A 35% Pre-Tax Beat, $14bn Out the Door in the Americas, and a 2026 Guide Two Points Below the Target Everyone Was Anchoring On

Results vs. Consensus

Q4 2025 Scorecard

The Sequential Bridge That Explains the Quarter

The single most useful calculation in this release is the one that puts the third and fourth quarters on the same basis. It requires no estimates and no judgment, only the filing's own underlying figures and the litigation release management disclosed in October.

Underlying revenues were identical to the dollar across the two quarters. Adjusted underlying operating expenses differ by $6m. The entire $719m sequential decline in underlying pre-tax profit is the $668m litigation release plus $57m of higher credit loss expense. In October we wrote that the third quarter's 69.7% underlying cost/income ratio "becomes roughly 75%" once the release is added back. The fourth quarter printed 75.2% clean. The adjustment was right, and it means the run-rate profitability of this bank has been stable at roughly $2.9bn of underlying quarterly pre-tax profit for two quarters, not rising and not falling.

Year-Over-Year Comparisons

Net interest income growing 18.2% year over year with the Swiss policy rate at zero is the most striking line in the table and the least discussed. It reflects higher average loan and deposit volumes, a better deposit mix, and the beginning of the benefit from replacing distressed-spread Credit Suisse funding. Fee income grew 9.5% and now runs at $7.2bn a quarter. The negative $412m other income line is a single identifiable item, the $457m net loss on the debt repurchase, and it is the reason the reported revenue line grew only 4.4% while underlying revenue grew 10.3%.

Quarter-Over-Quarter Comparisons

Every negative in this table has a named, disclosed cause. The 4.5% decline in CET1 capital is $4.1bn of shareholder-return accruals, $3bn for the 2026 buyback and $1.1bn for the full-year dividend, plus roughly $0.5bn from the liability management exercise. The 7.8% rise in underlying operating expenses is the absence of last quarter's litigation release. The tax rate swing is the reversal of a benefit that was always temporary. None of these is an operating deterioration, and all of them were signposted in October.

Full-Year 2025

The full-year figures above are the income-statement totals, and each reconciles to the sum of the four reported quarters. Net profit of $7,767m is the nine-month $6,568m plus the fourth quarter's $1,199m; full-year diluted EPS of $2.36 is $1.99 plus $0.37; operating expenses of $40,197m are $29,911m plus $10,286m. The one line that does not describe operating performance is the tax rate. An 11.9% full-year effective rate against 24.6% a year earlier reflects the litigation reserve releases in Non-core and Legacy plus tax planning around the legal-entity structure, and management has guided it to roughly double in 2026.

Quality of Beat

• Revenue: The best line of the quarter. Underlying revenues grew 10.3% year over year with net interest income up 18.2% and fee income up 9.5%, in a quarter where the Swiss policy rate was zero and the dollar was soft. The growth is organic. Reported revenue growth of 4.4% understates it by the $457m debt-repurchase loss and a $636m decline in purchase-price-allocation accretion, both of which are integration mechanics rather than trading performance.
• Margins: Real and now clean. The 75.2% underlying cost/income ratio is the first ratio UBS has printed in several quarters that requires no adjustment, and it is 670 basis points better than a year ago. Nine percentage points of positive jaws in the quarter is the operating leverage the integration was supposed to produce. The offsetting fact is that 75.2% is still five points away from the sub-70% exit-rate target with four quarters left to close it.
• EPS: Uninformative, and the estimate dispersion proves it. Published fourth-quarter EPS estimates spanned $0.25 to $0.67 around a $0.54 median, a range so wide that the $0.37 print can be described as a beat or a miss depending on which desk's model you use. The dispersion is a direct consequence of analysts disagreeing on the debt-repurchase loss, the integration expense and the year-end tax rate. This is why the scorecard above is anchored on pre-tax and net profit, where the estimates were consistent and the beat was unambiguous.
• Flows: The genuine miss, and the reason the stock fell. Group net new assets of $8.5bn is a 0.7% annualized rate on $4.8tn of invested assets. The composition is worse than the total: $23bn of inflows across EMEA, APAC and Switzerland against $14bn of outflows in the Americas. For the full year, $101bn of net new assets is 2.4% growth, below the roughly 3% a franchise of this quality should sustain and well below the greater-than-$125bn management has now guided for 2026.
• Capital: Lower by design, and slightly lower than expected. The CET1 ratio of 14.4% is where management said it would be, "closer to our targets of around 14%," after absorbing $4.1bn of shareholder-return accruals. It is nonetheless a 40 basis point sequential decline in a quarter when the market wanted evidence of capital headroom ahead of the ordinance.

Segment Performance

Underlying Pre-Tax Profit by Division

Every division grew underlying pre-tax profit year over year and every division declined sequentially. The year-over-year column is the signal; the sequential column is mostly the litigation release, which landed disproportionately in Non-core and Legacy ($440m of the $668m) and in the divisions' litigation lines.

Full-Year 2025 Underlying Pre-Tax Profit by Division

Global Wealth Management

GWM delivered $1,558m of underlying pre-tax profit, up 35.8%, on underlying revenues of $6,580m (+10.7%). Invested assets reached $4.8tn. Recurring net fee income rose 9% to $3.6bn on fee-generating assets of $2.1tn; transaction-based revenues were $1.2bn, up 20%, on structured products and cash equities; net interest income was $1.7bn, up 3% year over year and 4% sequentially. For the full year, pre-tax profit excluding litigation was $6.1bn, up 23%, with the cost/income ratio improving more than three points to 75.6%. All four regions grew, each contributing around $1.5bn excluding litigation. The Americas grew fourth-quarter pre-tax profit 32% at a 13% margin, capping a year of 34% profit growth; EMEA grew 27% in the quarter and 19% for the year; APAC grew 24% and 30%; Switzerland fell 4% in the quarter on rate headwinds but rose 2% for the year.

