the year, which was answered with a strategy narrative rather than a number.

1. The Liability Margin Rose 5bp, and Management Immediately Said It Will Not Again

The single largest driver of the sequential earnings improvement was liability net interest income, which rose €91M quarter on quarter to €1,795M on a combination of deposit volume growth and a 5bp rise in the liability margin to 1.04%. Lending net interest income added only €41M over the same period. Management volunteered the caveat in the prepared remarks rather than waiting to be challenged on it, and gave a specific mechanical reason: no large savings campaign ran in the quarter, so acquisition costs were abnormally low.

"What it also reflects is the absence of larger savings campaigns during the first quarter, meaning that the level of acquisition costs was relatively low this quarter and will likely normalize again in the coming quarters. As such, let me be clear that we should not expect a 5 basis points increase of the liability margin every quarter ahead."
— Ida Lerner, Chief Financial Officer

The guided range is the mid-point of 100 to 110bp for 2026, with the acknowledgement that 2027 and 2028 could temporarily exceed 110bp if the March forward curve materialises. Roughly 55% of the retail eurozone replicating portfolio has an average remaining maturity between one and fifteen years, which is what makes the tailwind multi-year rather than a rate call. The disclosed sensitivity is that every 10bp of pass-through on total savings and term deposits costs about €0.4bn of commercial net interest income.

Assessment: Pre-empting the extrapolation is the right call and it protects the guide, but it also means the cleanest driver of the quarter is explicitly flagged as non-recurring at this rate. What survives is the replication tailwind itself, which is contracted and visible. The competitive risk is the pass-through, and at 1.04% the margin is still a basis point under the 105bp mid-point of the guided range, with three quarters left to run.

2. The Guidance Change Is a Composition Swap Disguised as an Upgrade

Two guidance items moved in opposite directions on the same call. Commercial net interest income for 2026 was raised to €16.5–16.7bn, roughly €200M higher at the midpoint. All other income was reduced to €2.5–2.7bn. Total income for 2026 was confirmed at around €24bn, unchanged.

"Looking ahead, on the back of a very strong first quarter and especially the higher-than-expected volume growth, we can expect a slightly higher level of commercial NII than previously guided. We now expect commercial NII for the full year to be between EUR 16.5 billion and EUR 16.7 billion."
— Ida Lerner, Chief Financial Officer

Pressed on where the increase came from, management was unambiguous that it is not a lending story.

"So I think on the EUR 200 million, that is basically all -- the increase is all liability income. ... So it has nothing to do with lending or lending margins. It's just a matter of the volumes that we expect at higher margins and a better replication rate."
— Steven van Rijswijk, Chief Executive Officer

Assessment: This is the most important thing that happened on the call and it will be under-priced because the headline total did not move. Roughly €200M of income is migrating out of hedge ineffectiveness, Financial Markets and Treasury results, which are unpredictable quarter to quarter and carry no franchise value, and into deposit replication income, which is contracted, visible several years forward, and capitalised at a higher multiple by any sensible investor. Same level, better earnings. The market is generally slow to pay for a mix improvement at constant revenue, which is where the opportunity sits.

3. Only About a Third of the Replication Uplift Reached the Guidance

Analysts read the deck's replicating-income chart as showing gross replicating income on retail eurozone customer deposits rising by roughly €600M in 2026 under the March forward curve versus the December curve, a characterisation management engaged with rather than disputed. The commercial net interest income guidance went up €200M, implying a pass-through to depositors of close to 70% of the gross uplift. That gap became the most persistent line of questioning on the call, raised independently by two analysts, and management's answer was consistent: the deck is an illustration of a forward curve, not a forecast, and the difference is deposit competition.

"Taking purely the forward curve from March into account, you would say that, yes, we would potentially be higher than 100 basis points and 110 basis points. But we also know that there is a fierce competition. There's also a very rational behavior in the bank, focusing on profitability above growth over time."
— Ida Lerner, Chief Financial Officer

Assessment: Withholding two-thirds of a rate-driven gross uplift from the guide is conservative and consistent with how this management team has framed the replication portfolio, which they describe as a risk-management tool rather than a positioning tool. It also means the guide has genuine headroom if pass-through stays where it is. That is the single largest source of upside to the 2026 number, and it is deliberately not in it.

4. Cost Growth of 1.1% Is Not the Run-Rate the Guidance Implies

Expenses excluding regulatory costs and incidental items were €2,865M, up 1.1% year on year, against wage inflation across nine retail markets. Including regulatory costs and excluding incidentals, first-quarter operating expenses were €3,189M against €3,196M a year ago, essentially flat. The full-year guide of €12.6–12.8bn was confirmed unchanged, and that guide implies a step-up. Full-year 2025 operating expenses on the same basis were roughly €12.3bn, so the guide asks for 2.5% to 4.2% growth against a first quarter that delivered none of it.

