International Paper Company (NYSE: IP) — The Mega-Merger It Is Already Unwinding, Priced for the Guide With the CEO Buying the Lows

Independent equity research — published 2026-06-27. Initiation.

⚡ Author’s Take

This block is the author’s own subjective opinion and general information only. It is not investment advice. The analysis that follows (Sections 1–15) deliberately carries no recommendation and no price target; nothing below this block should be read as one.

Verdict: HOLD / accumulate-on-weakness for the patient contrarian — a commodity turnaround, not a quality compounder, and not a short. Accumulation zone roughly $30–36 (at/below the ~$27.85 book and the May washout); the stock at $38.76 is fair-to-slightly-cheap, pricing the FY26 guide with a free option on the 2027 ramp. Fair-value zone ~$45–55; bull case ~$60–65 if the back-loaded $5B-EBITDA target lands. Conviction: medium.

International Paper is two stories wearing one ticker. The first is a twenty-year record of value destruction: a serial acquirer of commodity containerboard capacity whose return on invested capital has sat below its cost of capital across an entire cycle — management’s own long-term plan scored three-year ROIC at 7.55% against a 9.0% target. The second is a genuine inflection: a Danaher-school operator, Andy Silvernail, importing the “80/20” system, winning box volume ~3 points faster than the industry, ripping out $700M+ of cost, and — most tellingly — buying $2M of his own stock at $40 the day after announcing a strategic split, alongside four directors at the lows. The market is paying for neither pattern decisively: at ~8.7x the cut FY26 EBITDA guide it capitalizes exactly what management is currently delivering and almost nothing for the $1.6B EBITDA bridge to 2027. The framing is contrarian-value with event-driven optionality — an abandoned dividend/value cyclical stabilizing off a multi-year washout (no momentum or growth factor loading; CEO-led insider conviction; a constructive +$100/ton 2026 price cycle). What flips me bullish: a clean fourth quarter that holds the FY26 guide and the EMEA segment turning profit-positive — proof 80/20 is structural, not cyclical lipstick. What flips me bearish: a fourth straight guidance miss or a dividend cut executed from weakness rather than choice, which would confirm the value-trap read. The single most honest thing about IP today is that the man running it bought the wreckage of his own mega-merger; the single most damning is that he bought a “global champion” for ~$9.9B and announced its break-up twelve months later. You are underwriting the operator, not the assets — and paying a fair, not a generous, price to do it. Tag: bought global scale, breaking it up a year later, and buying the dip on himself.

📈 Stock Price Action — Five-Year Event Map

Factual price history, not a recommendation. Price moves are FACT; attributed drivers are INTERPRETATION. No price target, no support/resistance levels.

Over five years IP has round-tripped a full commodity-and-deal cycle: a split-adjusted low of ~$25.50 (Jun-2023), a deal-euphoria high of ~$56.33 (Jan-2025), a slide to a washout low of ~$28.95 (19-May-2026), and a ~34% bounce to $38.76 today — roughly the midpoint of the range and ~31% below the high. The 52-week range is ~$29–58. The stock has reclaimed its 21-, 50- and 200-day moving averages off the May low. (Note: pre-2022 prices predate the Oct-2021 Sylvamo spin and are not strictly comparable to today’s pure-packaging entity.)

Cycle narrative. The 2021 Sylvamo spin reset IP as a packaging company; 2022–2023 was a textbook commodity de-stocking drawdown to a ~$29 trough. The 2024 recovery was deal-and-management driven — Silvernail’s arrival, Brazilian Suzano’s reported takeover approach, and the agreed DS Smith merger — cresting as DS Smith closed in January 2025. What followed was the painful part: across five quarters the stock halved as IP missed guidance three times, took a $2.5B impairment against the EMEA business it had just bought, cut its FY26 EBITDA outlook one quarter after setting it, and stoked fears the ~4.7% dividend is no longer covered. The January-2026 split announcement and the CEO’s next-day open-market purchase marked the sentiment low; the spring 2026 bounce tracks two containerboard price hikes (+$70 Mar-1, +$50 Jun-1) holding despite flattish demand, plus a visible insider-buying cluster. Each move ties to a documented event (earnings prints, 8-Ks, the published containerboard price benchmark, transcript guidance).

1. Executive Summary

International Paper is the largest fiber-based packaging company in the world and, on the evidence of two decades of financial results, one of the least efficient large-cap converters of capital into value. Following the January-2025 all-stock acquisition of UK-listed DS Smith plc and the January-2026 divestiture of its Global Cellulose Fibers (pulp) business, IP is now a ~$24B-revenue pure-play corrugated-packaging company spanning North America and EMEA, with ~65,000 employees and a new operating leadership team led by CEO Andrew Silvernail (ex-IDEX, May-2024) and CFO Lance Loeffler (ex-Halliburton).

The investment question is unusually clean. Containerboard is a commodity: fungible linerboard and medium, GDP-bound cyclical demand, administered pricing off the Fastmarkets/RISI benchmark, and returns that mean-revert toward the cost of capital for the average operator. IP has been the average-to-below-average operator: its consolidated ROIC peaked at just ~6.8% in 2021 and has otherwise sat in the 3–5% range — chronically below an ~8–9% WACC — even as best-in-class peer Packaging Corp (PKG) earns ~21% EBITDA margins and ~12% ROIC on identical products. The ~9-point margin gap to PKG, on the same end-product, is the central indictment: in this industry, being the biggest has not meant earning the most.

The bull case is entirely forward-looking and rests on three legs: (1) 80/20 — Silvernail’s IDEX/Danaher operating system, which is already showing real traction (NA box volume +2.5% vs an industry −0.3%, three straight quarters of ~3-point share outperformance; $710M of run-rate cost-out toward a >$1B target); (2) a constructive capital cycle — the industry removed ~10% of NA capacity in 2025 and pushed through ~$100/ton of 2026 price increases that are sticking; and (3) a planned break-up into a North American IP and an EMEA (legacy DS Smith) spin-off by late-2026/early-2027, intended to surface value the conglomerate structure obscures.

That last leg is also the thesis’s deepest scar. Announcing a split twelve months after closing a ~$9.9B (realized) transformative merger — and after writing off $2.5B of the EMEA goodwill in FY25 — is an explicit admission that the “global scale” rationale did not create the value underwritten. Set against that, the most credible bull datapoint is not a number management produced but one the insiders did: a cluster of open-market purchases at the lows, led by the CEO buying ~$2M at $40.

Financially, FY25 was a kitchen-sink trough: a GAAP net loss of −$3.5B (−$6.95) driven by the EMEA impairment, ~$958M of accelerated depreciation on mill closures, and restructuring; adjusted operating earnings were roughly break-even; and free cash flow was negative ~$159M, leaving the ~$1B dividend funded by asset sales — a payout the company itself flags as at risk. Balance-sheet leverage rose to ~3.0x reported net-debt/EBITDA (~2.3–2.5x on run-rate), but the rating is investment-grade (BBB/Baa2) and tangible book is positive (~$10.34/share). On valuation, IP trades at ~8.7x the cut FY26 EBITDA guide — a deserved discount to PKG (~11.6x), roughly in line with its true economic twin Smurfit WestRock (~6.9x). The market is capitalizing the guide and almost nothing beyond it.

This is a turnaround/value name: mediocre business quality, a poor capital-allocation history, a real but unproven operating inflection, a constructive cycle, and genuine insider conviction at a washed-out price. The body that follows takes no position; it lays out why each of those statements is true and what would falsify them.