"Net new assets were $8.5 billion, with $23 billion of inflows across EMEA, APAC and Switzerland, partially offset by outflows of $14 billion in the Americas, primarily reflecting net recruiting-related impacts. For the full year 2025, we generated net new assets of $101 billion, representing 2.4% growth. We delivered this while absorbing the expected temporary flow headwinds from strategic actions taken to support higher pretax margins and enhance our return on equity." — Todd Tuckner, Group CFO

Net new fee-generating assets were $9bn, unchanged from the third quarter, with APAC at a 10% annualized rate. Mandate penetration rose for a fourth consecutive quarter, with the My Way discretionary solution nearly doubling year over year to more than $30bn of invested assets. Net new loans were $5bn on Lombard and securities-based lending, and Americas loan balances grew for a seventh consecutive quarter.

Assessment: The profit performance is excellent and the flow performance is the worst of the year. Both facts are consequences of the same decision. Management chose margin over assets in the US, the margin arrived, and the assets left faster in the fourth quarter ($14bn) than in the third ($9bn). What changed this quarter is that the trade finally has a timetable: outflows persist through the first half of 2026, then taper materially, with the Americas expected to be a positive contributor for the full year. That is a falsifiable claim we can grade in July.

Personal & Corporate Banking

P&C produced CHF 543m of pre-tax profit, down 5%, as net interest income fell 10% year over year and 2% sequentially. Non-net-interest revenues fell 3%, with growth in Personal Banking more than offset by lower Corporate and Institutional client activity. Operating expenses of CHF 1.1bn were down 1%. Credit loss expense was CHF 80m in the quarter, precisely the figure guided in October, and CHF 277m for the full year. In dollar underlying terms the division earned $678m, up 5.0% year over year, and $2,857m for the full year, down 4.8% and the only division to decline.

The forward guidance carried the quarter's one explicit target slip. Management said the sub-50% underlying cost/income ratio by end-2026 is likely delayed by Swiss franc rate headwinds, while committing to a reported cost/income ratio of around 48% by 2028 even if rates remain at zero, and to double-digit pre-tax profit growth in 2026. Full-year 2026 net interest income is modelled to rise by a mid-single-digit percentage in dollars, helped by currency translation, the liability management exercise and loan growth.

Assessment: A division running to stand still, and honest about it. Slipping the 2026 cost/income target while holding the 2028 one is the right call, and the willingness to say so at the same event as the 2028 ambitions is a credibility marker. The awkward arithmetic remains: getting from the current ratio to around 48% by 2028 is a very large move, and management's own answer on how attributes it to non-interest revenue growth plus the cost synergies that arrive once the Credit Suisse platform is retired. The second half of that is bankable. The first half is not yet evidenced.

Investment Bank

Pre-tax profit of $703m rose 55.5% on underlying revenues of $2,885m, up 13.3%, capping the strongest top-line year in the division's history at $11.8bn, up 18%, achieved with essentially no incremental risk-weighted assets and a full-year return on attributed equity of 15%. Global Markets revenues rose 17% to $2.2bn, the strongest fourth quarter on record globally and in every region, with equities up 9% against a hard comparative on prime brokerage, record cash equities market share and equity derivatives, and foreign exchange, rates and credit up 46% on the back of FX and precious metals. Banking revenues rose 2% to $687m, with advisory up 2%, capital markets up 1%, equity capital markets up 68% and leveraged capital markets lower on softer sponsor activity. Operating expenses rose 6%.

The strategic disclosures were as significant as the numbers. Management is no longer guiding to a risk-weighted-asset target, having lifted group revenues over RWAs to the roughly 10% ambition from under 8% two years ago, and reframed the division's discipline as a 15% reported return on attributed equity through the cycle while consuming no more than 25% of group RWAs. The ambition to double Global Banking revenues versus the 2022 baseline was restated on an annualized basis by end-2026 rather than as a full-year 2026 outcome.

Assessment: The best division in the group two quarters running, and the proof that the post-Credit-Suisse investment bank design works. Dropping the RWA target is a meaningful signal: it converts the IB from a resource to be constrained into a franchise to be grown, which is a change of posture rather than of policy. The restated banking target is a soft miss dressed as a clarification, and management said as much when it acknowledged that a slow first half of 2025 in banking "effectively delayed us a bit."

Asset Management

Pre-tax profit of $268m rose 19.6%, or 41% excluding the profit and loss from disposals, which included a $29m loss on the sale of the O'Connor business. Revenues rose 4% with net management fees up 11% on higher assets under management, and operating expenses fell 2%, producing a 66% cost/income ratio that meets the 2026 exit-rate ambition a full year early. Invested assets reached $2.1tn. Net new money was $8bn in the quarter and $30bn for the full year, a 1.7% growth rate held back by product rationalization as the Credit Suisse integration completed. Unified Global Alternatives took $9bn of net new client commitments including $8bn from GWM clients, with funded invested assets of $330bn, which management characterized as a top-five limited-partner position.

Assessment: Quietly the most improved division of 2025 and the first to reach its target. The 2028 ambition of a roughly 65% reported cost/income ratio and around 3% net new money through the cycle is modest by design, and modest targets from a division that just beat its last one a year early are the kind we prefer. The 1.7% full-year organic growth rate is the honest weakness, and management attributed it to deliberate product rationalization rather than to competitive loss.

Non-core and Legacy

NCL produced a $224m underlying pre-tax loss, against a $102m profit in the third quarter, which was itself a function of the $440m litigation release. Revenues were negative $10m as $86m of funding costs were partly offset by position marks and disposals. Operating expenses fell nearly 60% year over year. Risk-weighted assets closed at $29bn, or $5bn excluding operational risk, down $2bn sequentially, and the leverage ratio denominator fell $6bn, or 25%, to $19bn. Since formation the division has cut risk-weighted assets by two-thirds, freed nearly $8bn of capital and reduced operating costs by roughly 80%.

