The Rate Shock Reopened the Door: A Third Guide Raise and a Mechanical $1B 2027 NII Lift at 2x Tangible Book. Upgrading PNC to Outperform

Results vs. Consensus

Q3 2026 Scorecard

Year-Over-Year Comparisons (3Q26 vs. 3Q25)

Quarter-Over-Quarter Comparisons (3Q26 vs. 2Q26)

Revenue. Reported revenue fell $313M (−4.6%) sequentially to $6.562B, and $252M of that decline is the swing in integration costs and significant items (from +$218M in Q2, mostly the $448M Visa gain, to ($34M) in Q3). Excluding integration costs and significant items from both quarters, revenue slipped only $61M (−0.9%), as $99M of NII growth was more than offset by lower fees. Fee income fell $137M (−6.0%) to $2.142B, a little worse than the 5–5.5% decline guided in July, because capital markets and advisory fell 25% (not the guided ~20%) off the record second quarter. The other fee lines were steady to slightly better: asset management and brokerage +2%, lending and deposit services +3%, card and cash management flat (treasury-management growth offset by FirstBank conversion fee waivers), and mortgage −6% on lower residential volumes. Other noninterest income of $214M beat the $150–200M guide on higher private-equity revenue, despite ($25M) of net Visa items. Year over year, revenue grew 10.9%, with NII doing nearly all of the work (+15.3%) and fees up only 3.5% against a strong Q3 2025 capital-markets comparison.

Margins and operating leverage. NIM fell 2bp to 2.94%, the first sequential decline of 2026 and the main disappointment of the quarter. The mechanics are visible in the supplement: the total yield on earning assets rose 8bp to 4.90% (securities +14bp to 3.59%, loans +3bp to 5.50%), but the rate on interest-bearing liabilities rose 10bp to 2.49%, interest-bearing deposit costs rose 7bp to 1.98%, and average noninterest-bearing balances fell 2%. Management blamed three things: lower-spread, high-quality corporate loans "came in hotter" than planned, higher-spread loans paid down, and expected escrow balances did not arrive. On the expense side, adjusted noninterest expense fell $113M (−2.9%) to $3.724B on lower compensation, and the efficiency ratio improved to 57% from 60%. Adjusted revenue fell 0.9% while adjusted expense fell 2.9%, so the company still produced roughly two points of positive operating leverage in the quarter, and it guided 4–4.5% for the full year.

EPS. Adjusted EPS of $5.09 rose 4.9% sequentially and 17.0% year over year, and nine-month adjusted EPS now totals $14.26. The $0.12 gap to GAAP $4.97 is $36M of FirstBank integration costs plus ($25M) of net Visa items (an $11M gain on retained Visa shares, net of ($36M) of Class B-3 derivative marks tied to litigation-escrow funding). The tax rate of 20.0% was close to normal, so the beat was not tax-flattered. Our concern is composition, not size: with revenue and NII below the Street, the beat came from expense timing, a provision that ran $68M below charge-offs, and a lower share count (401M diluted, from 403M). Those are legitimate, but they are not where a premium bank multiple gets earned.

Segment Performance & Balance Sheet

PNC reports three operating segments plus an "Other" bucket that holds treasury, securities results and corporate items. The second quarter's Visa gain sat in Retail Banking and the securities-repositioning loss sat in Other, so sequential segment comparisons are noisy. Year over year, each segment grew revenue by roughly 6–7%, a balanced picture that says the growth is not concentrated in one engine.

Corporate & Institutional Banking: Share Gains at Record Pace, Thinner Spreads

C&IB again carried the balance sheet. Average segment loans rose $11.6B (+5%) to $247.7B and are up 17% year over year, led by Corporate Banking ($146.4B average, up about 24% from a year ago). Every named industry category in the C&I book grew on a spot basis (only the residual "other industries" line fell), led by financial services (+$4.8B) and followed by real-estate-related, wholesale trade and manufacturing. Segment revenue slipped 2.2% to $3.212B because capital-markets fees normalized off the record second quarter, but segment NII rose $53M and treasury-management revenue of $1.175B is up 8% year to date. Segment earnings of $1.552B were flat year over year, because expenses rose 13% and the provision rose to $76M from $44M.

"Expansion markets have added approximately 500 new client relationships year to date, contributing materially to a 5% average C&IB loan growth during the quarter and 17% growth year over year." — Bill Demchak, Chairman & CEO

Management was explicit that the expansion markets are no longer a side bet: in corporate banking, ex the national businesses, the majority of in-market loans now sit in those markets, and they grow two to three times faster than the legacy footprint. The trade-off is spread. The loans PNC is winning are high-grade corporate credits, often paired with syndication and treasury-management mandates, and they carry lower spreads than the higher-yielding loans that paid down.

Assessment: This is the share-gain engine at full speed, and it is the main reason we are comfortable upgrading into a tougher rate backdrop. The cost is a structurally lower loan spread, which is now showing up in NIM. That is acceptable as long as the fee attachment follows. Capital-markets fees up 40% year to date say it does, but C&IB earnings flat year over year say the expense and provision cost of this growth is not trivial. The segment's next test is whether Q4's guided capital-markets rebound arrives.

Retail Banking: Accounts Growing, Deposits Slipping

Retail Banking earned $1.425B, down 18% sequentially only because the second quarter held the $448M Visa share-exchange gain. Excluding Visa items from both quarters, segment noninterest income rose about 7% and total revenue was flat. Year over year, earnings rose 5% and revenue 6.5%, helped by FirstBank. The client numbers are healthy: 215 thousand net new checking accounts so far in 2026, total accounts up 2% year over year (6% in the Southwest), card balances up 12% and brokerage assets up 9%. The weak spot is deposits. Average retail deposits fell 2% sequentially to $267.1B, driven by lower consumer interest-bearing and noninterest-bearing balances, and segment loans fell 2% on mortgage runoff and the student-loan transfer.

