Record Print, Backlog $39.8B All-Time High, Prime Power Data-Center Pipeline Opens — Upgrading to Outperform

Rating Action

This print drives an upgrade to Outperform. We initiated at Hold at Q2 2025 acknowledging the Power Generation pull as real but largely priced; the change is that Q3 reveals the data-center prime-power pipeline opening explicitly (named anchor customers Joule and Hunt), Solar Turbine lead times extending, sales-to-users accelerating into double digits in E&T across all applications, and the FY 2025 algorithm now showing the operating leverage flowing through despite a worsening tariff drag.

• Q2 2025 (Initiating at Hold): Best-in-class cyclical industrial leader, real Power Gen pull, but the multiple already reflected the data-center narrative; tariff trajectory and CI cyclical reset offset.
• Q3 2025 (Upgrading to Outperform): The composition of the print re-rates the franchise. Specifically: (a) prime-power data-center pipeline opens with named anchors (Joule, Hunt) — the addressable market is broader than reciprocating-engine backup; (b) Solar Turbines lead times extending in the Titan 250 / 350 segment; (c) sales-to-users acceleration is broad — all three primary segments, all regions ex-APAC, all E&T applications growing 20%+; (d) backlog +39% YoY at $39.8B with sequential growth; (e) FY 2025 raised again, ME&T FCF guide moved up; (f) operating leverage is coming through despite peak Q4 tariff drag. The upgrade is also informed by Bonfield's Q2 framing — "good momentum into 2026, operating leverage" — which Q3 confirms is mechanically realizing.

Results vs. Consensus

A genuine double-beat against a Street that had been skeptical on tariff drag. Composition is the story: revenue beat by ~$800M on E&T volume + STUs strength; EPS beat by ~$0.40 despite a ~$0.18 tax-rate headwind from the recently enacted U.S. tax legislation (capitalization of R&D becoming immediate expensing — cash benefit, book negative).

Segment Performance

Construction Industries (CI) — Acceleration With STUs +7%

• Sales: $6.8B, +7% YoY.
• Segment operating profit: $1.4B, -7% YoY; segment margin 20.4% (-300bps YoY).
• Tariff impact: ~340bps drag on margin; CI bears ~55% of the Q4 tariff impact.
• Regional sales: NA +8%, EAME +6%, APAC +3%, LatAm -1%.
• Sales-to-users: NA +11% (better than expected on resi + non-resi); EAME up on Africa/Middle East; APAC down on softness in subregions; LatAm slightly below expectations. Dealer rental fleet loading was down slightly but dealer rental revenue continued to grow. Q4 fleet loading expected to increase YoY.
• Margin ex-tariffs: Slightly higher than the prior year and about in line with expectations.
• Q4 setup: Strong sales increase YoY (volume + dealer-inventory tailwind absent the prior-year $1.6B destock); ~flat price; margin ex-tariffs higher YoY; incl. tariffs lower YoY.

Resource Industries (RI) — STUs Inflect Positive on Mining Truck Timing

• Sales: $3.1B, +2% YoY.
• Segment operating profit: $499M, -19% YoY; segment margin 16.0% (-430bps YoY).
• Tariff impact: ~260bps drag on margin; RI bears ~20% of the Q4 tariff impact.
• Sales-to-users: +6%, slightly above expectations on timing of large mining truck and off-highway truck deliveries (pulled forward from Q4). Mining better-than-expected; heavy construction + quarry/aggregates in line.
• Order book: Healthy orders for large mining trucks, articulated trucks, and large track-type tractors. Coal price weakness continues to drive parked trucks and lower rebuild activity. Per Bonfield, order strength is in large mining trucks; gold + copper prices high, coal weak (especially Indonesia).
• Autonomous: Continued customer acceptance growing; more detail expected at the November 4 Investor Day on RPM (an autonomous platform reference).
• Q4 setup: Stronger sales YoY on volume (dealer inventory tailwind), but unfavorable STUs given the Q3 timing pull-forward. Price slightly less unfavorable than Q3; margin incl. tariffs lower YoY.