The wind-down now has an explicit terminal path: exit 2026 with annualized operating expenses excluding litigation of around $500m (roughly 40% of 2025 levels) and annualized net funding costs below $200m, with the resulting pre-tax loss run-rate halving again by 2028 and tapering to immaterial levels thereafter. Operational risk RWAs are the stubborn residual at around $24bn, with roughly 10% rolling off through 2030 and the remainder substantially between 2031 and 2035.

Assessment: The balance-sheet job is essentially finished and the cost job is now quantified, which is progress. The operational-risk RWA disclosure is the least comfortable item in the release and deserved more attention than it got: $24bn of risk-weighted assets that do not begin to roll off in earnest until 2031 is a decade-long drag on group returns that no amount of execution accelerates, because the runoff schedule reflects regulators' instructions on the ten-year loss history.

Key Group KPIs

Key Topics & Management Commentary

Overall Management Tone: Substantially more forthcoming than in October, and more willing to be held to specifics. Where the third-quarter call deferred every 2026 quantity, this one produced a full target architecture out to 2028, a division-by-division bridge, a dated timetable for the US flow inflection, and an explicit admission that one target has slipped. The register on the Swiss capital question remained procedural and unemotional, though the CEO's defence of AT1 as an instrument class was the one moment of visible conviction on the call.

1. Two Quarters, One Run-Rate

The most useful thing management did was disclose the fourth quarter's distortions with the same prominence it gave the third quarter's. Reported revenues carried $54m of net negative adjustments, comprising the $457m loss on the November debt buyback offset by other merger-related purchase-price-allocation accretion. Integration expenses were $1.1bn. The effective tax rate was 29%.

Put those alongside October's $668m litigation release and the picture resolves. Underlying revenues were $12,199m in each of the two quarters. Underlying operating expenses were $9,169m this quarter against an adjusted $9,175m last quarter. Underlying pre-tax profit before the litigation release was roughly $2.9bn in both.

Assessment: The bank's run-rate earnings power has been flat for two quarters at a level that produces an underlying return on CET1 capital just under 12%. That is the number from which the walk to 15% has to happen, and it is why the full-year 2026 guide of around 13% is credible while the exit-rate 15% requires the second half of 2026 to do most of the work.

2. The November Liability Management Exercise

UBS repurchased $8.5bn of legacy Credit Suisse debt instruments issued at distressed spreads before the acquisition, booking a $457m net loss in reported revenues and reducing CET1 capital by roughly $0.5bn. The economics were laid out clearly.

"Buying back this expensive legacy debt early and replacing it with low-cost funding is not only NPV accretive, but will also benefit the net interest income of GWM and P&C in the coming years and reduce the net funding drag in NCL." — Todd Tuckner, Group CFO

Pressed for the size, management quantified the benefit at roughly $100m a year net of purchase-price-allocation effects across each of the next three years, split across Wealth Management, Personal & Corporate Banking and Non-core and Legacy. Cumulative funding cost savings since the integration began now stand at around $1.2bn, exceeding the original $1bn target for 2026.

Assessment: A $457m one-time charge for roughly $300m of net interest income over three years plus a permanently better liability profile is a defensible trade, though not an obviously spectacular one on those numbers alone. The stronger argument is the one management did not lean on: retiring distressed-spread paper removes a visible reminder of the acquisition from the funding stack and is part of why total loss-absorbing capacity fell from $199bn to $187bn without any deterioration in resilience.

3. The 2026 Guide Versus the 2026 Target

This is the disclosure that reset the stock, and it is easy to misread. UBS gave two different 2026 numbers. The exit-rate targets, unchanged, are an underlying return on CET1 capital of around 15% and an underlying cost/income ratio below 70%, measured on an annualized run-rate basis at the end of the year. The full-year 2026 expectation, disclosed for the first time, is an underlying return on CET1 capital of approximately 13% and a cost/income ratio of around 73%, with the effective tax rate normalizing to around 23%.

Management's explanation for the gap is the phasing of the cost programme: all of 2026 is required to deliver the remaining integration milestones, with net savings building progressively and weighted toward the second half, because the most complex decommissioning work cannot begin until the Swiss client migrations finish at the end of the first quarter and then ramps from midyear.

Assessment: The logic is sound and the disclosure is honest, but it lands as a downgrade because consensus had been anchoring 2026 estimates closer to the exit-rate target than to the full-year path. The tax normalization compounds it: moving from an 11.9% to a roughly 23% effective rate is worth over a point of return on its own. Anyone who modelled 2026 by applying the 15% target to a full year was always going to be disappointed, and this call is where that happened.

4. The 2028 Ambitions

The new architecture extends two years beyond the integration: a reported return on CET1 capital of around 18% under the current capital framework and a reported cost/income ratio of around 67% by 2028. Beneath the group level, GWM targets a reported cost/income ratio of around 68% and more than $200bn of annual net new assets; P&C a reported cost/income ratio of around 48% even at zero rates; Asset Management around 65% with about 3% net new money through the cycle; and the Investment Bank a 15% reported return on attributed equity through the cycle at no more than 25% of group risk-weighted assets.

The word "reported" is doing real work here. Every 2026 target is stated on an underlying basis; every 2028 ambition is stated on a reported basis. That is only possible because integration expense is assumed to be gone, and it is the clearest statement management has made about what this bank looks like once the Credit Suisse chapter closes.

Assessment: An 18% reported return on CET1 capital would be a genuinely high-quality outcome and above pre-acquisition profitability. Two caveats deserve equal weight. First, the qualifier "under the current capital framework" means the ambition is stated on a denominator the Swiss authorities are actively rewriting. Second, management itself declined to defend the distinction between a target and an ambition when asked, which is a fair signal of how much precision to attach to a 2028 number.