"In retail banking, we continue to grow clients organically adding 215 thousand net new checking accounts so far in 2026. With total accounts up 2% year-over-year, including 6% growth in our Southwest markets." — Bill Demchak, Chairman & CEO

The student-loan exit is housekeeping: PNC stopped originating these loans last year, moved the $1.4B book to held-for-sale, and took $35M of charge-offs on the transfer. Management expects a de minimis effect from the sale.

Assessment: Account growth is the durable part of the retail story, and the deposit decline is the part to watch. A 2% sequential drop in consumer balances in the first quarter of a new hiking cycle is consistent with rate-seeking migration, and management did not address it on the call. Retail deposits are the cheapest funding in the company; if they keep leaking while corporate deposits and brokered CDs fill the gap, the beta argument in Key Topic 4 becomes harder to defend.

Asset Management Group: Steady Flows, Expansion-Market Tilt

AMG earned $128M (−5% QoQ, +7% YoY) on flat sequential noninterest income of $271M. Discretionary client AUM rose 1% to $250B on positive net flows and is up 10% year over year. Nondiscretionary AUA fell 3% on net outflows, so total client AUA was $499B. Expenses rose 2% on continued investment.

"Asset management fees are up approximately 10% year-to-date, and assets under administration totaled $500 billion. Favorable equity markets have obviously helped here, but once again, our expansion markets have accelerated our growth. In these markets, AUM is growing 5x faster than our legacy markets." — Bill Demchak, Chairman & CEO

Assessment: AMG is a confirming signal, not a driver, as it has been throughout our coverage. The useful detail is that the expansion-market tilt shows up here too, which supports the view that the franchise build is broad rather than concentrated in corporate lending.

Key Bank KPIs

The KPI panel splits cleanly in two. The earnings-power lines all moved the right way: ROTCE rose to 18.28%, above the 18% exit target with a quarter to spare; ROE rose 28bp to 13.89%; the efficiency ratio fell to 57%; PPNR grew for a second straight quarter on a reported basis; and delinquencies fell to 0.34%. The balance-sheet lines moved the wrong way, mostly for reasons tied to rates and growth. NIM dipped. Tangible book fell 3.1% because AOCI deteriorated by $2.4B to ($6.5B) as rates rose across the curve. CET1 slipped to 9.8% as RWA grew with loans. The allowance ratio fell again, to 1.42% from 1.48%, even as nonperforming loans rose 8% and new nonperforming-asset inflows roughly doubled to $692M from $335M. The NPL rise traces to one large credit and reserved office loans, not to rates, but it lands in the same quarter. That divergence between rising returns and a weakening balance-sheet profile is what the start of a hiking cycle tends to look like at a bank. It also explains why the stock de-rated while earnings estimates rose.

Key Topics & Management Commentary

Overall Management Tone: Confident on growth and noticeably less committal on the margin. Management led with client momentum and the raised guide, and treated the start of a Fed hiking cycle as a manageable funding question rather than a threat. The one visible shift from July was on NIM: last quarter management was "on record" for 3%, and this quarter it called NIM "an outcome" and the 3% figure "a calculation." Where the call was thin was on the weaker balance-sheet lines (retail deposit outflows, the falling reserve ratio, the AOCI-inclusive capital picture), which management either did not raise or answered only when asked.

1. The Third Straight Guide Raise: Loans ~13.5%, NII ~15.5%, Revenue ~13.7%

For the third quarter in a row, management lifted the full-year outlook (all ex-integration costs and significant items, versus 2025). Average-loan growth went to ~13.5% from ~12.5%, NII to ~15.5% (the top of the prior 15–15.5% range), noninterest income to ~10.5% from ~9%, and total revenue to ~13.7% from ~13%. Expenses also went up, to ~9.5% from ~8.5%, partly on revenue-linked costs in capital markets and private equity, and the tax rate stayed at ~19.5%. On our arithmetic the revenue guide implies roughly $26.3B for 2026, and the NII guide implies about $16.6B, which requires Q4 NII of roughly $4.37B (+4% sequentially, exactly what management guided).

"We expect a full year average loan growth of approximately 13.5%. We expect full year net interest income to be up 15.5%. We expect noninterest income to be up approximately 10.5%. Taking the component pieces of revenue together, we expect total revenue to be up approximately 13.7%." — Rob Reilly, EVP & CFO

The macro assumptions under the guide changed more than the guide did. In July, management assumed no Fed moves in 2026. Now the base case is a 25bp hike in December and another in March 2027, taking the funds rate to 4.25–4.50%, with GDP growth above 2% and steady unemployment. The guide went up anyway.

Assessment: A third consecutive raise during a quarter in which the rate backdrop turned against banks is the strongest fundamental datapoint in the report. The quality caveat is that the raise is volume-led (loans, fees) while the margin is going the other way, and the expense guide rose a full point, close to the 1.5-point raise in noninterest income. The full-year positive operating leverage of 4–4.5% is still excellent. But this is a bigger, busier bank, not a more profitable-per-dollar one.

2. NIM Slips to 2.94%: Mix, Paydowns and Late-Quarter Volume

NIM fell 2bp when the Street expected it to rise toward 3%. Management gave three reasons. First, the high-grade, lower-spread corporate loans PNC is winning came in faster than planned. Second, the loans that paid down were the higher-spread, riskier ones, so the average spread on the book fell about 3bp. Third, expected escrow balances (which would have added noninterest-bearing funding) did not arrive. A fourth, timing factor helped explain the NII shortfall: half the quarter's loan growth was booked in the final two weeks, so it barely contributed to Q3 NII but carries fully into Q4.