Energy & Transportation (E&T) — The Engine Continues to Inflect

• Sales: $8.4B, +17% YoY — CAT's largest segment by sales and profit, getting larger.
• Segment operating profit: $1.7B, +17% YoY; segment margin 20.0% (+10bps YoY).
• Tariff impact: ~140bps drag on margin; E&T bears ~25% of the Q4 tariff impact.
• Application breakdown (sales):
  • Power Generation: +31% YoY. Reciprocating engines for data-center applications + turbines and turbine-related services. STUs in PG +33%.
  • Oil & Gas: +20% YoY. Solar Turbines + turbine services + gas-transmission demand. STUs +20%.
  • Industrial: +5% YoY (sales) / +20% STUs. Electric power applications driving growth from a low base.
  • Transportation: +5% YoY (sales) / +20% STUs. International locomotive deliveries.
• Solar Turbines — lead times extending: Larger Titan 250 / Titan 350 lead times extending; smaller turbines still respond faster. Capacity is being evaluated but not yet announced — CAT able to meet current order pace. Per Creed, "almost unprecedented interest" continues.
• Prime power inflection: Joule Capital Partners and Hunt Energy Company explicitly named as new prime-power customers. Per Creed, prime power is "a great opportunity ... it creates services opportunity as we move forward as well." Capacity additions partly going to serve prime power. Lead times "pretty extended" for data-center demand — planning years in advance with hyperscalers and colos.
• Margin ex-tariffs: Higher YoY and slightly stronger than expected; price realization remained favorable (E&T able to take regular price increases in a constrained-capacity environment).
• Q4 setup: Strong YoY sales (Power Gen + O&G); favorable price realization; margin ex-tariffs higher YoY; incl. tariffs slightly lower YoY. Sequential Q4 vs. Q3 growth slightly lower than typical seasonality on Q3 strength.

Financial Products (Cat Financial) — Best Q3 Past-Dues in 25+ Years

• Revenues: ~$1.1B, +4% YoY.
• Segment profit: $241M, -2% YoY; higher provision for credit losses + higher SG&A + unfavorable equity securities at Insurance Services.
• Credit quality: Past-dues 1.47% (-27bps YoY) — the lowest third quarter in over 25 years. Allowance rate 0.89%, near historic lows.
• Activity: Retail credit applications +16% YoY; retail new business volume +7% YoY.
• Read-through: The merchandising programs are still pulling demand; credit quality is exceptional; the slight margin compression is on insurance equity securities + a reserve build, not credit deterioration.

Key Topics & Management Commentary

Prime Power Data Centers: Pipeline Opens with Named Anchors

The single most thesis-relevant commentary of the call. The Power Gen story has been multi-year and broadly understood; what changes at Q3 is the explicit opening of the prime-power pipeline — data centers buying Solar Turbines for grid-independent prime power, not just reciprocating engines for backup:

“We're definitely really excited about the prime power opportunity with data centers and more broadly, just the demand for power that data centers and broader trends in the industry are putting on to the grid. We're going to see a lot more of this, I believe. Prime power is a great opportunity for us because it creates services opportunity as we move forward as well. ... With Solar, we're seeing a lot of ordering activity, and it's really healthy, as you suggest. ... Lead times are starting to get a little more extended at Solar.”
— Joe Creed, CEO

Two other operative details: (a) Joule Capital Partners and Hunt Energy Company are explicitly named in the prepared remarks; (b) Solar's services profile is direct (CAT field technicians, not dealer-routed), so prime-power Solar wins translate into direct services growth on CAT's P&L rather than the dealer's. The implication: every prime-power Solar unit is a multi-decade services annuity for CAT directly. That is the long-cycle algorithm change worth re-rating on.

Capacity: Lafayette Recip + Solar — Both Ramping

Reciprocating engine capacity at Lafayette is "coming on" per Creed; more units shipped in Q3 vs. Q2 as capacity ramps and supply-base coordination improves. On Solar, Bonfield was direct that capacity decisions will be made when needed, current order pace can be met. The signal: CAT is not yet reaching for capacity beyond the current expansion plan; longer lead times are partly a planned consequence of demand running ahead of conservative expansion. We read this as constructive — CAT is sized for the current order book and adding capacity in measured tranches rather than gold-plating.