2. Business Overview

International Paper manufactures containerboard — linerboard, corrugating medium, white-top liner, recycled grades and saturating kraft — and converts the majority of it into corrugated boxes and specialty packaging. The economic engine is a vertically integrated mill-to-box network: large, capital-intensive paper mills produce board that is shipped (within an economic freight radius of roughly 150–250 miles) to a dense network of corrugating/converting plants that turn board into customer-specific packaging. Integration captures the converting margin and stabilizes mill utilization; it is the core of whatever cost advantage the company possesses (see Competitive Position).

Post-transformation, IP reports two geography-based segments:

• Packaging Solutions North America (PS NA) — FY2025 sales $15,175M, segment operating profit $572M (down from $891M in FY2024, depressed by the $958M accelerated depreciation). ~11M tons of US containerboard capacity across ~15 mills and ~170 NA converting plants. End markets: e-commerce/parcel, processed and fresh food, protein, beverages, agriculture, durables and general industrial. No single customer exceeds 10% of revenue.
• Packaging Solutions EMEA (PS EMEA) — FY2025 sales $8,451M, an operating loss of −$236M (versus +$60M in FY2024). This is the legacy DS Smith business plus IP’s pre-existing European operations — corrugated packaging weighted toward UK/Continental fast-moving consumer goods and e-commerce, a more recycled-fiber-intensive, customer-design-led model than NA.

The Global Cellulose Fibers (fluff/market/specialty pulp) business — historically IP’s second segment, serving diapers, tissue and absorbent-hygiene converters — was sold to American Industrial Partners for $1.5B and closed in late January 2026; it is reported as discontinued operations and completes IP’s pivot to pure-play packaging. The 2021 Sylvamo spin had already exited printing papers.

Revenue character: this is repeat-but-cyclical staple-consumable demand. Corrugated boxes are destroyed in use and re-ordered, so volume is recurring in a practical sense — but it is not contractual, it correlates tightly with industrial production and non-durable consumption, and price resets with the published benchmark. There are no subscriptions, no installed-base lock-in, and near-zero switching costs at the box level. The customer base is broad and diversified, which removes concentration risk but also denies the company any captive pricing power.

Verdict: A straightforward, understandable, vertically integrated commodity-packaging business with diversified end markets and recurring (but cyclical and uncontracted) demand. The model is sound and easy to analyze; the question is not what IP does but how well the economics work — addressed in the Financial Quality and Capital Allocation sections.

3. Industry Dynamics

Structure — a consolidated oligopoly on a commodity product. North American containerboard is one of the more consolidated heavy-industrials: the top three producers control roughly two-thirds of >40M tons/year of US capacity — IP #1 (~29% share), Smurfit WestRock #2 (~20%), and Packaging Corp ~16% (PKG having added Greif’s containerboard assets for ~$1.8B at ~8.5x, late 2025). Barriers to entry are high in the capital sense — a greenfield mill runs $1B+ and multiple years — and the sustainability narrative (renewable, recyclable fiber displacing plastics) is a genuine secular tailwind for fiber-based packaging.

But the structure’s attractiveness is bounded by the product. Linerboard and medium are fungible commodities; pricing is administered off a published benchmark (Fastmarkets/RISI Pulp & Paper Week), so individual producers are price-takers to an industry-level number, not price-makers to their own customers. Demand is GDP- and IP-correlated and cyclical: box shipments were soft across 2023–2025 and 2026 industry demand has been guided to roughly flat versus a ~1.5% long-run trend. The result, historically, is that returns mean-revert to ~WACC for the average player — exactly IP’s experience.

The capital cycle — constructive but inherently self-undermining. 2025–2026 is a textbook favorable phase. North American producers removed an estimated ~3.9M tons (~10%) of capacity in 2025 — described by management as “unprecedented” — including IP’s own closures of Riceboro and Savannah (~1M tons). Tightened supply against flat demand allowed two 2026 price increases totaling ~$100/ton (IP +$70 March 1, Smurfit WestRock +$50 June 1) to stick. This is the supply-side discipline that supply-cycle analysis prizes — high prices following capacity withdrawal. The catch is structural: the same majors that just closed capacity are simultaneously converting and adding board (IP’s $250M Riverdale, Alabama conversion comes online Q3-2026; others are expanding), which seeds the next oversupply. The bullish phase plants the next glut; the open question is whether post-merger consolidation (Smurfit Kappa+WestRock in 2024; IP+DS Smith; PKG+Greif) has permanently improved pricing discipline or merely created larger participants who will, as ever, fill their mills.

Input costs and regulation. Key variable costs are recovered fiber (OCC), virgin wood, energy and freight — OCC and diesel were ~$200M headwinds called out in early-2026 guidance. Regulation is a modest net positive (anti-plastics, recyclability mandates) offset by mill emissions/permitting costs. EMEA carries incremental regulatory and energy-cost intensity versus NA.

Verdict: structurally FAIR — neither a good nor a bad industry. High barriers, consolidation and a sustainability tailwind are real positives; commodity economics, GDP-bound cyclical demand and a self-correcting capital cycle cap them. The industry rewards the low-cost, disciplined operator (PKG) with above-WACC returns and punishes the average one with sub-WACC returns through the cycle. IP has been the latter. An attractive industry position is available here; an attractive industry in the abstract is not.

4. Competitive Position

The moat, named precisely: a real but narrow cost advantage plus economies-of-scale-with-local-density — and nothing else. IP’s defensible economics, to the extent they exist, come from (a) a fleet of large, generally low-cost integrated mills with fiber access, and (b) the mill-to-converting integration and freight-radius density that let it serve box customers reliably within an economic transport distance and capture the converting margin. That is a genuine barrier — a new entrant cannot cheaply replicate a national mill-and-plant network — but it is the commodity kind of moat: it protects a cost position, not a price.

What IP conspicuously lacks: switching costs (a box buyer can re-source on price and reliability), network effects, brand/captivity (no end consumer asks for an IP box), and any proprietary technology. There is no mechanism by which IP can charge more than the benchmark for an equivalent box. The moat, such as it is, defends margin via cost and reliability, not via pricing power.

The decisive test is the financial outcome, and it fails the strong-form version. A moat that mattered would show up as durably superior returns on capital. IP’s do not: consolidated ROIC has been chronically sub-WACC (peak ~6.8% in 2021; ~4.4% in 2023; management’s own 2022–2024 long-term incentive plan scored three-year adjusted ROIC at 7.55% versus a 9.0% target). The most damning comparison is intra-industry:

PKG earns roughly nine points more EBITDA margin on the same end-product. That gap is not industry luck — it is operating discipline, a tighter ~95%-integrated NA-only network, and decades of refusing value-dilutive scale. It is the clearest possible evidence that in containerboard, scale is not the moat; cost discipline and integration are. IP is the biggest and has earned the least.

Does DS Smith + 80/20 change the moat type? No — it changes the operator. The DS Smith acquisition added geographic scale but not a new moat mechanism; EMEA corrugated is, if anything, a slightly better business (more design-led, customer-collaborative) but IP’s EMEA segment is currently running an operating loss. The 80/20 program (see Growth) is an attempt to close the operating gap to PKG within the existing cost/density moat — better SKU/customer mix, reliability, commercial discipline — not to build captivity or pricing power. The tell that scale-for-its-own-sake failed is IP’s own decision to split the company back apart twelve months after assembling it.

Verdict: COMMODITY PRICE-TAKER with a real-but-narrow cost/local-scale moat that has historically protected the best operator (PKG) and only partially the biggest (IP). Durable advantage in the weak-form sense (the network is hard to replicate); no durable advantage in the sense that matters — the ability to earn above the cost of capital through a cycle. Until ROIC clears WACC on a sustained basis, the honest characterization is “crowded commodity market in which IP has been a structurally sub-par returner.”

