Tronox Holdings

Phase A — Understand the business

Lens 1 · Company Overview

Tronox is the world's leading vertically-integrated manufacturer of titanium-dioxide (TiO2) pigment. The model is unusual in the sector: it both mines titanium-bearing mineral sands (in Australia and South Africa) and converts that feedstock into finished pigment at seven plants (United States, Australia, Brazil, UK, France, Saudi Arabia). The stated strategy is to be "as self-sufficient as possible" in feedstock so it can deliver "low cost, high-quality pigment" to ~1,200 customers in ~120 countries.

What it sells (FY2025 revenue $2,898M):

• TiO2 pigment — $2,298M (79%). A "quality-of-life" whitener/opacifier used in paint & coatings, plastics, paper. Management asserts there is no effective substitute. Sold under TIONA® and CristalActiv® brands. ~90% of capacity is chloride-process, ~10% sulfate.
• Zircon — $274M (9%). A mining co-product; ceramics glazes, refractories, foundry sand.
• Other — $326M (11%). High-purity pig iron (smelting co-product), monazite (the rare-earth feedstock), titanium tetrachloride (TiCl4), and other mining products.

Customers & contract structure: Highly fragmented demand — top-10 customers were 36% of FY2025 sales and NO single customer reached 10%. Contracts are short (≤1 year), recognized at point of control transfer; there is a "margin stabilization program" and a "long-term partnership strategy" intended to dampen price volatility, but pricing is fundamentally cyclical and spot-influenced. Top-10 customers have been supplied >10 years — relationships are sticky even if price is not.

Suppliers: Self-supplied on the hardest input (titanium feedstock). External inputs: chlorine, coke/carbon, process chemicals, energy, freight — bought on short/long-term contracts.

Competitors: Chemours, LB Group (Lomon Billions), Kronos Worldwide, INEOS, plus regional Eastern-European and Chinese producers.

The one-line model: A mine-to-pigment industrial. Earns a spread between feedstock cost (largely internalized) and pigment price (cyclical, currently depressed). ~5,700 employees across six continents.

Lens 2 · Supply Chain

Map the actual chain, upstream → downstream, with named nodes:

Upstream (in-house — the vertical-integration moat):

• Mines (titanium feedstock + co-products): KwaZulu-Natal Sands (Fairbreeze mine, 2 smelting furnaces) and Namakwa Sands (2 concentration plants, 2 furnaces) in South Africa; Northern Operations (Cooljarloo dredge + Chandala synthetic-rutile plant), Eastern Australia (Atlas mine + Broken Hill separation), and Perth Basin (Wonnerup) in Australia.
• Production capacity: ~832,000 MT/yr titanium feedstock (182kt rutile/leucoxene + 240kt synthetic rutile + 410kt titanium slag), plus ~297kt zircon and ~250kt pig iron.
• 2025 mine reinvestment: Fairbreeze extension commissioned; Namakwa East OFS under commissioning — both replace end-of-life mines and secure feedstock for "years to come."

Midstream (conversion — in-house): Seven pigment plants. Footprint is being cut in the downturn: Botlek (Netherlands) permanently closed (announced Mar 2025, ~240 jobs); Fuzhou (China) permanently closing (announced Jan 2026, ~550 jobs); one Namakwa smelter furnace idled. Going to five core pigment sites: US (Oklahoma/Hamilton), Australia, Brazil, UK (Stallingborough), France (Thann, ultrafine), Saudi Arabia (Yanbu).

Key external nodes:

• Chlorine, coke, process chemicals — multiple suppliers, contracted.
• ATTM (Advanced Metal Industries Cluster × Toho Titanium) — JV buyer of excess TiCl4 adjacent to the Yanbu plant; also the counterparty on the MGT loan/supply agreement.
• AMIC — JV partner on the Yanbu titanium-slag smelter; related to Cristal.

Downstream (the customer): ~1,200 pigment customers (paint, plastics, paper makers) across ~120 countries; market leaders in each end-use; zircon to ceramics/refractory/foundry buyers; pig iron to ductile-iron foundries (auto parts).