Backlog Extension — "I'd Like to Try to At Least Get That Stabilized or Brought Back In"

Jamie Cook (Truist) asked the most analytically rich question on backlog dynamics: should the Street model continued backlog growth, or is the backlog approaching a ceiling as capacity comes online? Creed's answer is the most thesis-relevant strategic comment of the call:

“We continue to see significant demand, particularly in E&T. ... We talk to our large data center customers, both hyperscalers and colos, frequently, monthly mostly. ... So we have great confidence in the pipeline that's out there, and that's why we're putting the capacity in. ... I'd like to maybe if the reason the backlog goes up a little less is that I'm able to get out more and more product and keep up, I think that's a good thing. ... We are getting pretty extended on some of our products in E&T, and I'd like to try to at least get that stabilized or brought back in if we can.”
— Joe Creed, CEO

The signal: backlog moderation in the future is not a demand concern — it's a capacity success. We model continued backlog levels at $35–40B through 2026 as throughput catches up.

2026 Algorithm: Tariff and Tax Setup

Bonfield walked the 2026 puts-and-takes carefully in response to David Raso's (Evercore ISI) framing question: (a) price headwind laps starting Q4 — "we don't expect price to have an impact on the fourth quarter" — positive for 2026; (b) tariffs remain a headwind in 2026 magnitude TBD; (c) the tax rate increase to 24% in 2025 from R&D capitalization changing to immediate expensing is a 2025 book-tax negative but a 2025 cash-flow positive; the 2026 tax rate is expected to benefit from foreign tax adjustments and be lower than 2025. The framework hint: 2026 algorithm is set up for operating leverage as price laps, capacity comes online, and tax rate normalizes — tariff trajectory is the principal swing factor.

Tariff Mitigation: "All Factors on the Table, Including Price"

Mig Dobre (Baird) pressed on whether CI's tariff drag could be offset by 2026 pricing. Creed's framing was the same as Q2 — long-term sourcing/footprint moves require certainty; "no regrets" mitigation only at the margin. But Bonfield's add for the first time put pricing explicitly on the table:

“The timing of mitigation actions, as Joe mentioned, will really be determined by the greater degree of certainty that we get out there. Obviously, all factors are on the table, and that will include potential for price.”
— Andrew Bonfield, CFO

The implication: 2026 will see CAT lean on price as part of the tariff offset, especially in segments where merchandising lap creates pricing room (CI in particular). Whether the unilateral pricing power exists at the margin is a competitive question we cannot answer ahead of the cycle — but Cat's brand, dealer network, and service depth argue for material pricing latitude.

FY 2025 Guidance & Outlook

Q4 specifics: Strong YoY sales growth across all three primary segments. Machine dealer inventory expected to decline slightly Q4 (vs. -$1.6B prior-year) — a meaningful YoY tailwind. Enterprise price ~flat. Q4 net tariff cost $650–800M (peak). Margin ex-tariffs higher YoY; incl. tariffs lower YoY. CI: strong sales growth, ex-tariff margin higher, incl-tariff margin lower (CI bears 55% of Q4 tariff impact). RI: stronger volume on inventory, unfavorable STUs from Q3 timing pull-forward. E&T: strong PG + O&G growth; favorable price; ex-tariff margin higher YoY.