5. Financial Quality

Quality of earnings — FY2025 is a kitchen-sink trough; read it through the adjustments. Reported FY2025 was a GAAP net loss of −$3,516M (−$6.95/share). The bridge:

• Loss from continuing operations: −$2,838M (−$5.61).
• Add back net special items of ~+$3,237M (pre-tax), dominated by a $2,467M Packaging-EMEA goodwill impairment (a DS Smith over-payment admission), plus ~$500M+ of severance/restructuring and acquisition/integration costs; less ~$487M of tax on those items.
• Result: an adjusted operating loss of roughly −$100M (−$0.20) for the year.
• The −$678M discontinued-operations loss (GCF) bridges continuing-ops to the −$3,516M total.

A critical, easily-missed nuance: the $958M of accelerated depreciation on the Red River/Savannah/Riceboro mill closures is not in special items — it sits inside segment/continuing-ops profit and is left in the adjusted figure. So IP’s −$0.20 adjusted loss is conservatively stated on that item; normalizing the non-cash $958M back puts underlying operating earnings clearly positive. The adjusted numbers are therefore fair-to-conservative on cash flow, but flattering in one respect — they exclude the $2.5B EMEA write-off, which is a real economic signal about the price paid for DS Smith, not noise.

Run-rate. Management’s FY2026 framework began at adjusted EBITDA from continuing operations of $3.5–3.7B and was cut to $3.2–3.5B after Q1-2026. FY2026 D&A is guided to ~$2.0B (versus $2.88B in FY2025), confirming the $958M does not recur. FY2025 reported EBITDA of $2,872M is the trough; the clean forward run-rate is the ~$3.2–3.5B guide.

Income statement and margins. Revenue was $23,634M in FY2025 (+$7.8B year-on-year, essentially all DS Smith; organic NA roughly flat). Segment margins tell the quality story: PS NA produced a $572M operating profit (3.8% margin, depressed by accelerated depreciation), while PS EMEA lost −$236M — and within that, the acquired DS Smith pieces lost money in 2025 (NA-DS Smith −$346M; EMEA-DS Smith −$321M per segment disclosure). This is an acquisition that, in its first full year, reduced consolidated profitability. Gross margin held ~28–30% across the cycle, but the path from gross to operating margin — where IP’s cost structure and capital intensity bite — is where it lags PKG by ~9 points.

Cash flow — the soft underbelly. Free cash flow by year ($M): 2021 +1,481 / 2022 +1,243 / 2023 +692 / 2024 +757 / 2025 −159. Capex ran $1,857M = 7.9% of sales in FY2025 — this is a capital-hungry business, structurally. The negative FY2025 FCF means the $977M dividend was funded from asset sales, not operations. FY2026 FCF is guided to just $300–500M against a ~$1B dividend — a ~200–330% payout that is plainly unsustainable from internal cash. The 10-K explicitly pre-flags the risk (“no assurance dividends will continue at historical levels … reduction or suspension”). A dividend reset (likely at, or as part of, the spin) is a live event, not a tail risk.

Balance sheet — levered but investment-grade. Net debt rose to ~$8,686M (from $4,491M in FY2024) on the DS Smith assumption — ~3.0x reported net-debt/EBITDA, ~2.3–2.5x on run-rate EBITDA. Goodwill ($5.3B) and intangibles ($4.0B) from DS Smith are large, but tangible book is positive at ~$5.46B (~$10.34/share) — a meaningful correction to the assumption that a deal-heavy balance sheet must be tangibly insolvent; it is not. Liquidity is ample: ~$1.9B of revolver/securitization capacity, ~$1.1B cash, plus the ~$1.5B GCF proceeds (used substantially for debt paydown). Ratings: BBB stable (S&P) / Baa2 stable (Moody’s). Pension is manageable. There is no solvency question here; there is a capital-intensity-meets-dividend question.

Verdict: economics do NOT reliably improve with scale. Gross margins are stable and the balance sheet is sound, but operating economics are commodity-grade, ROIC is chronically sub-WACC, the business is capital-hungry (capex ~8% of sales), and free cash flow is thin-to-negative at the trough — with a dividend currently larger than the cash that supports it. The bull case requires the FY2026 run-rate and the 80/20 cost-out to convert into durable mid-teens-or-better EBITDA margins and positive covered FCF; the history says be skeptical until it shows.

6. Capital Allocation

This is the section where IP’s record is hardest to defend, and where the forward bet is most concentrated.

M&A — a twenty-year pattern that never compounded. IP has been a serial large-scale dealmaker and serial divestor: Temple-Inland (2012), the Weyerhaeuser pulp business (2016), the Russia/Ilim and Olmuksan exits, the Sylvamo printing-papers spin (2021), the GCF sale (Jan-2026), and now DS Smith (2025). Despite two decades of reshaping, consolidated ROIC never durably cleared the cost of capital. The pattern is buy-big, integrate, under-earn, then spin or divest — motion without compounding.

The DS Smith economics crystallize the problem. The deal was all-stock (0.1285 IP shares per DS Smith share; 178.1M new IP shares, ~45% dilution, DS Smith holders ending at ~34% of the combined company). Because IP’s stock rose between announcement and close, the realized consideration ballooned from the ~$5.8B headline to ~$9.9B at the 31-Jan-2025 close — roughly 12x pre-synergy EV/EBITDA, ~8x only if the full $514M synergy target lands. About $6.3B (~64%) was booked as goodwill and intangibles in a commodity business — of which $2,467M was impaired within the first year. Then, on 29-Jan-2026 — twelve months after closing — IP announced it will split into two public companies. Buying a “global champion” and announcing its break-up within a year, after a $2.5B write-down, is the single most expensive instance of the buy-big/under-earn/spin pattern. Round-trip transaction and separation costs (DS Smith deal costs of $237M+$86M, $37M advisory, plus coming dis-synergy and separation costs) are deadweight, and the ~45% dilution is largely permanent even after the spin.

Capital returns — strained. Buybacks are effectively dead (>$1.28B in 2022 → $65M in 2025) while 178M shares were issued for DS Smith — the opposite of per-share value creation. The dividend (~$1.84/share, ~$1B, ~4.7% yield) is uncovered by FCF and, on management’s own framing, near its EBITDA “breakeven” — explicitly at risk of a reset around the spin. Capital has flowed out via dividends funded by asset sales, not returned from surplus FCF.

Reinvestment — genuinely attractive at the project level, immaterial at the corporate level. The bolt-ons and conversions management is now doing — NORPAC (Longview, WA), the Riverdale conversion, the Dover converting facility — are pitched at high-teens ROIC and are credible. The problem is scale mismatch: a few hundred million of high-return projects cannot move the needle on a ~$9.9B deal and a multi-billion capex base, so the consolidated average stays sub-WACC while the marginal dollar earns well. This is the right kind of capital allocation, finally — but it is small relative to the M&A that defines the record.

Incentives — partial alignment, loose pay-for-performance. The short-term plan (STI) is Adjusted EBITDA 70% / Revenue 20% / Cash Conversion 10% — no ROIC. The long-term plan (LTIP) does gate 50% on three-year adjusted ROIC (the other 50% on relative TSR) — a real positive that few commodity peers match. But the mechanism proved loose: in 2022–2024 the ROIC half missed (7.55% vs 9.0%) while a strong relative-TSR result (86th percentile) bailed out an above-target payout. CEO Silvernail’s 2024 package totaled ~$20.7M (partial year): $1M base, ~150%-of-target AIP, $12.5M of annual PSUs, plus an $8.5M, 100%-performance inducement PSU with stock-price hurdles to lure him from IDEX. Governance is otherwise above-average — independent board, robust clawback, double-trigger change-in-control, no stock options — though it combines the Chair and CEO roles.