"over the past 12 months, we have grown our mobile primary customer base by almost 7%, our customer balances by more than 5%, our volumes in investment products by more than 15% and fee income even by 15.6%. But our FTEs, however, decreased by 0.6%, while our cost growth was limited to 2%."
— Steven van Rijswijk, Chief Executive Officer

Management was asked directly whether the first quarter put them ahead of the full-year guide, and declined to move the number, describing instead the levers available between investment and cost.

Assessment: The scalability evidence is real: 7% customer growth and 15.6% fee growth against a headcount that shrank is not a slogan, it is a disclosed set of numbers. But holding the cost guide after a flat first quarter is an implicit statement that spending accelerates from here, and management chose to characterise that as optionality on investment rather than quantify it. Either the guide is conservative, which is the favourable reading, or the second half carries a visible cost ramp. It matters because the roughly 52% cost/income ratio implied by the 2027 revenue and cost targets rests on which one is true.

5. The Capital Model Changed, and the Change Was Presented as Cosmetic

From this quarter ING reserves all potential distributions outside CET1 capital up front, in addition to the standing 50% payout policy, to comply with European Banking Authority guidance. The transition cost 23bp of CET1 as a one-off. The full €1,556M net result for the quarter went into reserved profits rather than into CET1 capital, and available CET1 capital was broadly flat at €44.7bn against €44.6bn at year-end even as €1.5bn of profit was earned.

"The implementation of this new reserving approach had a one-off effect this quarter of minus 23 basis points. In total, the additional distribution has an impact of roughly 29 basis points on our core equity Tier 1. This is merely a change in reserving approach. Our distribution policy remains unchanged."
— Ida Lerner, Chief Financial Officer

The economics are indeed unchanged; the timing of recognition is not. The consequence is that the reported CET1 ratio no longer builds visibly from retained earnings between distribution announcements, and the test that governs buybacks, whether there is structural capital above roughly 13%, is applied to a ratio that now sits at exactly 13.0% with a pro-forma of 12.9% after the announced distribution.

Assessment: Calling this "merely a change in reserving approach" is accurate about the economics and understates the change in how an outsider should read the capital ratio. Under the old approach an investor could watch CET1 accrete and infer the next distribution. Under the new one the accretion is pre-committed and the incremental distribution capacity has to come from somewhere else, which in practice means risk-weighted asset relief. That is a less automatic, more managed capital return.

6. The Buyback Runway Now Depends on Risk-Weighted Asset Relief

Two specific sources were quantified on the call. The Dutch mortgage floor expires on 1 December 2026 following a decision by the Dutch central bank, releasing €4bn of risk-weighted assets worth about 15bp of CET1. And significant risk transfers are planned to deliver a further 15 to 20bp of capital relief in 2026, following two completed transactions in Wholesale Banking that produced 12bp last November.

"the Dutch mortgage floor expires as per the 1st of December 2026. And that decision will lead to a EUR 4 billion lower risk-weighted assets. So that's about 15 basis points of our CET1 ratio."
— Steven van Rijswijk, Chief Executive Officer

"we have previously also said that we expect to do additional capital reliefs in 2026 of between 15 to 20 basis points and that still remains the plan. We have a very good and constructive dialogue with ECB."
— Ida Lerner, Chief Financial Officer

Combined, 30 to 35bp on €344bn of risk-weighted assets is roughly €1.0 to €1.2bn of distributable capacity, and management explicitly said the mortgage floor benefit will be assessed inside the semi-annual distribution review in October rather than treated as a separate event.

Assessment: The runway is real and quantified, which is more than most European banks provide. It is also finite and increasingly technical. A distribution policy that depends on model and regulatory changes rather than on earnings retention is more forecastable in the near term and less durable in the long term. The favourable reading is that a bank running at exactly its target ratio while retiring 5.7% of its shares a year has solved capital allocation; the sceptical reading is that the easy sources of relief get used once.

7. Risk Costs of 19bp Are Below Trend, and Two Large Items Cancelled Out

Group risk costs of €346M at 19bp sit just below the through-the-cycle average of around 20bp. Underneath, a €94M addition to management overlays for higher energy prices and the broader effects of the Middle East conflict was largely offset by a provision release following a large repayment of a Stage 3 loan in Wholesale Banking. Stage 3 risk costs were €297M against €389M in the prior quarter; Stage 1 and Stage 2 risk costs were €50M.

The divisional split tells a different story from the group number. Retail Banking rose to €275M and 21bp from €175M and 14bp a year ago. Wholesale Banking fell to €62M and 12bp from €138M and 29bp. One division's credit deteriorated and the other's was rescued by a single repayment, and the two moves happened to net to a benign group figure.

Assessment: The group print is the least informative number in the release. Retail credit is drifting in Belgium, Spain and Poland, and the Wholesale offset is a single-name event that will not repeat. Normalising both, group risk costs are closer to the 20bp through-the-cycle average than to a below-trend outcome, which is consistent with the €1,486M full-year consensus. This is not a credit problem; it is a reason not to extrapolate 19bp.

