A 16% ROTE, a €1.9bn distribution and a quiet fade of the old lending-margin promise — all in the same quarter
INGA.AS · Earnings Call · 2026-10-29
The claim, and the evidence behind it
ING spent this call arguing that its profitability improvement is not a rate cycle accident. Steven van Rijswijk walked through the ROTE bridge and was unusually explicit: “more than 80% of that improvement is supported by a number of strategic drivers” — Steven van Rijswijk, Chief Executive Officer · 2026-10-29. The mechanics check out. Income grew 6.6% over twelve months while costs grew about 2%, giving positive jaws north of 4 percentage points, EPS up 17% year-on-year, and a 16% return on tangible equity. Mobile primary customers — the mobile primary customer engine — rose 259,000 in the quarter and more than 1.1 million over the year, ahead of the Capital Markets Day target. Average fees per private individual client are up 14%.
What is genuinely new is that ING now describes its own growth as product-built-once, consumed-in-many-markets. The global investment platform, live in the Netherlands, Belgium, Italy and Romania and launched in Romania only at the end of September, is the proof point management keeps returning to — a single build, four markets, faster product development. That is a different claim from "we are digital, therefore we are cheap." It is a claim about marginal cost of growth.
The bill for that growth is now visible
The single loudest new theme in this quarter's language is restructuring cost, which arrives at the top of the company's keyword stack after several quarters in which the story was growth without extra resources. The Wholesale Bank becomes Corporate and Institutional Banking, roughly 400 roles go globally, and a provision of about €175 million lands in Q4 2026 for an annual cost reduction of about €75 million from 2028.
That arithmetic drew fire twice. Giulia Miotto called the timing "quite a surprise"; Shrey Srivastava went straight at the ratio, asking why the payoff looks thin against comparable bank restructurings. Steven's answer was that the comparison doesn't hold — different roles, different scope — and that the savings "trickle in greatly over 2027." But the pattern matters: earlier in the call he framed year-to-date provisions of about €130 million as buying roughly €100 million of annualised savings, and referenced 2025's ~€300 million of provisions producing ~€170 million excluding wholesale. A quarter that costs €175 million to save €75 million is a worse exchange rate than the bank's own recent history — and it lands precisely when the CEO is asking the market to believe in scalability.
He is candid about the trade:
we create a bit of short-term pain for long-term benefit
This is not new behaviour. Tanate Phutrakul set the policy two CFOs ago: “we take provision when we have a clear business case, it is concise and that it can be achieved over 12 to 18 months” — Tanate Phutrakul, Chief Financial Officer · 2025-10-30. The question investors should hold onto is whether the Wholesale Bank action is the tail of that programme or the start of a new, less efficient one ahead of the June 2027 strategy update.
Two quiet upgrades and one quiet disappearance
The numbers nobody headlined are arguably the most important. Liability margin came in at 108bps for the first nine months with ~110bps guided for the full year — against a January call where the guidance was “on the lower end of the 100 to 110 basis points” — Tanate Phutrakul, Chief Financial Officer · 2026-01-29, and a July call where Ida Lerner pointed to above 110bps for 2027 and 2028. The replicating portfolio is doing the work, and Ida noted a 6 basis points improvement in liability margin in the quarter alone. Commercial NII rose €161 million quarter-on-quarter and 13% year-on-year, and on the €16.8–17.0 billion range she steered to the top: “we are more looking at the upper end of that range” — Ida Lerner, Chief Financial Officer · 2026-10-29.
Meanwhile, lending margin sits at 123bps, down a basis point on mix — more mortgages, more investment-grade wholesale — and is guided "broadly stable." The old multi-year aspiration of 125–130bps, repeated on calls through 2025 and into 2026, has effectively been retired without a formal farewell. The word has dropped out of the top-30 keyword set entirely this quarter. Growing at 6% annualised on both sides of the balance sheet by lending into lower-risk-density assets is the right commercial decision; it is simply a lower-margin business than the one the 2025 guidance described.
Capital, and the market's own blind spot
CET1 rose 30bps to 13.3% while ING announced €1.9 billion of additional distribution — a €1.65 billion buyback starting today plus €250 million paid in cash in January — with 100% of six months' net profit reserved outside CET1. The enabling machinery is risk weighted asset velocity: €3 billion of SRT relief in the quarter across wholesale and, for the first time, retail Dutch business banking; 16bps year-to-date and 28bps cumulative, which Ida concedes is significantly lower than peers. Capital efficiency is now the bridge between growth and distribution, replacing the 12.8–13.0% prudence of the tariff-shock era.
Two risk lines deserve scrutiny rather than applause. Risk cost of €403 million included €111 million of IFRS model update, and Stage 3 additions of €239 million were mostly collective, tied to consumer-lending growth. And on the regulatory model question — the same IRB model theme that sits atop the entire market's keyword list this quarter, with JYSK.CO flagging conservative IRB risk weights in its own release — ING's answer was that its multiyear model review is complete and “we do not expect any material adverse capital or risk-weighted asset impact” — Unknown Analyst, Executive or Senior Management · 2026-10-29. That is a company-specific all-clear in a quarter where the market is visibly pricing model and capital risk across European banks.
Where ING is merely consensus is deposits. "Competition remains strong" is now boilerplate across the sector — BBVA on corporate deposit pricing, HSBC on the shift into time deposits, Santander on declining cost of deposits, all in the last five days. ING's deposit cost narrative adds nothing distinctive; its differentiated claim is fee alpha and capital velocity, not deposit pricing power. Worth noting too that the market tape's financial-adjacent losers — fee-paying AUM, management fee growth, direct lending platforms — are alternative-manager exposures ING simply does not carry.
The open question
The stock of unfinished business is now concentrated in one date: 14 June 2027, when ING presents 2030 targets. Guidance for 2027 has been raised to total income at least €1 billion above the original target, costs ~€300 million below, and ROTE above 16%. The uncomfortable comparison, raised on the July call when European peers were approaching ROTE near 20%, has not gone away. Steven's own hedge on loan growth — “higher interest rates and more uncertainty typically has a dampening effect on loan growth. We haven't seen that yet” — Steven van Rijswijk, Chief Executive Officer · 2026-10-29 — is the tell. This quarter's story is that the machine works. Next year's story is whether a bank buying €75 million of savings for €175 million can keep calling itself scalable without buying more.