"The modest decline in net interest margin was largely a function of the mix in loan growth, and, to a lesser extent, increased funding costs. Consistent with what we have seen throughout the year, loan growth during the third quarter continued to be concentrated in high credit quality, lower spread loans." — Rob Reilly, EVP & CFO

"And so what we saw was the mix of our existing loans to pay downs were biased towards riskier stuff paying off, which is a good thing, but it drops our spread." — Bill Demchak, Chairman & CEO

The Q4 path to 3% is now framed as arithmetic rather than a promise. Loan yields and deposit costs should rise together and roughly offset, so the lift has to come from fixed-rate repricing: about $5B of securities running off in Q4 at a 2.65% yield, and a swap book whose received-fixed rate rose 17bp to 3.92%.

Assessment: We accept the explanation. It is consistent with the loan detail, and the "high-grade loans dilute NIM but bring fees" argument has been true all year. But two of the three causes (spread compression on the book and missing noninterest-bearing balances) are exactly what a hiking cycle produces, and the July "on record for 3%" has softened to a forecast. We now treat 3% in Q4 as roughly a coin flip and model 2.97–3.02%. The thesis does not need 3%. It needs NII dollars, and those are guided up 4% in Q4.

3. 2027 NII: A $1.0–1.1B Lift Called "Mechanical"

The most important forward datapoint came in Q&A. Asked whether NII growth would slow sharply in 2027, the CFO said 2027 NII would exceed the annualized Q4 2026 run rate, and agreed with a Street range of $1.0–1.1B of growth over 2026. On our numbers that is roughly $17.6–17.7B, or +6.0–6.6% on a high 2026 base, and it comes before any benefit from a steeper curve or continued loan growth above GDP.

"Our received fixed swap portfolio was $83 billion at quarter end, with an average received fixed rate of 3.92%. Which increased 17 basis points from last quarter. Importantly, we have a clear path for continued fixed-rate asset repricing over the next several years." — Rob Reilly, EVP & CFO

"But the fixed-rate asset repricing and the mechanical nature of that is real. It continues for P and C and the magnitude is big." — Rob Reilly, EVP & CFO

Assessment: This closes one of the four re-engagement signposts we set in July (a quantified 2027 bridge). It is not the 20% ROTCE bridge we asked for, but it is the more important half: a bank with locked-in NII growth of ~6% into a year when peers may face margin pressure from higher deposit costs. Higher long rates make this repricing tailwind larger, not smaller, which is the counterweight to the AOCI damage in Key Topic 5.

4. The Rate-Hike Regime: Funding Costs and Deposit Betas

The Fed's 25bp hike on September 16 (to 3.75–4.00%, the first increase since 2023) changed the funding conversation. The rate paid on interest-bearing deposits rose 7bp to 1.98% (5bp to 1.96% excluding brokered CDs), average noninterest-bearing balances fell 2%, and PNC added $7B of brokered CDs as the cheapest marginal wholesale funding. Management expects deposit betas to behave like a normal cycle: high on corporate and wealth balances, low on lower-end consumer balances, settling at a cumulative beta in the mid-50s. It also stressed that it is roughly neutral to short rates and benefits from a steeper curve.

"In regard to deposit costs, our total rate paid on interest bearing deposits increased 7 basis points to 1.98%. Including brokered CDs. Excluding brokered CDs, the rate paid increased 5 basis points to 1.96%." — Rob Reilly, EVP & CFO

"Overall, we remain well positioned for a rising rate environment, which will support net interest income, capital generation and balance sheet flexibility." — Rob Reilly, EVP & CFO

Assessment: On NII, management's case is credible: the balance sheet was built to be short-rate neutral, and the repricing math improves when long rates rise. The risk is in the mix. Corporate deposits come in at high betas, retail balances fell 2% in the quarter, and brokered CDs are not core funding. A mid-50s cumulative beta is a reasonable base case, but the cost of growth funding is rising. That is why we are adding a rate-regime bear point rather than dismissing the hike as neutral.

5. AOCI, Tangible Book and Capital: The Visible Cost of Higher Rates

The rate move hit the balance sheet directly. AOCI deteriorated by $2.4B to ($6.5B), and net unrealized losses on available-for-sale securities nearly doubled to $4.0B as securities duration extended to 4.1 years from 3.6. Tangible book per share fell 3.1% to $107.64, flat year over year. AOCI had already drifted lower through the first half (from ($3.4B) at year-end 2025), but retained earnings outran it; this quarter's $2.4B hit did not. As a Category III bank, PNC excludes AOCI from CET1, which eased to 9.8% on loan-driven RWA growth. Including AOCI, management put CET1 in the "low nines." The buyback stepped down to ~$560M from ~$610M, and Q4 is guided at a similar level.

"Our tangible book value was $108 per common share, down 3% linked quarter entirely driven by the decline in AOCI. And was stable compared with the same period a year ago." — Rob Reilly, EVP & CFO

Two offsets matter. The losses are marks on high-quality securities and swaps that accrete back as they approach maturity, and PNC's stress capital buffer stays at the 2.5% regulatory minimum through December 31, 2027. The pending Basel III changes would bring AOCI into capital, but management estimates the associated risk-weighted-asset relief at roughly a point of CET1.

Assessment: This is the price of the new regime, and it moves our TBV-compounding pillar from on track to at risk. The damage is reversible: it is a mark, not a loss of earning power, and it pulls to par. But tangible book is the anchor of a bank valuation, and further increases in long yields would extend the drawdown. Capital is adequate, not abundant: CET1 below target and a slightly smaller buyback say the loan growth is now consuming capital faster than earnings replace it.