8. The Overlay Is Scenario-Driven and Scheduled to Unwind

Asked what oil price the €94M overlay assumes, the Head of Risk declined to give one and framed the overlay as an adjustment to quarter-end macroeconomic scenarios across a wider set of variables than the oil price alone, with a specific commitment on how it unwinds.

"From the coming quarter, we expect to revert to the normal process, whereby macroeconomic consensus is feeding naturally into our loan loss provisioning process. And therefore, this overlay should diminish, while the net impact on the loan loss provisions will be actually depending on how the, let's say, higher oil price will affect the macroeconomic outlook"
— Andrea Cesaroni, Head of Risk

Assessment: A dated commitment to revert to the standard provisioning process is a checkable promise, which is what makes it useful. It also transfers the risk rather than removing it: if the macro consensus deteriorates, the overlay unwinds into a model-driven charge of similar or greater size. The right way to hold this is as a forward test for the second quarter, not as a release already banked.

9. Fee Income Grew 13%, and Roughly a Quarter of It Was Not Alpha

Fee income of €1,236M grew 13.0% year on year and 1.2% sequentially, with growth in every product category and every market. Management characterised the growth as approximately 75% alpha driven, meaning customer-count and cross-sell effects rather than market levels. Investment products led at €354M against €310M in the prior quarter, and daily banking fees reached €441M.

Product-line figures are rounded and may not sum exactly to the total. Two composition effects matter for anyone extrapolating. The prior quarter's lending fee line carried a €66M positive one-off from a retroactive reclassification of brokerage expenses in Germany, which is most of the €52M sequential decline in that row. And from this quarter fee income reflects reclassifications between product categories and a structural shift of revenue from other income into daily banking fees, which is part of why daily banking rose €27M sequentially. Neither effect is quantified in the current quarter's disclosure.

Assessment: Fee income is the strategic centre of this story and the growth is broad enough to be credible: 8% more investing customers, 15% more assets under management with half of it net inflow, 13% more trades, 14% growth in insurance fees. The disclosure is the weak point. A bank running a diversification thesis on the fee line should quantify a reclassification that flatters its most-watched sub-line, and this one did not.

10. A New CFO, With No Change of Message

Ida Lerner joined as Chief Financial Officer on 1 April 2026 and this was her first results presentation. She took the majority of the technical questions, including all of the liability-margin and replication work, and answered them with specific numbers and specific caveats rather than deferring to the Chief Executive. The Head of Risk also presented, which is not universal on European bank calls.

Assessment: No detectable discontinuity in guidance philosophy, which for a first call is the outcome to want. The one tell worth logging for next quarter is that the caveats came before the questions rather than after them, which suggests a preference for guiding to numbers that can be met rather than to numbers that impress.

11. Moody's Downgraded ING Bank, and It Was Not About ING

On 21 April Moody's revised ING Bank N.V.'s long-term issuer and senior unsecured ratings from A1 to A2. The stated cause is the introduction of full depositor preference across the European Union under the Crisis Management and Deposit Insurance package, which structurally subordinates senior unsecured creditors relative to depositors. All other ratings were unchanged, and ING Groep N.V. remains A- at S&P and A+ at Fitch. The change was disclosed in the release and drew no questions on the call.

Assessment: A legislative reclassification affecting every European bank, not a credit judgement on ING. Marginally relevant to wholesale funding cost, immaterial to the equity. Worth logging only so that a future funding-cost move is not misattributed to it.

Guidance & Outlook

The 2026 and 2027 outlook was confirmed rather than upgraded, with two component-level changes inside an unchanged total. The outlook explicitly excludes the previously announced sale of the Russian business, where an expected post-tax P&L impact of around €0.8bn sits outside every guided figure below.

The commercial net interest income raise was quantified at around €200M at the midpoint and attributed entirely to liability income. The all other income reduction was described qualitatively as "slightly lower than our normal run rate" with a range attached but no prior figure restated. Since total income guidance held at around €24bn while one component rose €200M, the arithmetic requires the other to fall by a similar amount, which the €2.5–2.7bn range delivers: it sits roughly €250M below where the pre-print consensus had all other income for the year.

Implied quarter-over-quarter ramp. The commercial net interest income guide of €16.5–16.7bn against €4,060M delivered requires €12.44–12.64bn over the remaining three quarters, or €4,147–4,213M per quarter. That is a 2.1% to 3.8% step up from the first quarter, achieved with a liability margin management says will not keep rising at 5bp a quarter, which means the remaining growth has to come from volume. At 8.3% annualised net core lending growth against a roughly 5% ambition, the first quarter delivered that volume with room to spare.