Chokepoints / single-source dependencies:

1. South Africa concentration risk — the single largest asset base ($1.01B PP&E ) sits in a jurisdiction the 10-K itself flags for "unpredictable regulatory, political and physical security environment" and grid reliability (Eskom). The Namakwa furnace idle is partly an energy-cost decision.
2. Self-supply cuts both ways — vertical integration is a cost moat in an up-cycle and an operating-leverage trap in a down-cycle: you can't flex your single biggest cost (your own mines) the way a merchant pigment buyer can. (Sourcing note: no supply-chain.md exists in the commercial layer for critical-materials — this lens is built from the 10-K + web.)

Lens 3 · Competitive Advantages (moats)

The moat is real but narrow, and it is a cost moat, not a pricing moat.

• Vertical integration (the core moat). Tronox is "unique in the degree to which we produce our own high-grade titanium feedstock". In a tight feedstock market this is a structural cost and security-of-supply advantage. It is the reason the company can credibly target "lowest-cost producer."
• Process IP / "hidden factory." 67 US patents + ~470 foreign; proprietary chloride chlorination/oxidation know-how; perpetual worldwide smelting license from Anglo American at Namakwa. But management concedes "much of the fundamental intellectual property associated with both chloride and sulfate pigment production is no longer subject to patent protection" — the durable edge is operational know-how, not enforceable IP.
• Scale + global footprint. Seven (→five) plants, ~1,200 customers, 120 countries — lets it serve multinational coatings customers globally and arbitrage regional pricing/logistics.
• Switching costs / stickiness. Top-10 customers supplied >10 years; margin-stabilization + long-term-partnership programs. Real but modest — pigment is a commoditized input and feedstocks are "highly substitutable."

Bargaining power: WEAK on both sides in this cycle. Demand is fragmented (no customer >10%) which is good, but the product is a commodity facing Chinese oversupply, so Tronox is a price-taker on pigment. It has more power over its own feedstock cost (it owns it) than over its selling price.

Verdict: A genuine low-cost-position moat that protects relative margin versus non-integrated rivals — but it does not protect absolute profitability when the entire industry price curve collapses. The moat keeps Tronox alive and middle-of-the-pack on margin (FY25 gross margin 12.5% vs Chemours ~17.5%, Kronos ~15.5% ) — it does not make the equity safe.

Lens 4 · Segments

Tronox reports one operating/reportable segment ("Tronox"), so disaggregation is by product and geography.

By product (net sales, $M):

Trend: decelerating across the board. Revenue peaked 2024 ($3,074M) and fell 6% in 2025. Zircon was hit hardest (-15% in 2025) on a 28% price drop partially offset by volume.

By customer geography (net sales, $M, 2025):

• EMEA $1,166 (40%) · Asia Pacific $790 (27%) · North America $749 (26%) · South & Central America $193 (7%).

By country of production (2025): Australia $696M, US $702M, Saudi Arabia $384M, UK $357M, South Africa $323M, other $436M — a truly global manufacturing base.

The number that matters — gross profit collapse:

Cause: TiO2 + zircon price deflation (Chinese oversupply) outrunning volume, compounded by idle-facility/lower-of-cost-or-NRV inventory charges as plants were closed. The segment is one business riding one commodity cycle — there is no diversifying second leg (yet — see rare earths, Lens 11/12).

---

Phase B — Measure performance

Lens 5 · Earnings Result (latest print — Q1 2026, reported 2026-05-07)

Vs consensus: Revenue $760M beat the ~$753M estimate, but adjusted EPS −$0.55 MISSED the −$0.46 consensus (~20% miss). A revenue beat with an earnings miss = volumes are holding, price/mix and cost are killing the P&L. That is the worst combination for a cyclical: you're shipping product at a loss to keep furnaces lit.

Line drivers: TiO2 revenue +5% (+$32M volume, −$22M price, +$22M FX from a stronger Euro); zircon +$20M (+57% volume, −28% price); other −$30M on lower pig-iron volume. The entire sector is buying volume with price.