Insider alignment — the contrarian offset, and it is real. Insiders flipped decisively: 2024–2025 was sales only (officers at $46–58); then a cluster of open-market purchases in Jan–May 2026 at $31–40, led by CEO Silvernail buying 50,000 shares (~$2.0M) at ~$40 on 30-Jan-2026 — the day after the split announcement — plus Directors Connor (~$1.0M @ $40.37), Gustafsson (~$1.0M @ ~$38) and Tozier (10,000 @ $31.30). Buying high in the cycle and buying more at the trough is the cash-backed conviction signal that most cuts against the value-trap read.

Verdict: NEGATIVE-to-MIXED. Historically a value-destroyer whose deals never lifted returns above their cost; the DS Smith/split sequence is the costliest example. Against that sit three genuine positives — a 50%-ROIC LTIP gate, high-return bolt-on reinvestment, and a CEO-led insider-buying cluster at the lows. The entire bull case is a forward bet that a new operator breaks a twenty-year pattern. The record says demand proof; the insider buying says someone with better information is willing to bet on it.

7. Growth History and Forward Opportunities

History — low-quality, M&A-and-price-driven. Five-year revenue is a sawtooth dominated by deals and spins, not organic compounding: $17.6B (2020) → $19.4B (2021) → $21.2B (2022) → $16.0B (2023, post-Sylvamo) → $15.8B (2024) → $23.6B (2025, +49% almost entirely DS Smith). Strip the M&A and organic legacy volume has been flat-to-down, with cyclical swings driven by box demand and benchmark price. Near-term “growth” in 2026 is largely administered-price recovery (+$100/ton), not unit expansion. By any quality lens this is low-quality growth: acquired, cyclical, price-led, capital-intensive.

The 80/20 program — the heart of the forward case. Silvernail is importing the IDEX/Danaher 80/20 operating system: identify the ~20% of SKUs/customers that drive ~80% of profit, prune the unprofitable tail, redeploy resources to the core, and win share through reliability and commercial excellence (the “GlobalConnect” commercial transformation, disciplined pricing). The early evidence is real:

• Volume share gains: NA box volume +2.5% versus an industry −0.3% — a third consecutive quarter of ~3-point outperformance, consistent with reliability-driven wins.
• Cost-out: $710M of run-rate savings achieved, with a stated target >$1B.
• Pricing leadership: IP led the +$70/ton March-2026 increase and held EMEA pricing.
• High-return reinvestment: NORPAC, Riverdale and Dover bolt-ons at high-teens ROIC.

The counter-evidence is equally real. Management has missed guidance three straight quarters, cutting FY2026 adjusted EBITDA to $3.2–3.5B one quarter after setting it at $3.5–3.7B, with the CEO publicly conceding “we have missed numbers … we haven’t given ourselves enough breathing room.” EMEA is currently an operating loss. The 2027 $5B EBITDA target (NA ~$3.5B + EMEA ~$1.5B) requires a heavily back-loaded ~75% H1→H2 NA ramp and unbudgeted price upside ($90M per $10/ton) — i.e., it leans on the cycle as much as on self-help. And the announced split re-introduces dis-synergies 12–24 months after the merger that created the synergies.

Forward opportunities, ranked by credibility: (1) cost-out completion toward >$1B (high credibility, partly banked); (2) NA share gains from reliability (medium-high, demonstrated); (3) high-return bolt-ons/conversions (medium, small); (4) EMEA turnaround to profit (medium, unproven); (5) the full $5B target and SOTP re-rate from the split (lower, cycle- and execution-dependent).

Verdict: LOW-QUALITY growth historically, with a genuine but unproven self-help inflection. The 80/20 traction is not imaginary — the volume outperformance and cost-out are measurable — but it is cyclically entangled (favorable 2026 price cycle, ~10% industry capacity removed) and has not yet survived a clean four-quarter test. The growth that would re-rate the stock is margin/ROIC normalization, not revenue, and that remains to be demonstrated.

8. Changes and Headwinds — Last Two Years

The past 24 months have been the most eventful in IP’s modern history:

• Leadership reset (May-2024): Andrew Silvernail (ex-IDEX CEO) succeeds Mark Sutton as CEO; Lance Loeffler (ex-Halliburton CFO) joins. A deliberate import of the Danaher/IDEX operating culture into a commodity paper company.
• Takeover interest and the DS Smith deal (2024): Brazil’s Suzano reportedly approached IP; IP instead agreed to acquire DS Smith. The all-stock deal closed 31-Jan-2025, creating the world’s largest containerboard/corrugated player and a secondary LSE listing.
• Footprint surgery (2025): Permanent closures of Riceboro and Savannah containerboard mills and the Savannah packaging facility (~1M tons out by Sep-2025); exit from molded fiber; the $250M Riverdale (AL) conversion to containerboard (online Q3-2026). ~$958M of accelerated depreciation booked.
• GCF divestiture (closed late-Jan-2026): Global Cellulose Fibers sold to American Industrial Partners for $1.5B; IP becomes pure-play packaging; ~$660M of proceeds to debt paydown.
• The $2.5B EMEA impairment (FY2025): A $2,467M goodwill write-down against the just-acquired DS Smith EMEA business — a first-year admission of over-payment.
• The split announcement (29-Jan-2026): IP will separate into a North American “International Paper” and an EMEA (legacy DS Smith) spin-off, dual-listed, targeting completion in ~12–15 months, with IP retaining ~20% of EMEA for 12–18 months. Both entities targeted investment-grade.
• Three guidance misses and a cut (2025→Q1-2026): Culminating in the FY2026 EBITDA cut to $3.2–3.5B and a candid CEO mea culpa.
• Insider buying cluster (Jan–May 2026): CEO and multiple directors buy in the open market at $31–40 after years of selling.

Headwinds into 2026: soft-to-flat box demand; OCC and diesel cost inflation (~$200M); reliability/specialty and weather disruptions (~$125M of the Q1 cut); the back-loaded 2027 ramp; dividend-cut overhang; and separation/dis-synergy costs from the coming split. Tailwinds: ~$100/ton of sticking price increases; ~10% of NA capacity removed; cost-out momentum; share gains; and a deleveraging path aided by GCF proceeds.

Verdict: net thesis-NEUTRAL, high-variance. The changes are simultaneously the source of the bull case (new operator, 80/20, constructive cycle, value-surfacing split) and the bear case (a costly merger being unwound, repeated misses, an uncovered dividend). They have raised the dispersion of outcomes rather than clearly strengthening or weakening the thesis — which is precisely why the stock trades on the operator’s credibility rather than on the assets.

9. Risk Analysis (Risk Matrix)

Net read: the high-likelihood risks (dividend cut, cyclicality) are medium-impact and substantially priced in; the high-impact risks (execution miss, capital-cycle reversal, 80/20 proving cyclical) are medium-likelihood and are the real swing factors. A catastrophic loss is unlikely given the investment-grade balance sheet and positive tangible book. This is a dispersion/execution risk profile, not a solvency one.

10. Valuation Discussion (Embedded Expectations)

No price target; no recommendation. The figures below are embedded-expectations and scenario analysis.