Implied cost ramp. First-quarter operating expenses excluding incidentals were €3,189M. The €12.6–12.8bn guide leaves €9.41–9.61bn for the remaining nine months. Regulatory costs are heavily front-loaded into the first quarter by Belgian accounting and back-loaded into the fourth by the Dutch bank tax, so the underlying quarterly expense line has to rise from €2,865M toward roughly €3.0bn to reach the guide. That is the number to test at the half.

Street position. The pre-print consensus had 2026 total income at €24,219M, above the guided figure of around €24bn, 2026 return on tangible equity at 13.9% against a guide of above 14%, and 2027 return on tangible equity at 15.2% against a guide of above 15%. Consensus is therefore already at or slightly above the guide on the headline profitability metrics, which means the guide itself is not the upside path. The upside path is the withheld portion of the replication uplift and the risk-weighted asset relief.

Guidance style. Conservative and mechanically defended. Management raised the one item it could evidence with a delivered quarter, cut the one item it could not control, held the total, and refused to move the cost line despite being invited to. That is a team guiding to numbers it expects to beat rather than to numbers that flatter the quarter.

Analyst Q&A Highlights

Whether the Liability Margin Keeps Climbing Into the Second Quarter

The first question of the call went straight to the quarter's cleanest driver, asking whether the short end of the curve offers further support in the second quarter and how deposit pricing behaved through April. Management confirmed competitive conditions were rational, declined to extend the 5bp trajectory, and reframed the answer around a full-year range rather than a quarterly path.

Q: "The first one will be on the liability margin, the 104 bps. Clearly, we should not replicate the plus 5 bps quarter-on-quarter. But objectively, looking into the second quarter, yes, it looks like there's further support from the short end of the curve. So I wanted to confirm that with you, if you see that as well."
— Benoit Petrarque, Kepler Cheuvreux

A: "we expect to be in the mid-range of between 100 and 110 basis points this year, also driven by a hedging tailwind, which comes in gradually, but not exactly linear and particularly a reflection of the lower-than-usual campaign-related deposit cost in the first quarter."
— Ida Lerner, Chief Financial Officer

Assessment: The question conceded the non-repeatability before management had to defend it, which tells you the Street had already worked out where the sequential improvement came from. The answer put a full-year band around a quarterly number, which is the correct way to guide a margin that moves on campaign timing. At 104bp the bank is a basis point below the 105bp midpoint of its own 100 to 110bp range, so a mid-range full-year average requires roughly 105bp across the remaining three quarters: flat to slightly higher from here, not a step down.

Why Only a Third of the Gross Replication Uplift Reached the Guidance

The most substantive exchange of the call, raised independently by two analysts, concerned the gap between a roughly €600M increase in illustrated gross replicating income and a €200M increase in guided commercial net interest income. The question was whether the haircut reflects uncertainty about where rates settle or an expectation of higher deposit competition.

Q: "of the EUR 600 million increase in replicating income in 2026 again on Slide 27, I know that's a gross number, but how much is included in the new commercial NII guidance? And the haircut you're taking in deciding how much of that EUR 600 million to embed in the new guide? Is that because you're waiting to see where rates really settle this year? Obviously, there's a huge amount of volatility, or because you actually see more price competition coming through on deposits and there being a bigger difference between the gross and the net number?"
— Multiple analysts incl. Chris Hallam, Goldman Sachs; Shrey Srivastava, Citi

A: "some of the benefit from higher short-term rates is from current account volumes and therefore, structurally accretive to NII. However, most of the benefit for us comes from the savings volumes, which is more sensitive to competition and historically has shown that the margins are fairly stable over time and is expected to also come down."
— Ida Lerner, Chief Financial Officer

Assessment: Management answered the question that was asked, which was competition rather than rate uncertainty. The distinction between current-account benefit, which is structural, and savings benefit, which competes away, is the right frame and is the reason the haircut is roughly two-thirds. For the investment case this is the most encouraging exchange on the call: the guide contains the durable third and excludes the contestable two-thirds, so a benign competitive environment is upside rather than base case.

Whether Commercial Momentum Survives the Macro Backdrop

A question on whether the quarter's lending momentum was holding up into a deteriorating macro drew a segmented answer: retail demand is driven by unemployment and housing shortage and is largely insulated, while wholesale demand is the exposed line.

Q: "the commercial momentum was very strong in Q1. And Steven, you called out momentum in mortgages, also growth in business banking. How is this evolving now considering the change in the macro backdrop? So are you still seeing good demand for loans or has that slowed down?"
— Giulia Miotto, Morgan Stanley

A: "I think the biggest impact that we could potentially see, but it's too early to call, is that when we look at the lending demand in Wholesale Banking. ... But with all the uncertainty going on, yes, that could be more muted in the quarters to come, but let's see what happens."
— Steven van Rijswijk, Chief Executive Officer

Assessment: The only forward-looking concession management made all call, and it was volunteered rather than extracted. Wholesale lending grew at an annualised 8% in the quarter against a 4 to 5% historical norm, so flagging it as the line most likely to moderate is both honest and a quiet warning that the volume half of the commercial net interest income guide leans on Retail from here.