Balance-sheet & cash flags (all red):

• Operating cash flow −$68M (vs −$32M Q1'25); capex $67M → quarterly FCF ≈ −$135M.
• Cash fell to $126M (from $199M at YE); funded the gap with +$85M net short-term debt (revolver draws).
• Inventories $1,577M — over 2x net sales-per-quarter; a working-capital anchor (a $67M Q1 inventory release was the only thing that softened the OCF burn).
• US holds only $7M of the $199M YE cash — cash is trapped at foreign subs and credit facilities restrict moving it to the US.

Guidance: No formal numeric guidance in the filing. Management's qualitative 2026 priority is "cash generation… improving pricing, efficient operations, reducing inventory," with capex guided down to ~$260M (from $341M FY25) and the cost program targeting $125–175M run-rate savings. CEO target: "resume debt paydown… long-term net leverage <3x." Note this is an aspiration, against a current 11.1x.

Market reaction: TROX has been in steady decline — ~$8.72 on the May 12 Goldman downgrade, ~$7.34 by mid-June 2026. The stock is a fraction of its ~$20+ 2021-2022 highs.

Unusual vs own history: TTM Adjusted EBITDA of $286M is the lowest in years, and the quarterly trajectory is still decaying: $93M (Q2'25) → $74M (Q3'25) → $57M (Q4'25) → $62M (Q1'26). A TTM net loss of $466M.

Lens 6 · Earnings Calls (sentiment trend)

Transcripts dir is empty; sentiment reconstructed from web coverage of the last several calls.

• Tone arc: From "constructive on a 2025 recovery" (early 2025) → a hard credibility break at Q2 2025 (the July 30 print that triggered the −38% one-day crash and the securities suit) → a defensive, self-help / liquidity-and-cost narrative through Q4'25 and Q1'26. Management stopped talking about near-term demand recovery and started talking about cost takeout, inventory reduction, plant closures, and "preserving optionality."
• Recurring phrases now: "cash generation," "sustainable cost improvement" ($90M realized, $125–175M target), "vertical integration," "rare earths optionality," "disciplined capital allocation," "balance-sheet resilience."
• Phrases they stopped saying: confident volume-and-margin guidance (precisely the statements the class action targets — "lofty margin projections… reliant on continually increasing sales volumes" ).
• Read: Management's messaging has shifted from offense to survival. The CEO's "<3x net leverage" line is a destination, not a forecast — credible only on a genuine pricing recovery they no longer promise to time.

Lens 7 · Comps

TiO2 is a Western oligopoly (Chemours / Tronox / Kronos control ~40% of global capacity ) versus a long tail of Chinese producers. Peer multiples below are ``, dated, and thin — the whole group is loss-making or trough-earning, so EV/EBITDA on depressed EBITDA is close to meaningless. Cited where found; otherwise n/a.

Equity market cap ($1.2B) vs net debt ($3.2B) → enterprise value ~$4.4B is ~94% debt-funded. This is a credit story wearing an equity ticker. On any normalized-EBITDA basis (say a mid-cycle $500–550M), EV/EBITDA is ~8–9x — not obviously cheap for a no-growth cyclical with a broken balance sheet; on trough EBITDA it's optically expensive. The real comp is not the P/E — it's where the bonds trade (see Lens 10: term loans at 18–23% discounts to par, unsecured 2029 notes at ~30% discount).

Lens 8 · Stock-Price Catalysts (what actually moves TROX)

The pattern over the cycle is unambiguous: TROX trades on TiO2 volume/price prints and on balance-sheet events, not on strategy.

• Jul 30–31, 2025: −38% in one day ($5.14 → $3.19) on the Q2 demand/volume miss — the single biggest move and the basis of the securities class action.
• Sep 2025: $400M 9.125% secured-notes raise — issued at par but at a punitive coupon; read by the market as a defensive liquidity move (the equity treated it as dilutive-to-credit-quality, not accretive).
• Q3 2025 dividend at $0.05 (the 60% cut took effect Q3'25) — confirmed the cash-preservation regime.
• May 12, 2026: Goldman Buy→Sell, −7% to $8.72; Truist Hold→Sell ($8 PT); Fermium Buy→Hold ($9).
• Dec 9, 2025: $600M EXIM + Export Finance Australia rare-earth letters — a positive optionality headline; modest, durable interest because it's non-binding/conditional.