Where the multiple sits. At $38.76 (EV ~$29.1B), IP trades at ~10.1x FY2025 trough EBITDA ($2,872M, depressed by $958M of non-recurring accelerated depreciation), ~8.7x the FY2026 guide midpoint (~$3.35B), and ~5.8x the 2027 $5B target. EV/sales is ~1.2x; P/B is 1.39x; FCF yield is thin (negative in FY2025). The ~4.7% dividend reads as a cut flag, not a value signal, given the payout math. Against its own multi-year history the stock is middling — neither washed-out nor rich versus its own range, with the price/book multiple (the cleanest read, below its own median) the most reliable tell.

Peer comparison. PKG trades at ~11.6x EV/EBITDA (21%+ margins, FCF-generative, ~12% ROIC, NA pure-play) — the quality premium. Smurfit WestRock, IP’s true economic twin (mid-teens margins, ~45% debt/EV, goodwill-heavy, mid-integration), trades at ~6.9x. IP at ~8.7x sits between the two, at a justified ~3-turn discount to PKG (warranted by the ~9-point margin gap, sub-WACC ROIC, higher leverage, integration/separation overhang, and a value-destructive M&A history) and at a modest premium to SW. IP is not obviously mispriced on a screen — any upside is earnings normalization, not the closing of a peer-multiple gap.

Embedded expectations. At EV ~$29.1B and ~8.5–9x, the market is capitalizing ~$3.2–3.4B of EBITDA — precisely the FY2026 guide, and essentially nothing more. It is paying ~zero for the $1.6B bridge to the 2027 $5B target, for the SOTP re-rate from the split, or for 2026 price hikes beyond what’s guided. The implied ROIC is ~8% — roughly the cost of capital, and consistent with management’s own 7.55% three-year LTIP result. After three straight misses, that skepticism is rational, not punitive: the market is underwriting the company IP currently is, and giving away the option on the company it says it will become.

Sum-of-the-parts (the announced two-way split).

• On near-run-rate EBITDA: NA ~$2.4B × ~9.0x + EMEA ~$0.95B × ~6.0x − $8.7B net debt ≈ ~$35/share — slightly below spot.
• On the 2027 target: NA $3.5B × ~9.5x + EMEA $1.5B × ~6.5x − $8.7B ≈ ~$65/share (+68%).
• The SOTP is not a hidden-asset unlock; its upside is the same earnings-recovery-plus-re-rate bet as the embedded-expectations gap, eroded by separation costs and the permanent ~45% dilution. The split is a focus/governance improvement, not a value teleport.

Scenario analysis (equity value = EBITDA × EV/EBITDA − $8.7B net debt ÷ ~528M shares):

Verdict. IP is fairly priced for what it is delivering and cheap only if you underwrite the turnaround. A defensible fair-value zone is ~$38–48; the reason to own at the trough is option value on the back-loaded 2027 ramp and the insider-signaled inflection, not a cheap multiple today. The asymmetry improves materially below book (~$27.85) and toward the May washout (~$29), where the downside is increasingly the dividend cut (medium impact, largely known) and the upside is the full re-rate.

11. Variant Perception

Consensus. The Street treats IP as a “show-me” turnaround in a commodity industry: respect for Silvernail and the 80/20 thesis, tempered by three guidance misses, an uncovered dividend, and a value-destructive deal history. Roughly fair-valued; ratings cluster Hold-to-modest-Buy. The factor tape agrees — DividendYield +0.61, Value +0.16, Quality +0.17, with no Momentum or Growth loading — this is an abandoned dividend/value cyclical-materials name, beta ~0.89, elevated idiosyncratic vol ~33% (event-driven), a poor long-run compounder (lifetime return ~2.7%/yr, max drawdown −90%) currently stabilizing off a washout (latest-quarter move ~+11% raw; price back above its 21/50/200-day moving averages). Factor-similar names confirm the neighborhood (SW, Smurfit Kappa, PKG, Sylvamo). Characterization: falling-knife-stabilizing / early-base-building, contrarian-value with event-driven optionality — not momentum, and not a value-trap to short.

Strongest bull case. A proven Danaher-school operator is applying a repeatable system to a structurally under-managed asset base, into a favorable capital cycle, at a washed-out price, with the CEO and board buying the lows. 80/20 is already producing measurable share gains and $700M+ of cost-out; the split sharpens focus and surfaces value; the constructive +$100/ton cycle plus EMEA recovery carry EBITDA toward $4–5B, where the stock is worth $55–65. You are paying ~zero for that bridge.

Strongest bear case. This is a commodity business that has earned below its cost of capital for a decade, and no operating system repeals commodity economics. The $9.9B DS Smith deal destroyed value (already $2.5B impaired) and is being unwound at further cost; the dividend will be cut; the 80/20 “gains” are a favorable price cycle in disguise; the 2027 target is back-loaded fiction; and when the cycle rolls over, IP re-rates to ~$23 with a sub-WACC ROIC intact. The insider buying is a single operator’s optimism against twenty years of corporate evidence.

The 3–5 assumptions that matter most:

1. Is 80/20 structural or cyclical? (Does the margin gap to PKG durably close, or revert when price softens?)
2. Where is the capital cycle going? (Does post-merger discipline hold, or does new conversion capacity glut the market?)
3. Does the dividend cut clear the overhang or signal distress? (Reset-from-strategy vs. cut-from-weakness.)
4. Does EMEA turn, and is the split accretive net of dis-synergies and costs?
5. Will management resist the next big, dilutive deal and let ROIC compound?

Falsification. Bull falsified by a fourth guidance miss, EMEA staying loss-making, or the margin gap to PKG failing to narrow as price softens. Bear falsified by two-to-three clean quarters holding the FY26 guide with EMEA turning profit-positive and ROIC visibly tracking toward the 9% LTIP gate.

Where consensus may be offsides: the tape’s complete absence of momentum/growth loading and the elevated event-vol suggest the market is positioned for “value trap” and has written off the optionality entirely. If the operator delivers even modestly — not the full $5B, just a clean year that holds the guide — the re-rate from “priced for the guide” to “priced for the trajectory” is the asymmetric move. The variant perception is not that IP is a great business (it isn’t); it is that a fair price + real insider conviction + free optionality on a credible operator is a better risk/reward than the value-trap consensus implies.

12. Fact vs. Interpretation

13. Open Questions

1. Dividend decision and timing — will the cut come at the spin, before, or be avoided via asset sales? Magnitude? This is the single largest near-term known unknown.
2. EMEA economics ex-purchase-accounting — what is the true underlying margin of the DS Smith EMEA business once integration/step-up noise clears, and is the operating loss structural or transitional?
3. Split mechanics and balance-sheet allocation — how is net debt apportioned between NA IP and the EMEA spin, and what are the standalone leverage/credit profiles?
4. 80/20 durability — does the NA volume/share outperformance hold when benchmark price softens, and does the margin gap to PKG measurably narrow?
5. Capacity discipline — net of the closures and the Riverdale/other conversions, is the industry adding or removing effective capacity into 2027?
6. Synergy realization vs. dis-synergy — how much of the $514M DS Smith synergy is banked before the split re-fragments the company, and what are the recurring dis-synergies of separation?
7. Capital-allocation guardrails post-spin — will the two boards commit to ROIC-first reinvestment and resist the next transformative deal?
8. Normalized through-cycle ROIC — can the combined self-help + cycle actually push sustained ROIC above the ~9% gate, the only test that matters for the moat verdict?

14. What Must Be True

Bull case — what must be true:

• 80/20 is structural: NA EBITDA margins move durably toward the mid-to-high teens and the gap to PKG narrows even as benchmark price normalizes.
• The capital cycle stays disciplined: post-merger consolidation holds pricing, and net new capacity does not glut the market through 2027.
• EMEA turns profit-positive and the split is net accretive after dis-synergies and separation costs.
• Cost-out reaches >$1B and converts into covered free cash flow, so a reset dividend is sustainable from operations.
• Management resists the next big deal and lets ROIC compound toward/above the 9% gate.
• Falsification test: a fourth consecutive guidance miss, EMEA remaining loss-making through 2026, or the IP–PKG margin gap failing to narrow as price softens — any one breaks the structural-inflection thesis and returns IP to value-trap status.