What Macro Assumptions Sit Behind the Overlay

A direct request for the oil price embedded in the €94M management overlay, and whether more would follow in the second quarter, was answered by describing the overlay as a scenario adjustment across multiple macro variables rather than a single-input calculation, with a commitment to revert to the standard process next quarter.

Q: "on cost of risk, the EUR 94 million overlay, what oil price do you assume there? And could we see more coming in Q2 considering how things are evolving literally as we speak?"
— Giulia Miotto, Morgan Stanley

A: "the primary purpose of the overlay, which we built was indeed to adjust the quarter end macroeconomic scenarios ... to reflect the potential deterioration linked to the ongoing escalation in the Middle East. And let's say, from the coming quarter -- but let's say, consider a wider set of assumptions and macroeconomic variables than the pure oil price."
— Andrea Cesaroni, Head of Risk

Assessment: A dodge on the specific number and a commitment on the process. The refusal to name an oil price is defensible because the overlay is genuinely multi-variable, but it also means there is no disclosed trigger for release. The commitment to revert to standard provisioning from the second quarter is the checkable half, and it converts the overlay from a discretionary buffer into a scheduled test of whether the macro actually deteriorated.

Whether the Mortgage Floor Release Accelerates Distributions

The question was whether €4bn of released risk-weighted assets from the expiring Dutch mortgage floor would translate into a larger buyback or simply be absorbed. Management routed it through the standing framework rather than treating it as a discrete event.

Q: "on capital. Just wanted to get your thoughts around like the change on the mortgage floor in terms of the impact that you have on your CET1 ratio and your distribution policy. You want to run around 13%. So seeing a bit of a positive impact, would that change how much you distribute in terms of buybacks?"
— Delphine Lee, JPMorgan

A: "we are looking at a target of around 13%. We use our capital for growth and for normal distribution. And if there is any structural amount over that around 13% that we have in capital, then we'll pay it back to shareholders. And so we'll treat it any -- in the same way as we normally do."
— Steven van Rijswijk, Chief Executive Officer

Assessment: The answer is a restatement of policy, which is the point: nothing about a regulatory release changes the framework, and the 15bp lands inside the October semi-annual review rather than triggering an announcement. What the exchange confirms is that distribution capacity is now assessed against a single ratio test, so every basis point of risk-weighted asset relief is directly convertible into buyback. It also confirms there is no buffer above the target being held back.

Whether First-Quarter Cost Performance Puts the Full-Year Guide in Play

An observation that the cost line looked ahead of the full-year guidance invited management to upgrade it. They declined, and reframed the answer around reinvesting efficiency gains rather than harvesting them.

Q: "first on cost. It seems like Q1 good cost control and you are a bit ahead of your full year guidance. Just maybe you can comment a bit more on that, whether it was FX and how the benefits of the operations restructuring should help in the rest of the year?"
— Benjamin Goy, Deutsche Bank

A: "the more we're able to use -- to have efficiencies coming from our scalability, both from the digitalization and our scalable tech and ops, that we can then reuse to get better customer experience by making investments into broaden our products ... But the outlook remains the same at this point in time."
— Steven van Rijswijk, Chief Executive Officer

Assessment: The question was about arithmetic and the answer was about strategy. Neither the foreign exchange contribution nor the restructuring savings phasing was quantified, and the guide was held. Read charitably, management is preserving optionality to spend an efficiency windfall on growth, which is the right instinct for a bank with a fee-diversification thesis. Read sceptically, holding a guide that now implies a visible cost ramp while declining to explain the ramp is the one soft spot in an otherwise numerate call. Grade it at the half-year.

Whether Wholesale Exposure Is Vulnerable to Sustained Energy Prices

A follow-up distinguishing a temporary uncertainty shock from a sustained energy-price regime drew the most specific sector-level risk disclosure of the call.

Q: "But what about if we have a more sustained higher energy prices, lower consumption and maybe higher inflation on your wholesale lending. If we see something more structural rather than the reverse uncertainty, which areas you see and what could be impact on your lending?"
— Tarik El Mejjad, Bank of America

A: "Sectors in Wholesale Banking that could be affected are sectors that are, one, linked to the oil price, i.e., that has the oil price and energy price is quite an input factor on the cost base. You could think about the chemical sector or fertilizers or construction or transport and logistics, those are sectors that are typically impacted."
— Steven van Rijswijk, Chief Executive Officer

Assessment: A named list of exposed sectors with no exposure numbers attached. The secondary point about Asian supply-chain dependence on the Strait of Hormuz is the more interesting one because it describes a transmission channel that would not show up in a direct-exposure screen. The honest summary is that management can describe the shape of the risk and has not sized it, which is normal one quarter into a conflict and should be pressed next quarter if energy prices hold.