Read-through: This stock reacts to (1) the direction of the TiO2 cycle (volume + price surprises), and (2) solvency signals (debt raises, dividend, leverage ratio, covenant headroom). It barely reacts to the long-dated rare-earth story. To own the equity is to take a dated view on the cycle turning before the balance sheet breaks.

---

Phase C — Judge people & books

Lens 9 · Management

• CEO: John D. Romano. Sole CEO since Apr 1, 2024; Co-CEO from Mar 2021; with Tronox/predecessors since 1988 (37 years) across sales, marketing, pigment ops, commercial strategy. Archetype: a lifelong commercial/operations insider — not a founder, not a financial engineer, not a hired-gun turnaround CEO. Deep industry knowledge; no demonstrated track record of navigating a balance-sheet restructuring.
• Track record: Ran the commercial side through the Cristal acquisition (2019) and the vertical-integration build-out. The cost program ($90M run-rate, "3x original target") is a genuine operational win and the clearest evidence of competent execution under pressure. The flip side: the team's demand-forecasting credibility is the subject of an active securities suit (Lens 10), and leverage tripled on this management's watch.
• Capital-allocation history — mixed, recently defensive: Built feedstock self-sufficiency (Fairbreeze, Namakwa OFS) — strategically sound, returns "above cost of capital" per management. But the company paid $48M of dividends in FY2025 while burning $281M of FCF and carrying 9x leverage — continuing to pay any dividend in that posture is a questionable use of scarce cash. A $300M buyback authorization sits unused (zero repurchased 2025) — correctly dormant. Net: reinvestment was rational; shareholder returns were continued slightly too long into the downturn.
• Skin in the game: Insider ownership data not in research layer (insider-transactions.csv absent). 2025 RSU grant-date fair value collapsed to $7.56 (from $16.69 in 2024) — comp is now deeply underwater, which at least aligns management with a recovery. n/a — insider %-ownership not sourced.
• Red flags: (1) the securities class action over demand/margin statements; (2) Cristal owns ~24% with two board seats and preemptive/registration rights — a concentrated holder whose interests "may not always coincide" with minority shareholders and which entrenches against takeover; (3) related-party MGT loan and AMIC/ATTM dealings with Cristal-affiliated entities (small, disclosed, but worth watching).

Lens 10 · Forensic Red Flags

Forensic-analyst lens. Every figure labeled.

Accounting-quality observations (filing-grounded):

1. Cash flow ≠ earnings, in the wrong direction. FY2025 net loss −$473M but operating cash flow +$60M — the gap is $635M of non-cash add-backs, $232M of which is restructuring (Botlek/Fuzhou). That restructuring is mostly real cash that will be paid (cash payments scheduled through 2026), so the "Adjusted EBITDA $336M" overstates sustainable cash generation. Adjusted EBITDA adds back $232M of restructuring; reported EBITDA was only $27M. Treat Adj EBITDA with suspicion — the adjustments are doing enormous work.
2. Receivables outrunning sales + off-balance-sheet financing. AR rose to $331M (Q1'26) even as the company runs a securitization facility (raised $230M→$255M→$275M across Mar–May 2026) that derecognizes sold receivables and books the proceeds as operating cash. The serial facility increases are a liquidity tell — they're factoring receivables harder to fund operations. Plus a $50M inventory-financing (repo) arrangement. Both flatter "operating cash flow" relative to the underlying.
3. Inventory risk. $1,577M of inventory with $139M of FY2025 lower-of-cost-or-NRV/idle charges already taken — in a falling-price environment, further write-downs are a live risk.
4. Deferred-tax assets $834M with full valuation allowances already on Australia/Brazil/Netherlands/UK losses (0% effective tax rate; no benefit recognized on losses). The DTAs are only worth anything if profitability returns — a "<recovery-contingent" asset.
5. Mark-to-market on the debt itself is the loudest red flag. The 10-K's own fair-value table shows the market pricing distress: 2024 Term Loan carried $741M / fair value $605M (−18%); 2024-B carried $893M / FV $684M (−23%); 4.625% Senior Notes due 2029 carried $1,075M / FV $754M (−30%). The credit market is pricing a real probability of restructuring on the unsecured notes.
6. Credit ratings deteriorated hard: Moody's Ba3 → B2 (negative); S&P B → CCC+ (negative) over 2025. CCC+ is substantively distressed.