Bear case — what must be true:

• Commodity economics dominate: 80/20 gains prove cyclical, and ROIC reverts to sub-WACC when price rolls over.
• The dividend is cut from weakness, signaling distress rather than strategy.
• The DS Smith unwind continues to bleed value (further impairment/dis-synergy), and the split fails to re-rate.
• The next cyclical downturn drops EBITDA toward $3.0B and the stock toward ~$23.
• Falsification test: two-to-three clean quarters that hold the FY2026 guide, EMEA turning profit-positive, ROIC visibly tracking toward the 9% LTIP gate, and a dividend reset executed from strategy (with covered FCF afterward) — that combination falsifies the value-trap call and validates the operator-led re-rate.

15. Source Appendix

Full citations with URLs, dates and filing sections are listed below. Primary sources prioritized over secondary.

Key primary sources: International Paper FY2025 Form 10-K (ip-20251231, filed 27-Feb-2026) and prior 10-Ks (2021–2024); FY2025/Q1-2026 earnings releases and 8-Ks (DS Smith close 31-Jan-2025; GCF sale 23-Jan-2026; split announcement 29-Jan-2026); Q4-2025 (29-Jan-2026) and Q1-2026 (30-Apr-2026) earnings-call transcripts; DEF 14A proxy statements (LTIP/STI metrics, executive compensation); Form 4 insider filings via SEC EDGAR (Jan–May 2026 purchase cluster); Fastmarkets/RISI containerboard pricing and AF&PA box-shipment industry data; peer filings (PKG, Smurfit WestRock).

APPENDIX A — Standard Diligence Questionnaire — International Paper Company (NYSE: IP)

Supplemental to the research memo. Grounded in the FY2025 10-K (ip-20251231, filed 2026-02-27), the Q4-2025 (2026-01-29) and Q1-2026 (2026-04-30) earnings-call transcripts, the 2025 DEF 14A, and EDGAR Form 4 filings. Fact / Interpretation / Assumption labels applied where the distinction matters. As-of price $38.76 (2026-06-26).

General

What thoughtful questions have other investors asked about this company? The genuinely contested questions cluster around six issues. (1) Is the 80/20 operating system structural or cyclical lipstick? — i.e., can CEO Silvernail’s IDEX/Danaher playbook durably close the ~9-point EBITDA-margin gap to Packaging Corp (PKG ~21% vs IP ~12%), or do the gains evaporate when the benchmark price softens? (2) Will the dividend be cut? — FY26 FCF guided to $300–500M against a ~$1B payout (~200–330% of FCF). (3) Was DS Smith a value-destroying mistake? — buying a “global champion” for ~$9.9B (realized) in Jan-2025, impairing $2.5B of its goodwill in FY25, then announcing a break-up twelve months later. (4) Is the 2027 $5B EBITDA target credible after three straight 2025 guidance misses and a heavily back-loaded H2-26/2027 ramp? (5) How is the split structured — debt apportionment, tax-free status to US holders, which entity keeps the IG rating? (6) Why is the CEO buying his own stock at the lows (~$2M at $40, Jan-2026) if the assets are as troubled as the write-offs suggest? (Interpretation: synthesized from the Q1-2026 analyst Q&A — Staphos/BofA, Weintraub/Seaport, Ng/Jefferies pressed repeatedly on credibility and the back-loaded ramp.)

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Low — a kitchen-sink trough. (Fact) FY25 GAAP net loss −$3,516M (−$6.95); adjusted operating loss ~−$0.20; reported EBITDA $2,872M is depressed by a one-time $958M of accelerated depreciation on mill closures that does not recur. (Interpretation) The clean forward run-rate is the FY26 adjusted-EBITDA guide of $3.2–3.5B (cont. ops, post-GCF), and management’s stated 2027 target is $5B — so FY25 sits well below mid-cycle. The trough is partly cyclical (soft box demand 2023–25) and partly self-inflicted (integration drag, EMEA loss, footprint surgery).

Driven by the external environment or internal actions? Both, roughly equally. (Fact, per Q1-26 NA bridge) External: macro/inflation (diesel, OCC, chemicals, soft demand) ~−$200M; weather (Jan-26 ice storm) ~−$50M. Internal: 80/20 cost-out (+$510M NA run-rate in 2025), footprint optimization, the deliberate “box go-to-market” pruning of unprofitable tail volume, and DS Smith integration cost. The recovery lever is internal (cost-out, share gains, mix) layered on an external price cycle (+$100/ton 2026 administered price increases).

How stable are revenues? Repeat but cyclical — not contractual. (Fact, 10-K MD&A) Demand correlates with non-durable industrial production, e-commerce, processed foods, protein and agriculture. Boxes are consumable (destroyed in use, re-ordered), giving a soft GDP-bound volume floor, but price resets off the Fastmarkets/RISI benchmark and swings with the cycle; there are no subscriptions, no backlog of note, near-zero switching costs. (Interpretation) “Repeat-but-cyclical staple demand,” not high-quality recurring revenue. No single customer >10% of sales — low concentration risk, but also no captivity.

Outlook for products/services? Stable-to-modestly-growing in volume; price-led near term. (Fact) 2026 NA industry demand guided to ~flat (revised down from flat-to-+1%); IP expects to outgrow the industry by ~2% (third straight quarter of ~3-point NA share outperformance, +2.5% vs −0.3% industry in Q1-26). EMEA softer (~0.5–1% growth, IP modestly underperforming as it “held pricing”). Secular tailwind: fiber displacing plastic (recyclability/sustainability).

How big will this market be — growing, shrinking, domestic or international? A large, mature, low-single-digit-growth, GDP-correlated market. (Fact) US containerboard capacity >40M tons/yr; long-run box-shipment trend ~+1.5%/yr (recently flat). Both domestic (NA, ~$15B IP sales) and international (EMEA, ~$8.5B); post-split the two pieces become regional pure-plays. (Interpretation) Not a growth market — the investment case is margin/return normalization within a flat-volume industry, not market expansion.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Less, on consolidation — but the product stays a commodity. (Fact) Top 3 NA producers ≈ two-thirds of capacity after a wave of consolidation: Smurfit Kappa+WestRock (2024), IP+DS Smith (2025), PKG+Greif containerboard (~$1.8B, late 2025). 2025 saw ~10% (~3.9M tons) of NA capacity removed (“unprecedented”), enabling ~$100/ton of 2026 price increases to stick. (Interpretation — supply-cycle lens) This is a constructive but self-undermining capital cycle: the same majors that closed capacity are converting/adding board (IP’s Riverdale comes online Q3-2026), which seeds the next glut. Consolidation has improved discipline; whether permanently is the open question.

How profitable is the business (ROIC, ROE)? Poor — chronically below cost of capital. (Fact) Consolidated ROIC: ~2.8% (FY20), ~6.8% (FY21 peak), ~4.4% (FY23), n/m/negative (FY24–25). Management’s own 2022–2024 LTIP scored 3-yr adjusted ROIC at 7.55% vs a 9.0% target — i.e., its own scorecard confirms sub-WACC returns. ROE: 5.5%/19.5%/2.8%/5.6%/−45.8% (FY20–25; FY25 distorted by impairment). (Interpretation) Even at the cyclical peak IP barely earned its ~8–9% WACC. This is the central bear fact.