6. Loan Growth: Organic Momentum Running Well Ahead of Plan

Average loans rose $9.0B (+2.5%) and spot loans $10.8B (+2.9%), both above the 1–2% July guide. Year to date, PNC has added $47B of loans (+14%), of which about $15B came with FirstBank and more than $30B was organic. C&I rose 4% on new production and higher utilization, CRE rose 4% as the asset class keeps re-accelerating, and consumer loans fell 1% on mortgage runoff and the student-loan transfer. The Q4 guide calls for another ~2.5% increase in average loans, helped by the late-quarter surge.

"When we step back and look at the year as a whole, we have grown considerably more than we anticipated coming into 2026." — Bill Demchak, Chairman & CEO

"While approximately $15 billion came through the FirstBank acquisition, more than $30 billion of the growth has been generated organically." — Rob Reilly, EVP & CFO

Assessment: More than $30B of organic loan growth in nine months, on a ~$331B year-end starting book, is the clearest confirmation yet of the expansion-market thesis. The honest caveat came from management itself: some of the late-quarter rush may be clients borrowing ahead of higher rates, and a slowdown in AI-related capital spending would slow loan growth across the economy even though PNC does not lend directly to data centers. We treat the pace as above trend, not as a new run rate.

7. Capital Markets: The Step-Down Was Worse, the Rebound Is Guided

Capital markets and advisory revenue fell $144M (−25%) to $433M, worse than the ~20% decline guided in July, as the record second quarter did not repeat. Year to date the line is still up roughly 40% ($1.473B vs. $1.059B), and management guided a ~15% sequential rebound in Q4 on strong pipelines, including Harris Williams. Put together, that implies full-year capital markets up roughly 27%, inside the 25–30% range management gave in July.

"We had a really strong second quarter that we did not expect to continue into the third quarter. But the pipelines are good." — Rob Reilly, EVP & CFO

Assessment: This played out almost exactly as our July bear point described: Q2 was a peak to normalize from, not a floor. The useful news is that the full-year number holds and Q4 is guided up even as larger banks describe a cooler post-Labor Day tape. We keep the cyclicality bear point as "emerging," because the line can swing by $100M+ in a quarter, but the fee base is clearly larger than it was a year ago.

8. Credit: NPLs Up 8% on One Name and the Office Pull-Through

Credit was mostly benign, with one blemish. Delinquencies fell 9% to $1.3B (0.34% of loans), and net charge-offs of $257M (0.28%) included $35M from the student-loan transfer, so underlying charge-offs of ~$222M matched Q2. But nonperforming loans rose $167M (+8%) to $2.2B. Service-provider NPLs jumped to $269M from $131M on one large commercial credit, and CRE NPLs rose to $559M from $464M as previously reserved office loans moved through the pipeline. Gross inflows of new nonperforming assets roughly doubled to $692M from $335M.

"Nonperforming loans were up $167 million or 8% to $2.2 billion. The increase was primarily driven by the pull through of previously reserved CRE office loans as well as 1 large commercial credit." — Rob Reilly, EVP & CFO

Management described the large credit as a company with cash-flow problems, not a theme, and said the office losses have long been reserved.

Assessment: We take the idiosyncratic explanation at face value for now. The office pull-through is the expected end of a cycle PNC reserved for long ago. But the timing is uncomfortable: NPLs turned up again in the first quarter of a hiking cycle, while the allowance ratio kept falling. Credit is not yet a thesis problem, and it is now on the watch list in a way it was not in July.

9. Expenses and Operating Leverage: The Line That Carried the Beat

Adjusted expense fell $113M (−2.9%) to $3.724B, at the better end of the 2–3% decline guided, mainly on lower compensation after the capital-markets-heavy second quarter. Integration costs dropped to $36M from $127M as FirstBank costs run off. For Q4, management guided adjusted expense up ~5%, partly revenue-driven (capital markets, private equity) and partly seasonal, while keeping full-year positive operating leverage at 4–4.5%. The $350M continuous-improvement program is on track and funds technology investment.

"The important part obviously and you know this is the positive operating leverage that we are pointing to still in excess of 4%. between 4% and 4.5%, which is fantastic." — Rob Reilly, EVP & CFO

Assessment: Four-plus points of full-year operating leverage, while integrating an acquisition, is best-in-class execution and confirms the operating-leverage pillar. The Q4 step-up of ~5% shows that a meaningful part of the Q3 expense decline was timing, so we do not extrapolate the Q3 cost line.

10. AI: "Purposefully Quiet," With the Infrastructure Already Built

Asked whether AI saves more than it costs, the CEO described a multi-year plan. The groundwork (a unified data lake, digitized policies, HR and access-control systems that recognize AI agents) is built. The visible cost today is rising depreciation from data-center rebuilds that include PNC's own GPUs. The next phase is re-engineering technology development and then product delivery. He was explicit that PNC will not publicize results until they show up, and that the goal is to build the capability in-house rather than rent it from vendors.

"What comes next is the reengineering of the company using AI harnesses which would speed the development and ultimately lower the cost with technology development and maintenance followed quickly by product development delivery that aids revenue. We have been purposefully quiet on all of the above." — Bill Demchak, Chairman & CEO

Assessment: A credible, unflashy posture, and the in-house emphasis fits PNC's cost culture. It also means there is no number to model. We treat AI as a long-dated efficiency option (it is the stated reason PNC is avoiding large M&A, below), not as a 2027 earnings driver.

11. M&A: Organic Growth and the AI Rebuild Come First

Asked about acquisition appetite after FirstBank, the CEO effectively ruled out a large deal. He said organic growth is surprising even management, and that a big integration would slow an AI-driven reorganization he views as strategically more important than scale. Small deals were not excluded.

"So put differently, I think the world's gonna change in the next couple of years. And we want to lead that. If that gets slowed down because we otherwise got into some big M&A deal, I think you could destroy massive value." — Bill Demchak, Chairman & CEO

Assessment: Positive for the stock. It removes the deal-dilution overhang that often caps super-regional multiples during consolidation waves, and it keeps capital for organic growth and buybacks. It also bears on our scale bear point: PNC is choosing to compete on growth and technology rather than size, which keeps that risk contained.