Leverage / liquidity (the core forensic story):

• Total debt $3.3B; net debt $3.17B; net debt / TTM Adj EBITDA = 11.1x (was 9.0x at YE, 4.8x at YE'24). A clean reverse-deleveraging.
• Total liquidity $406M ($126M cash + $280M revolver). Thin for a business burning >$100M/qtr.
• Maturity wall: scheduled long-term-debt maturities are 2026 $39M, 2027 $37M, 2028 $31M, 2029 $1,833M, 2030 $415M, thereafter $859M. The $1.075B 4.625% unsecured notes + the $736M 2024 Term Loan both mature in 2029 — this is the bar the equity must clear.
• Covenants: No financial covenants on the term loans or bonds; ONE springing covenant on the Cash Flow Revolver (max first-lien net leverage 4.75x, tested only if revolver draw >35% of commitments). In compliance now. This covenant-lite structure is what keeps the company out of default today — and is exactly why bulls argue there's no near-term crisis. But covenant-lite delays a reckoning; it doesn't cancel the 2029 refi, which will reprice ~$1.8B of debt at distressed spreads if the cycle hasn't turned.

Regulatory findings (required sub-section):

• SEC Litigation Releases / AAERs: None. Verified via SEC EDGAR EFTS (LR + AAER) search 2021-06-20→2026-06-20.
• Securities class action (material): Keller v. Tronox Holdings plc, et al., No. 3:25-cv-01441, filed Sep 3, 2025, D. Connecticut. Alleges false/misleading statements Feb 12–Jul 30, 2025 that the company "was ill-equipped to forecast demand" and pushed "lofty margin projections… reliant on continually increasing sales volumes." Stock fell $5.14→$3.19 (−38%) on the Jul 31 corrective disclosure. Early stage; company "intends to defend vigorously"; no damages quantified. Lead-plaintiff deadline was Nov 3, 2025.
• Non-SEC enforcement (FTC/DOJ/FDA/etc.): Web search surfaced no material agency enforcement actions, consent decrees, or fines against Tronox. Ordinary-course environmental/legal matters disclosed (Hawkins Point Plant remediation, $41M provision — legacy Cristal site).
• Net: No SEC accounting-fraud findings; one active securities class action over forward-looking demand statements; large legacy environmental liabilities ($538M ARO + environmental, undiscounted) but well-disclosed and long-dated.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection

Built bottom-up from FY2025 actuals; every input labeled. Output ``. No forecast.ts logged (watchlist rule). TROX is loss-making at the net line, so the meaningful projection is Adjusted EBITDA → free cash flow → leverage trajectory, not EPS (which stays negative in base/bear).

FY2025 anchors: Revenue $2,898M · Adj EBITDA $336M · capex $341M (→$260M guided 2026) · interest ~$189M · TTM Adj EBITDA now $286M (Q1'26).

The only projection that matters: does the cycle turn before 2029? At ~$286M TTM EBITDA against ~$190M cash interest + ~$260M capex, the base case is roughly FCF-neutral-to-slightly-negative — survivable, but with no deleveraging. The equity is worth multiples of today only in the bull path; it's an impairment-to-zero candidate in the bear path. This is a binary, cycle-timing call, not a compounder.