How profitable is the industry — how many competitors, what barriers to entry? (Fact) Barriers are high in the capital sense (a greenfield mill is $1B+/multi-year), which is why the industry is a consolidated oligopoly. But profitability varies enormously by operator on identical economics: PKG earns ~21% EBITDA margin / ~12% ROIC; IP ~12% margin / sub-WACC ROIC; Smurfit WestRock ~15% / mid-single ROIC. (Interpretation) The barrier protects a cost position, not a price. There is no demand captivity (boxes are multi-sourced spec’d commodities), no network effects, no brand. The industry rewards the low-cost disciplined operator and punishes the average one through the cycle.

Can the business be easily understood? Yes — unusually clean. A vertically integrated mill-to-box commodity-packaging company; two geography segments (NA, EMEA); pure-play after the Sylvamo (2021) and GCF (Jan-2026) exits. The economics are transparent: tons × benchmark price, minus fiber/energy/freight, times an integration rate. No opaque financial-engineering or hard-to-model end markets.

Can it be undermined by foreign low-cost labor? Largely no — geography is the moat input. (Fact/Interpretation) Corrugated is bulky and low-value-density; boxes serve a ~150–250-mile freight radius, so production is inherently local/regional and not import-exposed at the box level. Containerboard (the board itself) trades internationally to a degree (export markets are a leading price indicator), but the converting business is geographically protected. EMEA energy costs (not labor) are the bigger competitiveness risk there.

Do brands matter? No. No end consumer asks for an “IP box.” Brand/captivity is irrelevant; the relevant differentiators are cost, reliability, service, and design capability (more so in EMEA’s design-led model). (Interpretation) The absence of brand is one reason there is no pricing power above the benchmark.

What is the nature of competition? Administered-price oligopoly on a commodity. (Fact) Pricing is set off the published Fastmarkets/RISI benchmark via list-price moves led by the majors (IP led the +$70/ton March-2026 increase, “second out”). Competition is on cost, reliability, integration, and selectively on price/service for accounts. It is not auction-based and not a price war in normal conditions, but individual producers are price-takers to the industry number.

Customers’ switching costs? Near zero. (Fact) Boxes are spec’d commodities, routinely multi-sourced and re-bid on price and reliability. (Interpretation) This is the decisive moat weakness — IP’s only defense is to be the low-cost, most-reliable supplier within the freight radius (the 80/20 “win on reliability” thesis), not to lock customers in.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? Modest. (Interpretation) The owned timberland/fiber-supply relationships and the integrated mill-network density carry economic value beyond carrying cost, but IP is not a large timberland holder post-spins. The high-teens-ROIC reinvestment projects (NORPAC, Riverdale conversion) create value not yet visible in reported returns. No large hidden-asset story — this is not a “trapped assets in a conglomerate” SOTP; the split unlocks focus, not hidden value (see Valuation).

Off-balance-sheet liabilities? (Fact) Standard items: operating leases, a $449M net underfunded pension portion (manageable), ~$5.9B of non-current other liabilities including environmental remediation and postretirement obligations, and a $500M receivables-securitization facility (used). UK stamp tax (~$50M) on the DS Smith deal. (Interpretation) Nothing alarming or hidden; the environmental/postretirement tail is normal for a heavy-industrial paper company. Separation/dis-synergy costs of the coming split are an unquantified future obligation.

How conservative is the accounting? Fair-to-conservative on the key item. (Fact/Interpretation) The single most important QoE nuance: the $958M of accelerated depreciation on mill closures is left inside continuing-ops/adjusted earnings (not added back to “adjusted operating earnings”), so IP’s −$0.20 adjusted loss is conservatively stated — normalizing the non-cash $958M puts underlying operating earnings clearly positive. The offset: adjusted figures exclude the $2.5B EMEA goodwill impairment, which is a real economic signal (over-payment), not noise. The accelerated-dep / purchase-accounting D&A step-up are the main distortions; both are transparent in the 10-K. (Assumption: no aggressive revenue or capitalization games detected in the filings reviewed.)

How CapEx-hungry is the business? Very — structurally. (Fact) FY25 capex $1,857M = 7.9% of sales (FY24 5.8%, FY23 7.1%); normalized run-rate ~$1.5–1.9B/yr. Against ~$1.7–1.9B OCF, this leaves thin FCF. (Interpretation) Capital intensity is a core reason FCF is fragile and the dividend is uncovered at the trough; it is intrinsic to mill-based heavy industry, not a one-off.

Capital Allocation & Management

How much FCF does the business generate, how does management use it, what is the philosophy? (Fact) FCF (OCF − capex, $M): FY21 +1,481 / FY22 +1,243 / FY23 +692 / FY24 +757 / FY25 −159. FY26 guided $300–500M. Uses: dividends (~$977M FY25, funded by asset sales not FCF that year), capex, and debt paydown (~$660M of GCF proceeds). Buybacks effectively dead since 2022 ($1,284M 2022 → $65M 2025). (Interpretation) The historical philosophy has been “buy-big / integrate / under-earn / spin,” funding a large dividend through the cycle. The new regime (Silvernail) emphasizes 80/20 cost discipline, high-return bolt-ons, footprint pruning, deleveraging, and the value-surfacing split — a credible forward pivot against a poor historical record.

Significant acquisitions recently? (Fact) DS Smith plc — all-stock, closed 31-Jan-2025; 178.1M new IP shares (~45% dilution), DS Smith holders ~34% of IP; realized consideration ~$9.9B (ballooned from the ~$5.8B headline because IP’s stock rose between announce and close), ~$6.3B (~64%) booked as goodwill+intangibles, ~12x pre-synergy EV/EBITDA (~8x on full $514M synergy). Bolt-ons: NORPAC (Longview WA paper mill, “high-teens ROIC”), Dover converting facility. (Interpretation) DS Smith is the costliest instance of the buy-big pattern — impaired $2.5B within a year and slated for break-up twelve months after closing.

Buying back shares? No. (Fact) Repurchases collapsed to $65M (2025) from $1,284M (2022); meanwhile 178M shares were issued for DS Smith. Share count rose ~363M (2022) → ~528M (2025), ~+45%, while the stock fell — the opposite of per-share value creation.

Issuing large amounts of new shares to insiders? No abnormal insider issuance. (Fact) The large issuance was the DS Smith deal currency, not insider grants. Executive equity is PSU-based; routine RSU vesting (code A) and tax-withholding (code F) churn at annual vesting dates. All directors+officers as a group held <0.4% of shares (2025 proxy) — low absolute insider ownership, typical of a large-cap with a new CEO.

Compensation policy of directors/management? (Fact) STI = Adjusted EBITDA 70% / Revenue 20% / Cash Conversion 10% — no ROIC; 2024 STI paid 179.1% of target. LTIP = 3-yr Adjusted ROIC 50% / Relative TSR 50% — ROIC is gated (a positive), but in 2022–24 the ROIC half missed (7.55% vs 9.0%, 63.75% earned) while an 86th-percentile TSR (200% earned) bailed out an above-target blended payout. CEO Silvernail 2024 total ~$20.7M (partial year): $1M base, 150%-target AIP, $12.5M annual PSUs, plus an $8.5M 100%-performance inducement PSU with stock-price hurdles vesting May-2027. Ownership guideline 6x salary; combined Chair+CEO offset by an independent lead director; say-on-pay ~96%. (Interpretation) Governance structure is above-average; pay-for-performance is only partially tight (TSR can rescue a ROIC miss). The inducement being 100%-performance, not a guaranteed make-whole, is a positive.