Guidance & Outlook

Full-Year 2026 Guidance (vs. 2025, ex-integration costs and significant items): Raised Again

Fourth-Quarter 2026 Guidance (vs. Q3 2026)

The Q4 guide adds up to adjusted revenue of roughly $6.84B (NII ~$4.37B, fees ~$2.22B, other ~$250M) and adjusted expense of ~$3.91B, for adjusted PPNR near $2.93B, about 2% above Q3's ~$2.87B. Net charge-offs are guided up to ~$250M from ~$222M underlying, so the provision likely rises too as loans grow ~2.5%.

Implied 2026 EPS. Nine-month adjusted EPS is $14.26 ($4.32 + $4.85 + $5.09). Using the Q4 guide, a provision of ~$230–270M, a Q4 tax rate near 18.5% (consistent with the ~19.5% full-year guide), and ~399M shares, we get Q4 adjusted EPS of roughly $5.15–$5.25 and full-year 2026 of roughly $19.40–$19.50, above the $18.75–$19.25 we carried in July.

Street at: Entering the print, Q4 consensus sat near $5.13 of EPS and $6.74B of revenue; the guide implies revenue about 1.5% higher, so Q4 estimates should rise. Full-year 2026 consensus of roughly $19.00–$19.23 also looks low against the nine-month result plus the guide. For 2027, the Street sits near $21.4–21.5. The newly endorsed $1.0–1.1B NII bridge supports that, and we model $21.25–$21.75.

Guidance style: Conservative on volumes, as usual. Q3 loans beat the July guide, and the full year has been raised three times. But the guide has not been conservative on the margin: Q3 NII missed the company's own 3–3.5% guide and fees landed slightly below. The pattern this year is that management under-calls loan growth and over-calls spread. We expect that pattern to continue in a hiking cycle.

Analyst Q&A Highlights

Why the Margin Fell, and Whether Low-Spread Loans Are Worth It

The opening question went straight at the margin: deposit costs up, NIM down, and a loan book tilting toward lower-spread credits. Management declined to defend NIM as a metric. The CEO said he would talk only about NII, and the CFO argued that the corporate loans PNC is winning bring fee income that the margin cannot show.

Q: "I appreciate the color that you gave on the margin dynamics You saw the compression in the quarter and deposit costs up 7 bps or 5 bps ex the brokered." … "And I guess maybe just a little bit of color on what the trade off is, how you are looking at lower spread, hard-quality loans in terms of the impact on NII and as you look out from here?"
— John Pancari, Evercore ISI

A: "I am not going to talk about NIM because it is an outcome. So Rob kind of says, we will be over 3. that is on a forecast. Many things could change that. I will talk about NII, though." … "So together, accretive to NII, accretive to earnings, to ROTCE, not necessarily NIM because of the fee portion of that. So if you just look at NII and NIM in isolation, you are not seeing the whole transaction."
— Bill Demchak, Chairman & CEO; Rob Reilly, EVP & CFO

Assessment: The economics argument is sound, and capital-markets fees up ~40% year to date support it. But the posture changed: in July the CEO said the company was "on record" for 3%, and this quarter he called it a forecast that "many things could change." Read that as lower confidence in the margin, not as a change in strategy.

Whether Higher Rates Are Starting to Slow Borrowers

A natural question for the first quarter of a hiking cycle was whether higher rates had begun to affect client behavior, given that half the loan growth arrived in the final two weeks. The CEO said the company had discussed exactly that the night before without reaching a conclusion: no slowdown is visible yet, and some borrowing may be clients getting ahead of higher rates.

Q: "I was hoping you could speak sort of broadly to client sentiment and how if at all, their thinking is being impacted by higher rates." … "And then, you know, is there sort of a tipping point in your mind where you know, higher rates do start to impact the growth dynamic, and client health, etcetera."
— Scott Siefers, TD Cowen

A: "Intuitively, higher rates, you know, this goes forward, is going to crimp CapEx and otherwise slow borrowing, but we are not seeing any instance of that yet and our pipelines remain strong and we will see. You gotta imagine some of that is in anticipation of rates going up on the margin, at least a piece of it, but We do not know."
— Bill Demchak, Chairman & CEO

Assessment: An honest answer that cuts both ways. Strong pipelines support the Q4 loan guide, but management itself flagged that part of the surge may be pulled forward. If so, early-2027 loan growth will slow, and the NII bridge will lean more heavily on repricing.

2027 NII: How Much Does Growth Slow?

The most consequential exchange came when one questioner calculated roughly 10% NII growth ex FirstBank in 2026 against a Street forecast near 6.5% for 2027, and asked whether growth would really slow that much. The CFO, while declining formal guidance, effectively endorsed the Street's dollar range and tied it to repricing that is already contracted.

Q: "And correct me if I am wrong, Robert, but I am calculating something like 10% ex FirstBank net interest income growth for 2026. The street is at about 6.5%." … "You know, should the rate of NII growth slow down by that much in 2027?"
— Erika Najarian, UBS

A: "So when we look at it, clearly in 2027, we will exceed in net interest income our annualized run rate of what we project to finish 2026 in. And when we look at the consensus in terms of where you are, you sort of hit it-- that is a billion to 1.1 billion above 2026 in terms of growth. And I think we would agree with that range."
— Rob Reilly, EVP & CFO

Assessment: The single most valuable sentence of the call for valuation. A $1.0–1.1B NII increase is roughly $0.8B after tax, or about $2 per share, before fee growth or cost savings. It gives the 2027 consensus near $21.4–21.5 a mechanical floor and is a key reason we are comfortable paying ~10.5x for it.