Brier forecast (logged here for the record, not via forecast.ts): "TROX FY2026 Adjusted EBITDA < $400M" — p≈0.62, reflecting a still-decaying TTM trend ($286M) partially offset by the cost program; resolves 2026-12-31.

Lens 12 · Bull vs Bear

Bull case. Tronox is the world's #1 vertically-integrated TiO2 producer — a genuine, irreplaceable industrial asset — priced for distress at a ~$1.2B equity cap against $4.4B of replacement-cost mines and plants. The TiO2 cycle is at or near a trough; Western supply discipline (Botlek/Fuzhou/Chemours closures) plus anti-dumping duties (Brazil, Saudi, EU) on Chinese imports are tightening the market. The cost program delivers $125–175M of durable run-rate savings; capex steps down $80M in 2026; the balance sheet is covenant-lite with no maturities until 2029, buying years of runway. And there's a free call option: the rare-earth/monazite business, backed by $600M of conditional EXIM + Export Finance Australia financing under the US-Australia critical-minerals framework — a non-Chinese heavy-rare-earth supply chain the West desperately wants, attached to a TiO2 company for free. If the cycle turns, EBITDA doubles, leverage halves, and the equity is a multi-bagger off a low base.

Bear case (the 2–3 things that could permanently impair).

1. The 2029 maturity wall meets a balance sheet that can't deleverage. ~$1.8B comes due in 2029; the unsecured notes already trade ~30¢ below par. If TiO2 prices haven't durably recovered by ~2028, Tronox refinances ~$1.8B at distressed spreads (or restructures), and the equity absorbs the dilution/impairment. At 11.1x leverage with negative FCF, time is the enemy.
2. Vertical integration is an operating-leverage trap in a prolonged down-cycle. You can't flex your biggest cost (your own mines) the way a merchant pigment buyer can; furnaces want to run, so you ship volume at a 5.8% gross margin (Q1'26) and bleed cash to avoid idling assets.
3. Chinese structural oversupply doesn't go away. China added pigment capacity for a decade; even with duties, exports re-route. A "recovery" may be shallower and later than bulls model — the market re-rated TROX down precisely because management's demand forecasts proved wrong (the class action).

Pre-mortem (it's Dec 2027, the thesis broke — what happened?): TiO2 prices stayed flat-to-down through 2027 as Chinese supply re-routed around duties and global construction/coatings demand stayed soft. Adjusted EBITDA hovered ~$300M against ~$190M interest + $260M capex, so leverage never fell below ~8x. With the 2029 wall approaching, the 4.625% notes (already at ~70¢) priced an exchange; Tronox launched a distressed liability-management / equity raise that diluted holders 40–60%. The rare-earth project, still pre-revenue and capital-hungry, couldn't close non-recourse financing fast enough to matter. The equity halved again.

Are multiples too high? On trough EBITDA, EV/EBITDA looks expensive; on mid-cycle EBITDA (~$500M) it's ~8–9x — fair, not cheap, for a no-secular-growth cyclical carrying 11x leverage. The equity is not "cheap"; it's optionally cheap, conditional on the cycle.

Contrarian view (what the market may be missing, both ways): Bullish miss — the rare-earth optionality is being valued at ~zero; if the West's non-China heavy-REE push accelerates and the $600M facility converts to binding, Tronox holds a strategic asset that could be spun, JV'd, or financed off-balance-sheet, transforming the credit story independent of TiO2. Bearish miss — the consensus "no maturities until 2029 = no problem" framing understates how early credit markets force the issue; ~$1.8B of 2029 paper starts repricing the equity in 2027–2028, not 2029.

Lens 13 · Devil's Advocate (short-seller)

I am dismantling the bull case.