5. Net New Assets: The Number That Moved the Stock

Group net new assets of $8.5bn is the weakest quarter of the year and follows the strongest. The $23bn of inflows across EMEA, APAC and Switzerland is a perfectly respectable underlying rate; the $14bn of Americas outflows is what turns it into a headline. Full-year net new assets of $101bn represents 2.4% growth.

What is new is the shape of the forward commitment: more than $125bn of net new assets in 2026, and more than $200bn per annum by 2028. Reaching $125bn in 2026 from $101bn in 2025 requires the Americas to swing from a full-year drag to a positive contributor while the other regions hold their run-rate.

Assessment: Guiding to $125bn in 2026 immediately after printing $8.5bn in a quarter is a bold sequence, and management has now attached its credibility to the US inflection happening in the second half. That is the single most gradeable commitment on the call. If Americas flows are still negative when the second-quarter results land in July, the 2026 net-new-asset guide fails and the 2028 ambition loses its foundation, since management conceded the US "has to contribute more" to reach $200bn.

6. The Cost Ambition Goes Up, Not Down

Cumulative gross run-rate cost saves reached $10.7bn at year-end, including $3.2bn delivered during 2025 and $700m in the fourth quarter. Against the 2022 baseline the cost base is down around 25% excluding currency, litigation and production-linked variable compensation, and around 12% overall. The planning process then produced something the market did not expect.

"In the planning process for 2026, we identified an additional $500 million in cost synergies. This allow us to increase our gross cost savings ambition to $13.5 billion. I'm particularly pleased that we will be able to produce these synergies at a very efficient cost to achieve multiple of 1.1." — Sergio Ermotti, Group CEO

The residual $2.8bn for 2026 breaks down as roughly 40% technology infrastructure and run costs, 40% workforce capacity and the remainder third-party spend and real estate, consistent with the split given in October. Integration expenses of $13bn have been incurred to date with around $2bn more expected in 2026, taking the final cumulative total to around $15bn at historical foreign exchange rates. Management guided net saves to about 75% of the $13.5bn gross figure.

Assessment: Raising a cost-synergy ambition three years into a programme, while holding the cost-to-achieve multiple at 1.1x, is the strongest evidence in the release that this integration is being run well. It is also the pillar of our thesis that has now been confirmed four quarters running, and the one we have the least doubt about.

7. Capital Upstreaming Accelerates

The parent bank's standalone CET1 capital ratio rose to 14.2% from 13.3% on $9bn of capital upstreamed from subsidiaries, roughly $4bn from Credit Suisse International in the UK and roughly $3bn from the US intermediate holding company. Because foreign subsidiary investments attract a 400% risk weight at the parent, the $6.5bn reduction in the investment-in-subsidiaries account cut parent-bank risk-weighted assets by $26bn. A further $3bn is expected predominantly from UK subsidiaries, and around $2bn more from the US IHC by 2028.

Management also reset its parent-bank operating guidance, saying it now expects UBS AG to run at a standalone CET1 ratio of around 14% for the foreseeable future, above the previously stated 12.5% to 13% range, because of foreign-exchange-driven headwinds on leverage ratios. Group equity double leverage closed the year at 104%, down five points from the second quarter, against a target near 100%.

Assessment: Materially more important than its billing on the call. This directly shrinks the foreign-subsidiary capital shortfall that has framed the entire Swiss capital debate, and management confirmed as much when asked. The counter-argument, which one questioner made explicitly, is that demonstrating the shortfall is manageable while parliament is still deliberating may weaken the industry's own negotiating position.

8. The Buyback: $3bn, With an Asterisk Pointing Upward

UBS repurchased $3bn of shares in 2025 and intends to repurchase $3bn in 2026 "with an aim to do more," subject to financial performance, maintaining a CET1 ratio of around 14% and further clarity on the Swiss regulatory regime. Around $2bn of the 2026 programme is expected in the first half. The full-year 2025 dividend is $1.10 per share, up 22%.

Two mechanics were clarified in the Q&A. First, of the $9bn accrued at the parent bank for upstreaming to the holding company, $4.5bn is being paid imminently and the second $4.5bn is being held pending Swiss capital developments, preserving optionality for a larger second-half programme. Second, UBS may begin accruing later this year for part of a 2027 buyback, which would be additional to any 2026 top-up.

Assessment: $3bn is at the low end and below the $4bn threshold we set in October as our upgrade trigger, so the trigger has not fired. But the structure around it improved in three ways: a stated ambition to exceed it, a retained $4.5bn of parent-bank capital explicitly available to fund the excess, and the prospect of a 2027 accrual. This is a company positioning to distribute more and waiting for permission, which is exactly the posture our thesis assumed.

9. The Swiss Capital Ordinance and the Operational-Risk Multiplier

The final capital ordinance is now expected to be published later in the first half, after which the Federal Council must confirm the effective date and phase-in. Management said it is reasonable to expect a phase-in and a prospective application date based on historical practice, while stressing that both remain for the Federal Council to determine.

The quarter's sharpest regulatory disclosure was quantitative and easy to miss. The Swiss application of an internal loss multiplier drives materially higher operational risk risk-weighted assets than the corresponding implementations in the UK, EU and US, where authorities are expected to set the multiplier at 1. On that basis UBS states its operational risk RWAs would be $40bn lower. Separately, final Basel III implementation in Switzerland has added around $60bn of RWAs cumulatively, and the output floor is estimated to have no impact in 2026, up to 1% in 2027 and around 2% in 2028.