Motivations of management? (Interpretation, evidenced) The strongest signal is cash-backed: insiders flipped from net sellers (2024–25 at $46–58) to a net-buying cluster (Jan–May 2026 at $31–40) — CEO Silvernail bought 50,000 sh (~$2.0M) at ~$40 on 30-Jan-2026 (the day after announcing the split), joined by Directors Connor (~$1.0M @ $40.37), Gustafsson (~$1.0M @ ~$38) and Tozier (10,000 @ $31.30). (Assumption) These code-P buys appear discretionary (opportunistic, immediately post-announcement) rather than 10b5-1-planned. Combined with the stock-price-hurdle inducement PSU, management is aligned to a re-rate and is betting real capital on the turnaround.

Valuation & Market Data

Is the stock an ADR, MLP, or K-1 issuer? No to all. IP is the US parent C-corp (NYSE primary; secondary LSE listing post-DS Smith). Ordinary 1099 dividend reporting; no K-1, no UBTI, no ADR. The EMEA spin will be dual-listed (LSE + NYSE) and is targeted investment-grade.

Dividend policy? (Fact) ~$1.84/share, ~$1B aggregate, ~4.7% yield; 43-year increase streak (per company). Uncovered: FY25 FCF −$159M and FY26 guided $300–500M vs a ~$1B payout (~200–330% of FCF). Management framed dividend “breakeven” at ~$3.6–3.7B EBITDA — i.e., at/below the FY26 guide. The 10-K explicitly pre-flags “reduction or suspension” risk; management is “maintaining the policy through 2026” but reviewing it for the post-spin structure. (Interpretation) A reset/rebase (likely at or around the spin) is a live event, not a tail risk — the yield is a cut flag, not a value signal.

How profitable is the business? Covered above — sub-WACC ROIC, ~12% trough EBITDA margin vs PKG ~21%, ~0% operating margin in the FY25 loss year. Mediocre.

Is net income diverging from cash from operations? Yes, sharply — for explainable reasons. (Fact) FY25 GAAP NI −$3,516M vs OCF +$1,698M — a ~$5.2B divergence driven by the non-cash $2,467M EMEA goodwill impairment, the $958M accelerated depreciation, purchase-accounting D&A, and the discontinued-ops (GCF) loss. (Interpretation) The divergence is the right direction (cash >> GAAP loss), confirming the GAAP loss is dominated by non-cash one-timers — but OCF minus capex still left FCF negative, so it is not a “cash machine masked by accounting.” Read cash through FCF, not OCF: FCF was genuinely −$159M.

Risks & Downside

What factors would cause the stock to decline? (Fact/Interpretation, per the risk matrix) (1) a fourth straight guidance miss (medium likelihood / high impact); (2) a dividend cut from weakness rather than strategy (high / medium); (3) box-volume + benchmark-price cyclicality rolling over (high / high); (4) the capital cycle reversing as converted/new capacity gluts the market (medium / high); (5) 80/20 proving cyclical not structural (medium / high); (6) input-cost inflation (OCC/energy/diesel); (7) EMEA turnaround failing / value-destructive split; (8) capital-allocation recidivism (another big dilutive deal); (9) key-person (Silvernail) departure. Bear scenario: EBITDA ~$3.0B × ~7x → ~$23/share (−41%).

Risk of a catastrophic loss? Low. (Fact) Investment-grade (BBB/Baa2 stable), positive tangible book (~$10.34/share), ~$1.9B liquidity plus ~$1.1B cash and ~$1.5B GCF proceeds, ~3.0x reported leverage (~2.3–2.5x on run-rate). (Interpretation) No solvency path absent a severe, prolonged shock. This is a dispersion/execution-risk profile, not a balance-sheet-risk one.

Chance of a total loss? Very low. Tangible book is positive, the balance sheet is IG, and even the bear case is a de-rate to ~$23 (a dividend cut + cyclical trough), not impairment of the equity. A total loss would require a multi-year demand collapse plus failed deleveraging — not supported by current evidence.

Recent News & Events

Has the business environment changed recently? Yes — both the company and the cycle. (Fact) Company: pure-play pivot (GCF sold Jan-2026), the announced two-way split (29-Jan-2026), three guidance misses culminating in the FY26 EBITDA cut to $3.2–3.5B (Apr-2026), and the insider-buying cluster. Cycle: ~10% of NA capacity removed in 2025 and ~$100/ton of 2026 price increases sticking against flat demand — a constructive (if fragile) supply-side turn. The event timeline is built from 8-Ks, press releases and transcripts.

Significant acquisitions? DS Smith (closed Jan-2025, ~$9.9B); GCF divested ($1.5B to American Industrial Partners, closed Jan-2026); bolt-ons NORPAC (Longview WA) and Dover converting. The defining recent corporate action is the announced break-up into NA IP + EMEA spin.

Change in accounting policies? (Fact) No fundamental accounting-policy change; the major presentation changes are reclassification of GCF to discontinued operations (restating enterprise figures) and DS Smith purchase accounting (goodwill $5.3B, intangibles $4.0B, D&A step-up). The $2.5B EMEA goodwill impairment and $958M accelerated depreciation are events within existing policy, not policy changes.

Recent changes — new markets, facilities, management? (Fact) Management: CEO Silvernail (ex-IDEX, May-2024), CFO Loeffler (ex-Halliburton, Apr-2025), board refresh (Beggs, Robbie, Tozier). Facilities: closed Red River/Savannah/Riceboro mills (~1M tons out, 2025); Riverdale (AL) conversion to lightweight containerboard (~$250M, online Q3-2026); NORPAC and Dover added; ~31 EMEA site closures/in-process. Markets: secondary LSE listing (from DS Smith); EMEA spin to be dual-listed. (Interpretation) The most eventful 24 months in IP’s modern history — raising outcome dispersion rather than clearly strengthening or weakening the thesis.

APPENDIX B — Source Appendix — International Paper Company (NYSE: IP)

Primary sources prioritized over secondary. “Primary” = SEC/regulatory filings, company press releases, earnings-call transcripts, and audited financials. “Secondary” = trade press, industry data and peer filings. SEC CIK 0000051434. As-of date 2026-06-27; price $38.76 (2026-06-26).

1. SEC / Primary Filings (PRIMARY)

Filing history available via SEC EDGAR full-text search (https://efts.sec.gov/LATEST/search-index?q=…) and the company filing index (https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000051434). Form-type counts over the trailing five years: Form 4 ×208, 8-K ×76, SC/SCHEDULE ×36, DEFA14A ×18, Form 3 ×15, 10-Q ×15, 144 ×14, 11-K ×10, 10-K ×5, DEF 14A ×5, plus merger proxies.

2. Earnings-Call Transcripts (PRIMARY)

Speakers: CEO Andrew Silvernail; CFO Lance Loeffler; President PS EMEA Tim Nichols (Q4-25). Treated as management hypothesis, validated against the 10-K/press-release bridges. Transcripts available via company IR and public transcript providers (e.g., Seeking Alpha / Motley Fool) for the same call dates.

3. Quantitative Data (SECONDARY — third-party aggregated; reconciled to filings)

Third-party aggregated/statistical data are not primary; for this US filer EDGAR and the 10-K/10-Q/press releases are authoritative, and every material number reconciles to a filing.

4. Industry / Peer Data (SECONDARY — trade press, industry data, peer filings)

Reconciliation discipline: third-party aggregated/statistical data are not primary; for this US filer EDGAR and the 10-K/10-Q/press releases are authoritative, and every material number in the memo reconciles to a filing. Management transcript commentary is treated as hypothesis and validated against the financial statements. No third-party analyst target was adopted as a price target.