Analyst Q&A — Notable Exchanges

Q&A was thesis-pivotal this quarter, with the strongest single thread on Power Gen prime-power capacity and the second strongest on backlog dynamics. Notable threads:

• Kyle Menges (Citi) opened on the prime power data center opportunity. Drew the explicit Joule reference, the Solar Turbines lead-time extension, and the Titan 250 / 350 segmentation. The most thesis-relevant question of the call.
• Angel Castillo Malpica (Morgan Stanley) probed E&T price and pull-through margin (whether Power Gen incremental margins sit above the typical 30% E&T pull-through). Bonfield deflected to the enterprise margin target framework but acknowledged Power Gen volume was the principal margin driver in Q3 and that target margin pull-through is over 30% across the range.
• David Raso (Evercore ISI) pressed for 2026 puts-and-takes — price flat in Q4 implications, capacity efficiency, mix, tax rate. Bonfield politely deflected to January but laid out the directional algorithm: price flat is positive; tariffs still a headwind; tax rate normalizes lower; foreign tax adjustments positive.
• Tami Zakaria (JPMorgan) on the "light switch" STUs acceleration in every segment ex-APAC. Creed: E&T capacity ramping, RI mining-truck delivery timing, CI merchandising programs in NA. Read: the broad STUs strength is a real demand step-up, partly amplified by capacity capture in E&T.
• Mig Dobre (Baird) on CI demand sustainability and 2026 tariff mitigation. Drew Bonfield's first explicit "potential for price" on tariffs. Creed: sourcing/footprint moves require certainty; CAT's heavy U.S. footprint argues against rapid relocation.
• Robert Wertheimer (Melius) on RI commodity exposures and autonomy. Bonfield: large mining trucks the area of strength; gold + copper high, coal a drag (Indonesia); autonomy acceptance growing — more detail at Investor Day on RPM.
• Kristen Owen (Oppenheimer) on backlog composition (E&T vs. CI/RI sequential) and Solar Turbines accounting (data center prime power = power generation segment, not oil & gas). Useful disclosure that prime-power Solar = power generation in the segment reporting.
• Michael Feniger (Bank of America) on Solar Turbines revenue scale — CAT declined to disclose. Bonfield: oil & gas applications doing pretty well, gas transmission orders strong, the historical 2010 framework was offshore-heavy and the world has moved on with shale.
• Jamie Cook (Truist) on backlog growth ceiling and Q4 tariff peak. Drew Creed's most strategic comment of the call — backlog moderation is a capacity success, not a demand concern; CAT is "getting pretty extended" and would prefer to bring backlog back in. Tariff peak in Q4; mitigation actions including price are on the table.
• Stephen Volkmann (Jefferies) closed on Solar's direct services model — the most under-appreciated structural tailwind of the call. Solar units run continuously, CAT services them directly (not via dealers), so every prime-power Solar win is a multi-decade direct services annuity for CAT.

What They’re NOT Saying

• No 2026 framework yet. Multiple analysts pressed; Creed and Bonfield both deferred to January (Q4 print) and to the November 4 Investor Day. The implicit signal: framework is being built in real time as the order book extends.
• No quantification of prime-power data-center backlog. Creed named Joule and Hunt but did not size them. The buyside-natural question is whether prime power is now 5%, 15%, or 30% of the E&T backlog. Investor Day natural setting for this disclosure.
• No specific Solar capacity announcement. Capacity expansion is being evaluated but not committed. The hint: CAT is sized for the current order book; longer lead times are a planned consequence rather than a constraint signal. Investor Day may surface the next capacity tranche.
• No quantification of pricing latitude in 2026. Bonfield explicitly opened the door to price as a tariff-mitigation lever — first time on this call series. But the magnitude was not sized. We model 50–150bps of incremental price across 2026 as the lap completes; this is the principal upside lever to operating margin.
• No autonomous truck commercial roadmap. "Continued customer acceptance" qualitative framing; Bonfield deferred to Investor Day for RPM detail. Autonomy is not yet in the model contribution but represents long-cycle multi-year optionality on RI.
• No discussion of non-recip / non-Solar power generation. The conversation centered on engines and turbines; no commentary on, e.g., battery storage tied to data-center prime power. CAT is a powertrain company — the boundaries of where power-gen demand pulls outside its current product portfolio are not yet articulated.