Deposit Betas in the New Hiking Cycle

With the Fed hiking again, a question asked whether deposit betas would behave differently from the 2022–2023 cycle. Management said the last cycle was distorted by the migration of excess noninterest-bearing balances and that this one should look like a normal cycle, with betas higher on corporate balances and lower on lower-end consumer balances.

Q: "Can you talk about some of the deposit beta dynamics you expect from here? Just given that the rate paid on core deposit was up, we are going to get further Fed tightening. Anything different going on with the start of this rate hike cycle relative to the prior 1?"
— Manan Gosalia, Morgan Stanley

A: "Set that 1 aside in terms of typical rising rate environment We will see increased betas on the corporate side for sure on the wealth side less so on the lower end consumer retail side, but I think we will settle out in some sort of cumulative beta in the mid-50s like we have done in the past."
— Rob Reilly, EVP & CFO

Assessment: A mid-50s cumulative beta is a sensible base case and consistent with PNC's history. The open question is mix: growth is being funded disproportionately by high-beta corporate deposits and brokered CDs while retail balances fell, so the blended beta could run above the mid-50s even if each category behaves normally.

CET1 Including AOCI, and the Buyback

A capital question asked what CET1 would be with AOCI included, now that the company no longer shows it, and how long the buyback can continue with ratios below target. The CFO said CET1 including AOCI would be in the low nines, still above 9%, and called the AOCI-inclusive ratio stale given the expected Basel III changes.

Q: "With the AOCI move, it is no longer in your materials, but do you have what the CET1 was, including AOCI?" … "You are obviously generating a lot of organic capital. And the push and pull between where your capital ratios are and your ability to continue that type of buyback?"
— Ken Usdin, Autonomous Research

A: "So we feel good about our capital position. We rounded down to 9.8 Our target is 10. And that was simply a result of loans coming in higher than we expected, which drove up RWA." … "If we did that, you know, we dropped to low nines, but certainly above 9 and still a pretty healthy TCE ratio."
— Rob Reilly, EVP & CFO

Assessment: Adequate, not abundant. Loan growth is consuming capital faster than expected, the buyback has already stepped down ~$50M a quarter, and the AOCI-inclusive ratio sits closer to 9% than 10%. Capital is not a constraint on the dividend, but it limits buyback upside until the Basel III relief arrives.

How Much of the Loan Growth Is Share, and How Much Is AI Capex?

A multi-part question asked management to split commercial loan growth between industry growth and share gains, and between AI-related capital spending and everything else. The CEO said the excess growth in the expansion markets is pure share, declined to size the AI effect, and then volunteered the downside case.

Q: "So for every $100 of additional commercial loan growth, how much would you say is additional industry growth versus market share?" … "And then how much of it would you say is kind of AI CapEx related versus non AI CapEx related or other?"
— Mike Mayo, Wells Fargo

A: "So you know, take that you know, all that excess growth rate is pure share. Is 1 way to think of it." … "So if the CapEx broadly slows down, related to AI. Do we think even though we do not lend directly to that, that it would slow loan growth? Yes."
— Bill Demchak, Chairman & CEO

Assessment: The candid "yes" is the most important risk disclosure of the call. PNC's growth is part share gain (durable) and part a strong capex economy (cyclical), and management will not split the two. Our thesis rests on the share-gain part. The capex part is why we model loan growth slowing toward mid-single digits in 2027.

The Large Commercial Credit Behind the NPL Increase

A credit question asked which industry the large new nonperforming credit sits in and what the broader outlook is. The CFO named the service-provider category, called the problem company-specific, and pointed to the office loans as already reserved.

Q: "You mentioned about the NPL increase, CRE office and 1 large commercial credit. If you are able to share or willing to share what industry that large commercial credit in and your outlook from here on credit."
— David Chiaverini, Jefferies

A: "And on the 1 large credit, it is in the service providers, you can see that in our detail of loans. Nothing particularly thematic about it, just a company that is hit some cash flow issues." … "But, you know, as we had said for years, that will flow through non performers and ultimately charge offs. But, again, that is all been paid for."
— Rob Reilly, EVP & CFO

Assessment: The supplement supports the answer: service-provider NPLs rose by ~$138M, which matches one large name, and the office pull-through is visible in CRE. What the answer did not cover is why the allowance ratio fell while NPLs rose. That combination is acceptable once; in a hiking cycle it would not be acceptable twice.

AI Agents and the Future of Deposit Competition

A question asked whether AI agents that move deposits automatically will raise funding costs across the industry. The CEO rejected the idea that third-party agents will transact inside bank accounts in the near term, but accepted that deposit movement will keep getting more efficient and that PNC must match the technology.

Q: "On these third quarter calls about the AI agents moving deposits around and forcing deposit costs up for the industry. what is your take on that?"
— Gerard Cassidy, RBC Capital Markets

A: "The current attempts at this with some of the new agentic applications that have come out where you know, frankly, try to come in and transact on your behalf. Inside of your accounts. Incredibly unsafe, not going to happen. Bad idea." … "And to win in that period of time, we are gonna have to offer our clients technology that is as good as these third party providers so we can allow them to do what they want."
— Bill Demchak, Chairman & CEO

Assessment: Forceful, and right on the near term: liability and authentication rules make autonomous agents moving consumer money a slow-moving threat. The structural point he conceded matters more for a deposit-funded bank: cash will get cheaper to move, so the deposit franchise has to be defended with product, not inertia.