Whether a 52% Cost/Income Target Is Ambitious Enough

The most direct challenge of the call put ING's implied 2027 cost/income target against a peer group of European banks targeting lower ratios, and asked what prevents ING from doing the same. The answer was that the efficiency gains are being deliberately spent on revenue diversification.

Q: "do you think the cost income target of around 52% in 2027, just based on your revenue and cost targets, is ambitious enough, given there are 23 other European banks targeting a lower cost income between 2026 to 2028. I'm just trying to understand the main pillars stopping ING from getting to a lower cost income than 52%."
— Namita Samtani

A: "what we will largely save in terms of our operational efficiencies, we are investing in broadening and deepening our client relationships. That is helping in the end, that's what we're driving towards the ROE. ... And implicitly, that will then also have a cost-income decrease as a consequence. But the main driver is consistent RoTE at scale."
— Steven van Rijswijk, Chief Executive Officer

Assessment: The right answer, and it is also the answer that makes the cost guide unfalsifiable in the short run. Subordinating cost/income to return on tangible equity is defensible for a bank whose problem is income concentration rather than cost bloat, and 80% of revenue being interest-linked is a real problem worth spending money on. But it means the only enforceable commitment is the return target, so that is the one to grade. If return on tangible equity clears 15% in 2027, the cost answer was correct regardless of where the ratio lands.

What They're NOT Saying

1. The all other income reduction was never quantified against a prior number. The commercial net interest income raise came with a figure, roughly €200M at the midpoint, repeated and confirmed under questioning. The offsetting reduction to all other income came with a new range and no prior range to compare it to. Since total income guidance held at around €24bn, the two moves have to be roughly equal, but management let the positive number carry the headline and left the negative one to be inferred.
2. The Russia disposal sits outside every guided figure. The outlook footnote excludes the previously announced sale of the Russian business, where the expected post-tax P&L impact is around €0.8bn. That is more than half a quarter's profit, it is not in the 2026 or 2027 numbers, and it was not mentioned once on the call by management or by any analyst. No expected completion date was given.
3. The fee reclassification was disclosed in a footnote and not sized. From this quarter, daily banking fee income benefits from a structural shift of revenue out of other income, and there were also reclassifications between product categories. Daily banking fees rose €27M sequentially and €49M year on year. How much of that is the reclassification is not stated, which matters because fee growth is the single most-cited proof point of the diversification thesis.
4. Retail Belgium's credit deterioration drew no explanation and no question. Risk costs at 36bp against 16bp in both comparable quarters, on a business with a 2.4% return on allocated equity and an 81.3% cost/income ratio, described in one sentence as "primarily related to business lending." Restructuring provisions were taken this quarter for headcount reduction in the same business. Nobody asked.
5. The implied second-half cost ramp was not addressed. First-quarter operating expenses excluding incidentals were flat year on year, while the full-year guide requires 2.5% to 4.2% growth against 2025. Asked directly whether the quarter put them ahead of the guide, management answered with the scalability narrative and held the number. The phasing was never explained.
6. No number for the artificial intelligence contribution. More than 90% of AI pilots moved into production, more than 75% of customer chats resolved without human support, agentic mortgages live in the Netherlands, conversational banking about to roll out globally. No cost saving, no revenue uplift, no capitalised spend, no full-time-equivalent reduction attributed. The disclosed evidence is a 0.6% headcount decline over twelve months, which is not obviously an AI number.
7. The Stage 3 coverage ratio fell and was not discussed. Coverage declined to 33.5% from 34.4%, and the stock of Stage 3 provisions rose only €49M against Stage 3 outstandings rising €476M. The improvement in the Stage 3 ratio to 1.5% was highlighted; the decline in how much of it is provided against was not.
8. The size of the Stage 3 repayment that rescued Wholesale credit was not given. A single loan repayment more than offset a €49M overlay and took the division from 38bp to 12bp. The euro amount was never disclosed, which makes the underlying wholesale run-rate impossible to reconstruct from the outside.

Market Reaction

• Pre-print setup: The Amsterdam line closed at €24.75 on 30 April against €23.88 on 29 April. Entering the print the shares were down 0.5% year to date while the AEX was up 4.8%, but up 40.4% over trailing twelve months and up 8.0% over the trailing thirty days, sitting €2.25 below a 52-week closing high of €26.13 set inside the first quarter itself. The New York depositary receipt closed at $27.65 on 29 April, down 1.3% year to date against the S&P 500 up 4.2%, and up 42.7% over twelve months.
• Reaction session, 30 April, before-the-open print: Amsterdam opened at €24.12, a 1.0% gap, traded €23.89 to €24.91, and closed at €24.75, up 3.70% or €0.88. The depositary receipt opened at $28.41, a 2.7% gap, traded $28.38 to $29.14, and closed at $28.93, up 4.6% or $1.28.
• Volume: Amsterdam traded 13.1M shares against a 9.2M thirty-day average, or 1.4x. The depositary receipt traded 3.0M against a 3.2M average, or 1.0x. The conviction was in the home market, not in New York.
• Relative move: The AEX rose 1.70% and the Euro Stoxx 50 rose 1.12%, so the Amsterdam line beat its home index by roughly two points. Against European banking peers reporting into the same session the outperformance was wider: of seven large European banks sampled, only two closed higher and five closed flat to lower, in a range from +1.79% to −1.41%. This was an idiosyncratic move, not a sector bid.
• Intraday path: The depositary receipt was up about 2% in premarket trading before the US open and finished up 4.6%, so more than half the eventual gain accrued during American hours as the S&P 500 rose 1.0%.