Assessment: The $40bn figure is the most concrete quantification UBS has yet given of what Swiss regulatory divergence costs its shareholders. On roughly $493bn of group risk-weighted assets it is about 8% of the denominator, which at a 14.4% CET1 ratio ties up close to $5.8bn of capital that a UK- or EU-domiciled peer would not hold. That is the argument the company will be making in Bern, and it is now on the record with a number attached.

10. US Wealth: Conditional Charter, and a Margin Ladder Pulled Forward

The US national bank charter received conditional approval, delivering on the timeline given in October. Combined with seven consecutive quarters of US loan growth, management pulled its US wealth margin ambition forward by a year: around 15% pre-tax margin in 2026, around 16% in 2027 and around 18% in 2028, against roughly 13% delivered in the fourth quarter and a three-point improvement across 2025.

Management was again careful not to oversell the charter's near-term effect, describing the banking build-out as already underway and the charter as a way to leverage that progress rather than to replace it, and noting it will take time before the additional capabilities contribute meaningful margin.

Assessment: Pulling a margin target forward while flows deteriorate is a coherent position only if the outflows really are the least productive advisers leaving. The evidence so far supports that reading: Americas pre-tax profit grew 32% in the quarter and 34% for the year while assets left. The 2026 target of around 15% is the near-term test, and it is close enough to the current 13% to be gradeable within a few quarters.

11. AT1: A Defence of the Instrument, and Silence on the Appeal

The AT1 topic returned in a different form. Rather than the October court ruling, the question this quarter concerned the Swiss Finance Minister's published comments that AT1 may be unsuitable for the new capital reform because it costs as much as equity and could unsettle markets. The CEO's response was the most forceful passage on the call.

"I would just point out that without AT1, Credit Suisse would have gone through a resolution on Monday morning. So I mean, if one want to question the effectiveness of the AT1, we had a concrete and not theoretical example on how it was critical to restoring very rapidly financial stability in Switzerland and also globally. So from my standpoint of view, that's a first observation. The second one, I would say that the Basel committee has confirmed its total backing of the AT1 as a vital part of the capital stack." — Sergio Ermotti, Group CEO

On the funding side, UBS has issued roughly $13bn of AT1 since the acquisition, reaching 4% of risk-weighted assets against a 4.4% regulatory allowance, and has already placed $3bn of a targeted $3.5bn 2026 programme in January. What did not come up at all was the status of the appeal against the October partial ruling on the Credit Suisse AT1 write-off.

Assessment: The instrument-class debate and the write-off litigation are separate questions, and the call addressed only the first. The CEO's argument is strong on the merits, and the January issuance demonstrates the market has not repriced UBS's own AT1 access. The absence of any update on the appeal, one quarter after it was the dominant topic, is conspicuous.

Guidance & Outlook

Implied path to the exit rate: to reach an underlying return on CET1 capital of around 15% on an annualized run-rate basis at the end of 2026 while averaging around 13% across the year, the fourth quarter of 2026 has to run materially above the average. That is arithmetically consistent with the cost phasing management described, in which decommissioning ramps from midyear, but it concentrates the entire delivery risk into the second half of the year.

Where the Street sits: consensus had roughly $1.26bn of pre-tax profit and $919m of net profit for the quarter and got $1,700m and $1,199m. The forward problem is the opposite: models anchored on a 15% 2026 return and an effective tax rate near the 2025 level need to come down toward the ~13% and ~23% management has now specified.

Guidance style: Materially more generous than in October and more precise than the market expected. Management gave a full target architecture, dated the US flow inflection, quantified the operational-risk multiplier's cost, raised the cost ambition, and volunteered that one divisional target has slipped. The single item still withheld is the one that matters most, and it is withheld because Bern has not published it.

Analyst Q&A Highlights

Whether the Foreign-Subsidiary Capital Shortfall Has Shrunk

The call opened on the accelerated upstreaming of $9bn from subsidiaries and the resulting $26bn reduction in parent-bank risk-weighted assets, and asked directly whether re-running the widely cited foreign-subsidiary capital shortfall calculation would now produce a smaller number. Management confirmed it would, while framing the repatriation as always planned and merely accelerated.

Q: "So if we were to rerun the math that got us to the $24 billion foreign sub capital shortfall earlier in the year, is it fair to say that number today would be lower? And can you give us a sense sort of by how much lower and how much repatriation rebalancing you can still do to work that number lower from here?"
— Chris Hallam, Goldman Sachs

A: "Yes, it's fair to say that if you rerun the numbers that the uptick from 1Q '25 or indeed 2Q '25 would be lower for this. But naturally, as we've said, we always had every intention to upstream this capital. It's important to reiterate that this capital that we've been repatriating from the Credit Suisse subs was always part of our planning. Our strong progress in derisking the entities, as I mentioned, has in these cases, just simply accelerated the return of the capital... But certainly, mathematically, your inference is correct."
— Todd Tuckner, Group CFO

Assessment: A confirmation with real consequences for the regulatory debate. The headline number framing the Swiss capital argument has moved in UBS's favour through its own execution rather than through negotiation. Management declined to size the reduction, which is understandable while parliament deliberates.

Resilience of the 13% 2026 Return Guide in a Choppier Market

A second part of the same turn probed how durable the newly disclosed full-year 2026 return guide is if markets stay volatile rather than delivering the constructive backdrop the Street had assumed in January. The answer was notably unquantified and leaned on controllables rather than on scenario analysis.