What They're NOT Saying

1. A NIM number: Management moved from being "on record" for 3% to calling NIM "an outcome" and 3% "a calculation." The metric the Street keys on is now explicitly unguided, at the moment a hiking cycle makes it harder to forecast.
2. Why retail deposits fell: Average Retail Banking deposits fell 2% sequentially on lower consumer interest-bearing and noninterest-bearing balances. Neither the prepared remarks nor the Q&A addressed it, though it is the most rate-sensitive funding question for the year ahead.
3. The reserve logic: The allowance ratio fell to 1.42% while NPLs rose 8% and new nonperforming inflows doubled. Management did not explain the reserve methodology beyond the student-loan transfer and the previously reserved office book.
4. A 20% ROTCE path, or a 2027 outlook beyond NII: The 2027 NII bridge was endorsed, and the CFO agreed that the Q4 capital-markets expense step-up is not a read-through to 2027 expense growth. Beyond that there was no 2027 fee, expense or ROTCE framework, and no update on the previously discussed path toward a 20% ROTCE. With ROTCE already above 18%, the next return target is open.
5. Basel III timing and the AOCI-inclusive ratio: The capital answer ("low nines", Basel III "just around the corner or somewhere along those lines") left both the timing and the size of the net effect vague.
6. The escrow shortfall: Management cited missing escrow balances as one reason noninterest-bearing deposits underperformed, but did not size them or say whether they return in Q4.
7. The cost of the AI program: The CEO pointed to "creeping depreciation" from data-center rebuilds and said PNC has been "purposefully quiet." There is no spend figure, timeline or savings target to model.

Market Reaction

• Pre-print setup: PNC closed at $215.80 on October 14, up only +3.4% year to date against +12.8% for the S&P 500, down 10.4% over the trailing thirty days and up 18.4% over twelve months. It sat 16.0% below its 52-week closing high in a range of $178.37–$256.97. The thirty-day slide included a 3.9% drop on September 16, the day of the Fed's hike ($240.78 to $231.49). The stock entered the print at almost exactly 2.0x tangible book of $107.64 per share.
• Reaction day (October 15, before-open report): The stock gapped down to open at $208.49 (−3.4%), traded a range of $208.49–$218.90 during the call, and closed at $214.94, down −0.4% (−$0.86). Volume was 4.6M shares, 2.5x the 1.8M thirty-day average. The S&P 500 rose +0.3% and the regional-bank ETF rose +1.8%, so PNC lagged both on the day.
• Post-print sessions: By the October 23 close the stock had recovered to $224.46, +4.4% from the print-day close, while the S&P 500 fell 1.6% over the same six sessions and the regional-bank ETF rose 3.9%.
• Since our July downgrade: From the $254.15 July close at which we moved to Hold, PNC fell 15.1% to the pre-print close, while the S&P 500's year-to-date gain widened from +10.2% to +12.8%.

Gap-and-Recover: the Open Priced the Miss, the Close Priced the Guide. The 3.4% gap at the open was a direct read of the headline: revenue, NII and NIM all below the Street, and a margin that fell when it was expected to rise. The recovery through the call, to a nearly flat close on 2.5x normal volume, came as the raised guide, the Q4 capital-markets rebound and the 2027 NII endorsement landed. That is the right way to read the quarter. The headline quarter was soft at the top line, and the forward picture was better than expected.

Macro, Not Micro: the De-Rating Came Before the Print. Most of the damage happened in the month before the report, and it was sector-wide. Large-cap bank stocks fell into correction as the Fed restarted hikes and long Treasury yields rose sharply, with investors weighing slower lending, higher credit costs and higher funding costs. PNC fell more than the regional-bank ETF over the thirty days into the print (−10.4% vs. −7.4%), which is consistent with a stock that had carried the fullest valuation in the group. The post-print rebound, against a falling S&P 500, suggests that the market saw the print as confirmation that earnings power held up through the rate shock. Our July Hold was directionally right but too generous: the stock trailed the index by roughly 17 points over three months. The same de-rating is what reopens the upside now.

Street Perspective

Debate: Is the New Hiking Cycle a Tailwind or a Headwind for PNC?

Bull view: The bull case is that PNC built a short-rate-neutral balance sheet for exactly this environment. Higher long rates enlarge the fixed-rate repricing tailwind, management has endorsed $1.0–1.1B of 2027 NII growth, and the AOCI hit is a mark that accretes back to par without touching regulatory capital.

Bear view: The bear case is that hiking cycles hurt banks with a lag. Deposit costs are already rising, noninterest-bearing balances are shrinking, tangible book just fell 3%, and the credit cost of tighter policy has not yet arrived. On this view the September sell-off is the start of a de-rating, not the end of one.

Our take: Mostly bull on earnings, partly bear on the balance sheet. The NII bridge is contractual enough to trust, and it is the bigger number. The bear case is right that tangible book and credit are now the swing factors, which is why we add a rate-regime bear point and keep conviction moderate. At ~10.5x 2027 earnings, a good part of the bear case is already in the price.

Debate: Is the NIM Miss a Spread Problem or a Mix Choice?

Bull view: The bulls argue the margin is being diluted on purpose. High-grade corporate loans arrive with syndication, derivatives and treasury-management fees, so the full relationship return is higher than the spread suggests. Half the loan growth landed in the last two weeks, so Q4 NII catches up.

Bear view: The skeptics note that the average spread on the book fell ~3bp, partly because higher-yielding loans paid off, and that management has moved from "on record" for 3% to "it is an outcome." They read that as competitive spread compression wearing a mix label.

Our take: More mix than problem, but not only mix. Fee attachment is real (capital markets +40% year to date), and NII dollars are guided up 4% in Q4. Still, the shift in management's language is a signal, and we no longer give PNC credit for a 3%-plus NIM in our base case. The thesis is a NII-dollars story, and the dollars are intact.

Debate: Early Credit Cracks or Idiosyncratic Noise?

Bull view: The bull reading is that one large service-provider credit and previously reserved office loans explain the entire NPL increase, delinquencies fell to 0.34%, and underlying charge-offs were flat. PNC's book is skewed to high-grade borrowers, and its stress-test losses are the lowest in its peer group.