The two listings disagree by about 90 basis points on the same results, and the reason is mechanical rather than interpretive. The depositary receipt is one ordinary share, and the euro strengthened against the dollar over the session, which translates straight into the New York line. The euro close of €24.75 multiplied by the prevailing rate reconciles to the $28.93 depositary close within a cent. The Amsterdam close is the cleaner read on how the print was received, and 3.70% against a 1.70% index is a solid endorsement rather than a re-rating.

What the market paid for is worth isolating, because a 6.4% profit-before-tax beat is not by itself a 3.7% day for a European bank. Three things landed together. The return on tangible equity print of 13.6% against a 12.1% poll is a 150 basis point beat on the metric the whole 2027 target is expressed in, one quarter into the year. The commercial net interest income guidance raise is the first upgrade to the recurring revenue line since the full-year 2025 results, and management sourced all of it to liability income rather than to a lending assumption anyone would need to underwrite. And the €1.0bn buyback landed exactly on the poll median, which removed the risk of disappointment on the one number the market was positioned for.

The setup mattered as much as the print. This was a stock down 0.5% year to date against an index up 4.8%, having given back roughly €2.25 from a high made inside the same quarter, after a twelve-month run of 40%. That is a name where the marginal holder had already taken something off, which is why a good quarter produced a clean 3.7% move on 1.4x volume rather than a squeeze or a fade. The market discounted the all other income miss, correctly, as the mark-to-market line that management had already told it to look through, and paid for the two lines it will still be capitalising in 2027.

Street Perspective

Debate: Is the Replication Tailwind a Rate Call or a Contract?

Bull view: The bull case circulating on the Street is that roughly 55% of the retail eurozone replicating portfolio carries an average remaining maturity between one and fifteen years, so the reinvestment of maturing tranches at higher yields is largely locked in regardless of where policy rates go from here. On this reading the liability margin path to 2028 is closer to an amortisation schedule than to a forecast, and the guide contains only a third of the illustrated uplift.

Bear view: The sceptical framing is that the replicating portfolio determines the gross investment return and says nothing about what the bank pays depositors. Deposit pricing is a competitive variable, the disclosed sensitivity is that every 10 basis points of pass-through costs about €0.4bn of commercial net interest income, and a single quarter without a savings campaign is exactly what produced the 5 basis point margin improvement everyone is now extrapolating.

Our take: Both halves are true and management said so. The gross replication income is contracted; the net margin is not. What decides this is that the bank guided to only about a third of the gross uplift, so the bear case is already substantially embedded in the number. If pass-through stays where it is, the guide is beaten. If competition intensifies, the guide is met. The asymmetry favours the bulls, and it favours them because management chose conservatism rather than because the risk is absent.

Debate: Was This a Beat or a Cost Line?

Bull view: The optimistic reading is that revenue was in line only because a non-economic hedge-ineffectiveness charge of €81M sat inside it, that the company's own volatile-items reconciliation puts profit before tax at €2,359M rather than €2,258M, and that the two recurring revenue lines both beat. On that basis this was a clean operating quarter that the accounting obscured.

Bear view: The bear camp points out that roughly two-thirds of the profit-before-tax beat came from below the revenue line, that €66M of it was regulatory costs the Street estimates rather than models, and that a €18M provision beat rested on a single unquantified loan repayment. Adjust for all three and the beat shrinks to something ordinary.

Our take: The bears have the better description of this quarter and the bulls have the better description of the business. Regulatory cost variance is not earnings power, and a Wholesale provision line rescued by one repayment is not a run-rate. But the two things that decide the multiple, commercial net interest income and fee income, both beat on their own, and the mix shift out of market-driven income is a real improvement in what those earnings are worth. Grade the quarter as ordinary and the composition as good.

Debate: Has Capital Return Become Structurally Harder?

Bull view: The bull argument is that the reserving change is presentational, the distribution policy is unchanged, €4.4bn of additional distributions were made over the trailing twelve months on top of the 50% payout, and the pipeline of risk-weighted asset relief is quantified at 30 to 35 basis points for 2026 with more to come from significant risk transfers in retail portfolios that have barely started.

Bear view: The bear argument is that a CET1 ratio of 13.0% against a target of around 13%, with a pro-forma of 12.9% after the announced distribution, leaves no structural excess at all, and that a buyback funded by regulatory model changes rather than by retained earnings is a finite resource. The Dutch mortgage floor expires once.