Market Reaction

The print landed before the open on October 29, 2025. The stock opened sharply higher on the double beat, the FY 2025 raise, the record backlog, and the explicit prime-power data-center disclosure. Press coverage cited a high-single-digit to low-double-digit intraday rally, with the stock outperforming the broader industrial complex on the day. Volume was elevated; multiple sell-side desks moved to "Buy" or raised price targets in the days following. The market reaction was, in our read, an appropriate rerating on the prime-power disclosure plus the Q4 setup — not a parabolic move requiring contrarian sizing.

Street Perspective

The bull case being made on the Street post-print converges on four planks: (1) prime-power data-center pipeline opening explicitly with named anchors materially expands the addressable Power Gen revenue pool; (2) Solar Turbines lead-time extension + Lafayette recip ramp imply visible capacity-bound revenue through 2027; (3) sales-to-users acceleration across all three segments and most regions confirms broad cyclical pull, not just E&T-narrow; (4) the FY 2025 raise + Q4 setup despite peak tariff cost demonstrate operating leverage realizing through the algorithm. The November 4 Investor Day is positioned to crystallize a multi-year framework that supports the rerating.

The bear case being articulated centers on: (1) the multiple has now expanded to fully reflect prime-power optionality — further upside requires Investor Day delivering above-consensus framework; (2) tariff drag at $1.6–1.75B for 2025 with Q4 peak signals durable margin compression that pricing alone may not offset; (3) RI segment margin -430bps YoY exposes the cyclical reset risk if commodity prices weaken further or autonomous-truck commercial timeline slips; (4) the Q3 STUs acceleration in RI is partly a Q4-pull-forward, not a sustainable trajectory.

Our read sides with the bull framing on (1), (2), and (3) and treats the Investor Day risk and tariff risk as appropriately priced into the Outperform call. We treat (4) as scoped — STUs in RI are choppy quarter-to-quarter but the order book direction is clear.

Model Implications

• FY 2025 revenue: The trajectory implied by 9-month + the FY raise supports $66–67B FY 2025 sales & revenues, with E&T pulling and CI/RI now stabilizing.
• FY 2025 adjusted operating margin: Near bottom of target range (high-teens) inclusive of $1.6–1.75B net tariff drag.
• FY 2025 ME&T FCF: Above $7.5B as guided; supportive of $1.0–1.5B/Q capital return cadence.
• FY 2026 framework (preliminary — we update post-Investor Day): Mid-single to high-single-digit revenue growth driven by E&T capacity ramp + CI margin recapture as price laps. Margin recovery contingent on tariff trajectory: assuming flat-to-down tariff drag, FY 2026 enterprise adjusted operating margin moves into the upper half of target range. Tax rate benefits from 2025 base (24% ⇒ 22-23% in 2026 on foreign tax adjustments).
• Backlog conversion: Record $39.8B with E&T weighted; we model 1H 2026 visibility as essentially booked, 2H 2026 ~75% covered as of today.
• Capital return: We model continued $1.0–1.5B/Q capital return through 2026 with potential pace-up if Investor Day articulates structurally higher FCF generation post-2026.
• Prime power: We add an upside scenario for prime-power Solar contributing $0.5–1.0B of incremental annual revenue by 2027 with a multi-year direct services attach. This is not yet in our base case — we update post-Investor Day.

Thesis Scorecard

Overall: Thesis materially strengthens. Four of five bull pillars confirmed plus; the new fifth pillar (prime-power data-center inflection) adds a new addressable market. Bear case scoped — tariff drag has worsened but pricing latitude is now explicitly acknowledged; Power Gen multiple-already-priced concern is receding on prime-power expansion; CI cyclical reset is completing.

Action: Upgrading to Outperform on Caterpillar (CAT). The November 4 Investor Day is the principal near-term catalyst — expect framework articulation around (a) Power Gen + prime-power capacity through 2027, (b) multi-year revenue/margin algorithm, (c) RI autonomous and electric truck roadmap. We do not chase parabolic moves on the print but underwrite a 12-month total return above the S&P 500 from current levels.

Net: A genuine inflection quarter. Record sales, record backlog, prime-power data-center pipeline opens, and operating leverage realizing despite peak tariff drag. Upgrading to Outperform.