Bear view: The bear reading is that NPLs rose 8% and new nonperforming inflows doubled in the first quarter of a hiking cycle, while the allowance ratio fell. Banks rarely see credit turn all at once; it starts with "one large credit."

Our take: Idiosyncratic for now, with the burden of proof on the next quarter. The data support management's explanation, but the falling reserve ratio is a poor match for the macro direction. A second quarter of rising NPLs, or a Q4 charge-off figure well above the ~$250M guide, would move us toward the bear camp.

Debate: Has the De-Rating Created an Entry Point?

Bull view: The bulls point out that the stock is down ~12% from July while 2026 earnings estimates rose and the 2027 NII path firmed up, leaving it near 2.1x tangible book and ~10.5x 2027 earnings, a modest multiple for an 18%-ROTCE franchise.

Bear view: The bears argue that tangible book is the anchor and it is shrinking, that further rate increases would extend the AOCI drawdown, and that bank multiples do not usually re-rate higher while the Fed is still hiking.

Our take: Entry point. Our July downgrade was purely about price, and the price is now ~12% lower with better earnings visibility. On our numbers the twelve-month base case returns ~10–17% including the dividend, ahead of a mid-to-high-single-digit market expectation. The bears are right that the path may be bumpy while the Fed hikes, and that is why this is a moderate-conviction Outperform, not a high-conviction one.

Model Update & Valuation Framework

Q3 raises our earnings estimates, lowers our tangible-book path, and resets the price. Key revisions, at the October 23 close of $224.46:

Valuation framework. At $224.46, PNC trades at roughly 2.09x tangible book ($107.64), ~11.5x our ~$19.45 of 2026 adjusted EPS, ~10.5x our ~$21.50 of 2027 EPS, and a ~3.6% dividend yield on the $8.00 annualized payout. In July the stock traded at ~2.29x tangible book and ~13.4x forward earnings. The price-to-tangible-book multiple has fallen less than the P/E because tangible book itself fell. That is the honest complication of this upgrade: on book value, PNC is roughly where it was at our April Outperform (~2.0x), but on earnings it is about two turns cheaper than in July, with better 2027 visibility than it offered then. For a franchise earning an 18%+ ROTCE, 2.1x tangible book is a fair-to-cheap multiple, and ~10.5x forward earnings is a low multiple for a bank with contracted NII growth.

Price-target framework. Our base case compounds tangible book from $107.64 to roughly $117–120 over twelve months (about $2.9B of retained earnings after the dividend and a ~$560M quarterly buyback, plus modest AOCI accretion, on ~389M shares) and applies 2.0–2.15x to the ~$119 midpoint, for a 12-month value of roughly $240–255. That is ~11.2–11.9x our 2027 EPS, and +6% to +14% from $224.46. With the 3.6% dividend, the base-case total return is roughly +10% to +17%, ahead of the mid-to-high-single-digit return we expect from the S&P 500. A bull case (2.3x on ~$120 of tangible book as long yields stabilize, Basel III relief frees buyback capacity, and loan growth holds above GDP) supports $270–280 (+20% to +25%). A bear case (1.70–1.75x on ~$115–117 of tangible book if yields keep rising, retail deposits keep leaking and credit migrates) implies $195–205 (−9% to −13%). On price alone, base-case upside (+6% to +14%) and bear-case downside (−9% to −13%) are roughly balanced. The dividend tips it: in total-return terms the base case is +10% to +17% against −5% to −9% in the bear case, about 1.8 to 1 at the midpoints, and bull versus bear is about 2 to 1 on price. That is enough for an Outperform, and it is also why conviction is moderate rather than high.

Thesis Scorecard Post-Earnings

This scorecard grades the same standing thesis carried since initiation. The pattern of Q3 is the mirror image of Q2. In July every fundamental pillar was green and valuation was the binding constraint. In October the binding constraint has been relieved by a 15% fall in the price, two fundamental pillars took a rate-related hit (tangible book, and to a lesser degree capital), and a new bear point (the hiking cycle itself) enters the scorecard.

Overall: The thesis is intact on earnings power and weaker on the balance sheet. Three of five bull pillars strengthened or held cleanly (repricing, loan growth, operating leverage). The tangible-book pillar moved to at risk for a reason outside PNC's control, and capital is adequate rather than abundant. The binding bear point from July, valuation, has been relieved by the price. The new rate-regime bear point is the one to watch, and it explains both the de-rating and our moderate conviction. Net, the gap between business quality and price that closed in July has reopened, though with more macro risk than at initiation.

Action: Upgrade to Outperform from Hold. In July we stepped aside on price, not on the franchise, and named the signposts that would bring us back: a pullback toward ~2.0x tangible book, durable capital-markets strength, a quantified 2027 bridge, or Basel III relief. Two are substantially met (the stock entered the print at 2.0x tangible book, and the CFO endorsed a $1.0–1.1B 2027 NII bridge, if not the full ROTCE bridge we asked for), one is partly met (capital markets normalized but are guided up 15% in Q4), and one is pending (Basel III). Meanwhile 2026 earnings moved up to ~$19.45 and 2027 looks like ~$21.50, so the stock is two turns cheaper on earnings than at our downgrade. We expect a ~10–17% twelve-month total return in our base case. What would move us back to Hold: a second quarter of rising NPLs with a falling allowance ratio; retail deposit outflows that push the blended beta well above the mid-50s; a Q4 NIM below 2.95% alongside NII below the +4% guide; or a 2027 outlook in January that falls short of the $1.0–1.1B NII bridge. What would move us to Underperform: a genuine credit turn (charge-offs sustainably above 40bp), a capital shortfall that forces a buyback suspension, or a further rate shock that cuts tangible book by another 5% or more.

Signposts for Q4 2026 earnings (January 2027):