Q: "So if we spend much of 2026 with this current market backdrop, elevated volatility, dollar weakness, more questions around public and private market valuation levels, how would that impact your, the group across your various businesses? How resilient would the 13% target be in that context?"
— Chris Hallam, Goldman Sachs

A: "We certainly recognize and in our outlook statement talking about 2026, recognize that we entered the quarter with constructive markets continuing, still seeing higher dispersion and lower correlation in markets that informed constructive 2-way trading in our IB and still our wealth clients remaining risk on despite the need to continue to diversify across asset classes and geography. Of course, as volatility, as we've said, event-driven volatility from various things, whether it be geopolitics or some of what we've been seeing recently, naturally suggests that things can turn quickly. So we're focused just on what we can control, and the ambitions and targets we've laid out reflect that."
— Todd Tuckner, Group CFO

Assessment: The question asked for sensitivity and got posture. "Focused on what we can control" is a fair answer from a bank whose cost programme is the main 2026 driver, but it leaves the revenue side of the 13% guide untested. Since the walk from roughly 12% today to 15% at the exit rate depends on cost delivery rather than revenue growth, that may be the right emphasis; it is not a reassurance a modeller can use.

When US Adviser Attrition Ends

The question we flagged as unanswered in October was asked again and this time drew a dated answer, which is the single most gradeable commitment management made on the call.

Q: "Just coming back to U.S. Wealth Management and maybe just bottom up a little bit around the restructuring of the adviser side. When should we expect adviser attrition to end? And how should we think about net flows as we progress through 2026?"
— Kian Abouhossein, JPMorgan

A: "Asset inflows or outflows from adviser hires or exits, as I've said in the past, do occur several months after announcement. So we can model the impacts on NNA based on announced net recruiting data. And on that basis, we do expect further NNA headwinds through the first half of 2026, after which we expect net recruiting outflow impacts to materially taper. And as Sergio said, for the U.S. business to be a positive contributor to GWM net flows in 2026 overall."
— Todd Tuckner, Group CFO

Assessment: A significant upgrade in disclosure quality. Management has now committed to a mechanism (announced net recruiting data with a several-month lag), a timeframe (headwinds through the first half) and an outcome (positive Americas contribution for the full year). All three are checkable. The claim also carries real risk: it requires the Americas to swing from negative $14bn in a quarter to positive across a year.

The Path from $101bn to $200bn of Annual Net New Assets

A questioner asked how the group bridges from $101bn of 2025 net new assets to the more than $200bn per annum ambition for 2028, and which regions carry it. The answer reframed the last two years as a deliberate headwind period now ending.

Q: "Can you help us better understand the path to your ambition of reaching more than $200 billion net new assets by 2028? And maybe give us some more color around sort of the key regions."
— Antonio Reale, Bank of America

A: "It's important to remember that we guided to $100 billion in 2024 and 2025 because we flagged that there are a number of headwinds that we have to work through around this unprecedented integration, and that's going to create some offset to NNA... And indeed, that's played out over the course of '24 and '25. So what gives us confidence in terms of the build is the fact that we've worked through many of these headwinds we just talked about... Naturally, as we said, the U.S. will, is expected to be a net positive contributor in '26, but we know in '27 and '28, the U.S. has to contribute more in order to grow to greater than $200 billion."
— Todd Tuckner, Group CFO

Assessment: The most honest sentence in the answer is the last one. The 2028 net-new-asset ambition is explicitly contingent on the US, the one region that has produced outflows for two consecutive quarters. That makes the US inflection not merely a divisional issue but the load-bearing assumption of the group's growth ambition.

Gross-to-Net Conversion on the Raised Cost Ambition

Asked how much of the headline savings actually reaches the bottom line after reinvestment in technology, artificial intelligence and adviser retention, management gave a specific planning ratio for the first time at the full-programme level.

Q: "Can you talk us through a little bit more on sort of your expectations for net cost savings from here on?... If you could give us a sense of how much of these savings are reinvested in the business, IT, AI capabilities or SA retention? And how much can be the sort of net cost savings coming through?"
— Antonio Reale, Bank of America

A: "If I look at my gross to net in terms of what I plan for the end of 2026 on the $13.5 billion, I intend to deliver net saves of around 75% of that amount, excluding variable and FA comp, anything that, any headwind from that and effectively is excluded, but it's a 75% gross to net cost capture in how we think about getting to our end of 2026 targets."
— Todd Tuckner, Group CFO

Assessment: A 75% conversion on $13.5bn implies roughly $10.1bn of net savings against the 2022 baseline, and it is consistent with the 77% achieved to date. This is the arithmetic that has to hold for the sub-70% exit-rate cost/income ratio to arrive, and management has now put it on the record in a form that can be checked against reported costs.

What Would Unlock a Larger 2026 Buyback

Two separate turns pressed on the retained parent-bank dividend tranche and on the mechanics of a possible top-up. Management confirmed the $9bn parent-bank accrual is being split, with $4.5bn paid imminently and the rest held pending the ordinance, and that any decision to do more would be announced and accrued in a subsequent quarter.

Q: "I think you said that half of the $9 billion accrued between parent and group could be distributed in the second half of the year subject to bid to save the proposal. I just want to understand what outcome could drive essentially a $4 billion additional buyback in the year, if I understood this correctly."
— Giulia Miotto, Morgan Stanley

A: "So we're imminently paying up half of that or $4.5 billion to the holding company. The other $4.5 billion, we're just taking a prudent wait and see to see what happens in terms of the Swiss regulatory capital framework developments like we had last year, just retaining that optionality to either retain or to pay up."
— Todd Tuckner, Group CFO

Assessment: This is the most concrete thing said about capital returns all day. There is $4.5bn of parent-bank capital sitting explicitly ready to fund a larger distribution, held back solely on regulatory uncertainty. The optionality is real and quantified. It also means the $3bn headline understates what UBS is prepared to do if the ordinance permits, which is precisely why the stock's reaction to the buyback number was arguably an overreaction to the wrong variable.

How the 2026 Exit Rate Will Be Measured, and Target Versus Ambition

A question on the mechanics of exit-rate measurement, and on why 2028 carries the label "ambition" rather than "target," produced an unusually candid exchange in which the CEO overrode the framing of the question.