• What structurally breaks the model: A commodity producer with the cost structure of a miner (fixed, can't flex) selling into a market with the price behavior of a Chinese-oversupplied chemical. In a down-cycle that's maximum operating leverage in reverse — and Tronox is more exposed than non-integrated peers, not less. The "vertical-integration moat" is the bear thesis in disguise.
• Revenue concentration: Not customer concentration (healthy — top-10 only 36%). The concentration is single-commodity, single-cycle: 79% TiO2, one price curve, no second leg. There is nothing to offset a prolonged pigment trough.
• Why the moat is weaker than bulls think: Management admits the core process IP is largely off-patent; the edge is operational know-how, which narrows the cost gap to peers (12.5% gross margin actually trails Chemours and Kronos) but doesn't create pricing power. The moat keeps you alive at mid-pack margins — it does not protect the equity.
• Most dangerous competitor bulls underestimate: LB Group / Lomon Billions and the Chinese complex. Duties slow but don't stop re-routed exports; China's marginal cost sets the global floor, and that floor is below Tronox's all-in cost in this cycle.
• Worst capital-allocation / governance flags: Paying $48M dividends while burning $281M FCF at 9x leverage; Cristal's 24% block + 2 board seats entrenching against a takeover that might otherwise rescue minority holders; related-party MGT/AMIC dealings; an active securities suit alleging the team can't forecast its own demand.
• What must hold for today's ~$7 price: A TiO2 price recovery beginning within ~18 months, the cost program landing in full, and no further inventory write-downs and a 2029 refi at non-distressed spreads. Four things, all cycle-dependent.
• If growth disappoints 20–30%: At ~$220M EBITDA, leverage exceeds 13x, FCF is solidly negative, and the equity faces a dilutive raise or restructuring — a path to a fraction of today's price, plausibly toward zero in a 2029 restructuring.
• The single scenario that permanently impairs: TiO2 prices flat-to-down through 2028 → forced distressed treatment of the 2029 stack → equity dilution/wipe. Plausibility: moderate-to-meaningful (the bonds already imply ~30% impairment on the unsecured notes). This is a credit at the equity layer.

Lens 14 · Management Questions (ordered by information value)

1. At 11.1x net debt/EBITDA with $1.8B maturing in 2029, what is the specific plan and timeline to address the 2029 stack, and at what TiO2 price level does a refinancing become uneconomic / trigger a liability-management exercise?
2. What is your internal trough/recovery assumption for TiO2 and zircon prices through 2027 — and given the Q2'25 forecasting miss now in litigation, why should we trust this demand outlook?
3. At what leverage or liquidity threshold would you eliminate the remaining dividend entirely, and why is paying any dividend justified while burning FCF at >9x leverage?
4. The rare-earth $600M EXIM/EFA support is non-binding/conditional — what are the precise gating conditions, the expected timeline to a binding commitment, and the equity capital Tronox itself must contribute?
5. Walk us through the path to FCF-positive: at $260M capex and ~$190M interest, what Adjusted EBITDA level is breakeven, and what gets you there beyond cost cuts?
6. How much additional inventory write-down risk sits in the $1.58B balance if pigment prices fall another 5–10%?
7. You've raised the securitization facility three times in 2026 (to $275M) and run a $50M inventory repo — how much of reported "operating cash flow" depends on these, and what's the plan if they're pulled?
8. What is the all-in cash cost per tonne of your pigment versus the Chinese marginal producer, and at current prices, which of your five remaining plants are cash-negative?
9. South Africa is your largest asset base in a jurisdiction you flag for grid/political risk — what's the contingency if Eskom or regulatory disruption forces further furnace idling?
10. Cristal holds 24% with two board seats and preemptive rights — how do you ensure minority-shareholder interests are protected in any restructuring or strategic transaction?
11. The $300M buyback authorization is unused — confirm it stays dormant until leverage is below, say, 4x, and define the capital-allocation waterfall explicitly.
12. What proportion of the $125–175M cost-savings target is durable structural cost-out versus cyclical (volume/idle) that reverses when plants restart?
13. If the cycle has not turned by end-2027, what is the playbook — asset sales (which?), an equity raise, a rare-earth monetization, or a debt exchange?
14. How should we think about the ~$834M deferred-tax asset — under what profitability path does it become usable rather than a permanent valuation-allowance write-off?
15. What would have to be true for you to consider Tronox a takeover/merger candidate, and does the Cristal agreement practically foreclose that?

---