Our take: The bears are describing the mechanism correctly and drawing the wrong conclusion from it. Capital generation has not slowed; €6.4bn of net profit over twelve months contributed almost two percentage points of CET1, and the constraint is the target ratio, not the generation. What has changed is that the excess is now reserved up front rather than accumulated visibly, which makes the next distribution more predictable and the one after that less so. The relevant question for 2027 is whether risk-weighted asset growth stays below capital generation once wholesale lending normalises, and this quarter, with Wholesale customer lending up 3.3% and division risk-weighted assets flat, the answer was yes.

Debate: Does 1.44x Tangible Book Already Price the 2027 Target?

Bull view: The bull case is that 8.9x the 2027 consensus earnings line and 1.44x tangible book for a bank guiding to a return on tangible equity above 15% is cheap on any residual-income framework, and that the combined dividend and buyback yield near 7.6% pays the holder to wait for the arithmetic to work.

Bear view: The bear case is that a 40% twelve-month move has taken the price-to-book multiple from 1.06x to 1.40x, that consensus already sits at or above the guide on both 2026 and 2027 return on tangible equity, and that the marginal buyer is therefore underwriting execution rather than a re-rating.

Our take: The bears are right that the re-rating has happened and wrong that it exhausts the case. A bank sustainably earning 15% on tangible equity at a 10.5% cost of equity and 3% growth supports roughly 1.6x tangible book; at 1.44x the shares are not pricing the 2027 target, they are pricing something closer to 14%. The gap is modest, which is why this is an Outperform rather than a table-pounding call. The distribution yield is what makes the wait economic.

Model Framework & Valuation

This is an initiation, so the table below states the framework being adopted rather than a change to a prior model.

Valuation framework. At the €24.75 Amsterdam close the shares carry a market capitalisation of about €71bn on 2,876.6M shares. Shareholders' equity per share was €17.68 at 31 March, and deducting €1,552M of intangible assets gives tangible book of €17.14 per share. The shares therefore trade at 1.40x book and 1.44x tangible book. Trailing four-quarter earnings per share of €2.18 puts the multiple at 11.4x; the consensus 2026 line of €2.36 puts it at 10.5x and the 2027 line of €2.77 at 8.9x.

The residual-income arithmetic is the honest frame for a bank. A franchise sustainably earning 15% on tangible equity, growing tangible book at about 3%, and carrying a 10.5% cost of equity supports roughly 1.6x tangible book. Widening the cost of equity to 11% takes that to 1.5x; tightening it to 10% takes it to 1.7x. Applying 1.50x to 1.70x to an estimated year-end 2026 tangible book of about €17.80 per share, after retained earnings and the tangible-book dilution of buying stock back above book, gives a range of €26.70 to €30.30. We anchor at about €28.50, roughly 1.60x, which implies about 15% price appreciation from the €24.75 close.

Distribution. The final cash dividend for 2025 of €0.736 per share was approved on 14 April and paid on 24 April. Consensus has 2026 dividends per share at €1.18, a 4.8% yield at €24.75, and total structural excess-capital distribution at €2.0bn, worth a further 2.8% of market capitalisation. Combined, roughly 7.6% of market value is returned annually while the CET1 ratio holds at target. Adding the dividend to the price framework gives a twelve-month total return in the region of 20%, which is the basis for the rating.

What would change our view. Deposit competition intensifying to the point where the liability margin falls below the 100 to 110 basis point range rather than sitting in its mid-part. Retail risk costs continuing to climb from 21bp without the Wholesale offset repeating. Operating expenses running at the top of the guide alongside revenue at the bottom, which would break the positive-jaws commitment the 2027 return target depends on. Or a second consecutive quarter in which all other income undershoots, which would turn a mix improvement into a level problem.

Thesis Scorecard

This is our initiation, so the table below establishes the pillars we will carry forward and grade each quarter rather than scoring a standing thesis. The status tags reflect this quarter's evidence only.

Overall: Thesis established at a constructive starting point. Four bull pillars confirmed on this quarter's evidence, two bear points already emerging, one materialising and one contained. The structure is legible: one mechanism, deposit replication converting into liability margin, drives most of the recurring upside; one line, all other income, carries most of the quarterly noise; and one constraint, a capital ratio pinned at target, governs how much of the earnings comes back to shareholders.

Action: Buy. A 13.6% return on tangible equity rising toward a guided 15% by 2027, at 1.44x tangible book and 8.9x the 2027 consensus earnings line, with roughly 7.6% of market value returned annually, is an adequate rather than spectacular entry, and the composition of the earnings improved this quarter without the price of the earnings changing. Size the position for the mark-to-market line rather than for the credit line, and grade three things at the half: the liability margin against the 100 to 110 basis point range, Retail risk costs against this quarter's 21bp, and operating expenses against the implied ramp.