Q: "How do you want us to measure the exit targets for 2026? Is it a fourth quarter number? Or are there pro forma calculations involved?... And also, is there any particular reason why 2028 remains an ambition rather than a target?"
— Stefan Stalmann, Autonomous Research

A: "At the end of the day, target and ambitions are almost the same. I would say that the target are more short term. What we can see the look-through is for 2026, we can, we have a visibility to talk about targets versus '28 and going forward, it's more of an ambition, but I wouldn't be too bothered about overanalyzing that kind of aspect. We want to get there. So I mean, that's what it is."
— Sergio Ermotti, Group CEO

Assessment: Refreshing, and also the correct amount of weight to place on a 2028 number. On the measurement question, management said it will annualize the normalized run-rate at year-end for the numerator and construct a reasonable denominator looking back, then let the 2027 reported cost/income ratio settle whether the target was met. That is a sensible standard and it means the real verdict on the 2026 exit rate arrives with the 2027 reported results, not with the fourth-quarter 2026 release.

Bridging P&C's Cost/Income Ratio to 48%

With the sub-50% end-2026 target slipping, a questioner asked what supports the 2028 commitment to around 48% reported and how much of it is Credit Suisse system decommissioning. The answer split it into two drivers and confirmed the slip explicitly.

Q: "You said in your presentation that you continue to target that exit rate '26 of below 50% and then reported 48% for '28... So can you maybe elaborate again on what gives you the confidence that you can decline the cost-income ratio by so much over 10 points basically in '26? And how much of that would be driven specifically by the decommissioning of the IT system at Credit Suisse?"
— Flora Bocahut, Barclays

A: "As Sergio mentioned, we're, it's unlikely we would meet the less than 50% cost-income ratio on an underlying basis in 2026, given the NII headwinds. And as you rightly say, we said that the 48% can be achieved by '28 even in the current interest rate environment. So what gives us confidence on that? So I would say it's 2 things broadly. One, it's P&C building out their non-NII revenues... And then, of course, on the expense side... P&C will be a big beneficiary of the gross cost saves that we take out in 2026."
— Todd Tuckner, Group CFO

Assessment: The answer is honest about the slip and appropriately splits the bridge between a cost half we can underwrite and a revenue half we cannot. Growing non-interest revenues in a Swiss corporate franchise facing trade-policy pressure is the harder of the two, and it is the part on which the 2028 number depends most.

What They're NOT Saying

1. Anything about the AT1 write-off appeal. One quarter after it dominated the call and three months after UBS announced its intention to appeal alongside FINMA, the proceeding went unmentioned. The only AT1 discussion concerned the instrument's role in the future capital framework and the 2026 issuance plan.
2. The size of the reduced foreign-subsidiary capital shortfall. Management confirmed the number would be lower after the $9bn upstream and declined to say by how much, even as a range.
3. Any sensitivity around the 13% 2026 return guide. Asked directly how resilient it is to a volatile market backdrop, the answer described the environment and reasserted focus on controllables without quantifying downside.
4. What "aim to do more" is worth. The $3bn is committed; the increment is contingent on the ordinance and explicitly deferred to a subsequent quarter. The $4.5bn retained at the parent bank sets an upper bound on the funding, not on the intent.
5. The FRC revenue split between foreign exchange and precious metals. Asked for a rough breakdown of the division's 46% growth line, management committed to come back with it rather than answering. In a quarter where precious metals were unusually active, the split matters for judging repeatability.
6. Quarterly net-new-asset phasing within the >$125bn 2026 guide. The full-year number is committed and the Americas inflection is dated to the second half, but no path was given for how the quarterly progression should look, which makes the first two quarters of 2026 hard to grade against the guide.
7. Whether the November debt buyback's $457m loss was contemplated in prior guidance. The exercise was disclosed clearly and its benefits quantified at around $100m a year for three years, but no comment was offered on why the trade was struck in the fourth quarter, when it depressed the reported result the market was about to judge the year on.
8. How the 2028 reported ambitions change if the ordinance changes the denominator. The 18% return is stated "under the current capital framework." No alternative scenario was sketched, so the ambition is undefined for exactly the outcome most likely to alter it.

Market Reaction

• Pre-print setup: The ADR closed at $47.67 on 3 February, up 2.9% year to date against 1.1% for the S&P 500, up 47.1% over the trailing twelve months, and down 0.9% over the trailing thirty days. The 52-week closing range entering the print was $26.12 to $49.25, leaving the stock about 3.2% below its 52-week closing high.
• Reaction session: UBS reports before the US open, so the print day is the reaction day. The ADR gapped down hard, opening at $45.37 (-4.8%), traded a $44.53 to $45.71 range, and closed at $44.85, down 5.9% or $2.82. The S&P 500 fell 0.5% on the session, so nearly all of the move is stock-specific.
• Volume: 4.7m shares against a 2.1m thirty-day average, or 2.3x normal, and the heaviest session of the period. This is repositioning at scale, not a drift.
• Primary listing: the Swiss line fell more than 4% on the day, a smaller decline than the ADR's close-to-close move.
• Position versus range: the close leaves the ADR 8.9% below its 52-week closing high and 71.7% above the 52-week closing low.

A stock that beats pre-tax profit by roughly 35% and falls 5.9% on 2.3x volume is being repriced on something other than the quarter. Three candidates arrived together and all three point at the trajectory rather than the print. Net new assets of $8.5bn against $38bn a quarter earlier put a number on the cost of the US strategy that was easy to ignore when the Americas was losing $9bn and impossible to ignore at $14bn. The CET1 ratio fell 40 basis points to 14.4%, mechanically and as guided, but into a market that wanted evidence of headroom before the ordinance. And the first full-year 2026 guide, at around 13% underlying return on CET1 capital and a 73% cost/income ratio, sits roughly two points and three points respectively away from the exit-rate targets that had been doing the work in most models.

The tax line compounds all three. A 2025 effective rate of 11.9% moving to around 23% in 2026 is worth more than a point of return on its own, and it converts a year of 55% earnings-per-share growth into a much flatter 2026 for anyone who had not modelled the normalization. Set against that, the buyback landed at $3bn rather than the larger figure some had hoped for, so the one variable that might have offset a softer earnings path did not. The move looks to us like a rational repricing of 2026 estimates rather than a verdict on the franchise, and it is worth noting the market gave back only a fraction of a 47% twelve-month gain.

Street Perspective

Debate: Is the 2026 Guide a Reset or a Waypoint?

Bull view: Nothing was cut. The 2026 exit-rate targets of around 15% and below 70% are unchanged, and the newly disclosed full-year figures simply make the phasing explicit: the hardest decommissioning cannot start until the Swiss migration finishes in the first quarter and ramps from midyear, so savings are necessarily second-half weighted. Management raised the cost ambition in the same breath, which is not what a company resetting expectations does.

Bear view: A full-year 2026 return of around 13% against an exit rate of 15% means the fourth quarter has to run well above the average, concentrating all delivery risk into the final months of a three-year programme's hardest phase. Add a tax rate roughly doubling and the practical effect on 2026 earnings estimates is a downgrade regardless of what the target slide says.

Our take: A waypoint, disclosed late. The phasing logic is genuine and the cost-ambition raise is inconsistent with a company losing confidence. But management chose to publish the full-year path only after the market had spent a year anchoring on the exit rate, and the gap between the two was always going to land as bad news whenever it arrived. We treat the roughly 13% as the number to model and the 15% as the option.

Debate: Are the US Outflows Strategy or Symptom?

Bull view: The evidence is in the profit line. Americas pre-tax profit grew 32% in the quarter and 34% for the year while assets left, the margin rose two points to 13%, and loan balances grew for a seventh consecutive quarter. Least-productive advisers are leaving, the compensation grid is doing its job, and management has now dated the inflection to the second half of 2026 with a mechanism anyone can check.

Bear view: Outflows accelerated from $9bn to $14bn quarter over quarter, which is not what tapering looks like. The company is guiding to a positive full-year 2026 Americas contribution immediately after its worst quarter, and it has conceded the US must contribute more still to reach the 2028 ambition. A margin improvement bought by shrinking in the largest wealth market is not obviously repeatable.

Our take: Strategy, executed with worse flow consequences than management expected. The profit evidence is too strong to call this a symptom of franchise decay. But the acceleration from $9bn to $14bn is real, and management's dated commitment converts this from an open-ended concern into a specific test with a specific deadline. We will grade it in July.

Debate: Does the Ordinance Still Justify a Discount?

Bull view: The picture improved materially this quarter. UBS upstreamed $9bn from subsidiaries and cut the parent's risk-weighted assets by $26bn, shrinking the foreign-subsidiary shortfall through execution alone. It has $4.5bn retained at the parent explicitly available for a larger distribution. The ordinance publishes later in this half, and a phase-in with prospective application is the reasonable expectation. Once it lands, the discount has nothing to attach to.

Bear view: The publication date has already moved once, the effective date and phase-in are for the Federal Council rather than for UBS, and the Swiss internal loss multiplier alone costs $40bn of operational risk risk-weighted assets versus UK, EU and US treatment. Every 2028 ambition carries the qualifier "under the current capital framework," which is the framework being rewritten.

Our take: The discount is still warranted but it is narrower than it was in October, because UBS has demonstrably reduced the exposure through its own actions and has quantified the divergence cost for the first time. This remains the reason we own a Hold rather than an Outperform, and the ordinance's publication in the first half is the event that resolves it.

Model Update & Valuation Framework

Valuation. At the $44.85 close the ADR trades at 1.67x the $26.93 tangible book value per share reported for the quarter, against 1.44x when we initiated in October. Tangible book grew 1.5% sequentially net of buybacks and dividends. Rolling it forward at roughly that pace puts tangible book per share near $28.60 a year out. Our 12-month target moves to $47 from $40, applying a multiple of about 1.64x to that figure. From the $44.85 spot that implies +4.8%, which keeps the rating at Hold.

Why the target rises while the rating does not. Two things genuinely improved this quarter and both belong in the multiple rather than the near-term earnings: the 2028 architecture gives the market a credible terminal profitability level for the first time, and the accelerated capital upstreaming shrinks the regulatory exposure through execution. Neither changes 2026 earnings, which in fact got harder on the tax line. A higher multiple on a similar book value, with the same binary still unresolved, produces a higher target and the same rating.

What moves the target from here. An ordinance in the first half that lands at or better than the industry's proposals, combined with a buyback top-up funded from the retained $4.5bn, would support 1.85x or better and a target near $53. A punitive ordinance, or a second-half failure of the Americas flow inflection that undermines the 2026 net-new-asset guide, would argue for 1.35x to 1.45x and a target in the high $30s to low $40s.

Thesis Scorecard: Q3 2025 Signposts Revisited

We set seven signposts in October to be graded at this release. Six delivered.

Standing Thesis Scorecard

Overall: The thesis is unchanged in structure and better evidenced in both directions. The cost pillar strengthened to the point where it is now the least doubtful part of the story. The wealth-flow pillar weakened enough to move from At Risk to Challenged. The regulatory bear point contained slightly, for the first time, through UBS's own execution rather than through a political outcome.

Action: Hold. Our upgrade trigger, a 2026 buyback of $4bn or more, did not fire; the company committed $3bn. The other trigger, a benign ordinance, has not yet been tested because the ordinance has not been published. Meanwhile the multiple expanded from 1.44x to 1.67x tangible book while the 2026 earnings path got harder on tax. We would rather own this after the ordinance publishes in the first half, with the buyback increment known, than pay a re-rated multiple for the same unresolved binary.