Sociedad Química y Minera de Chile SQM

At a glance

One-line thesis. SQM is the Chilean brine giant behind ~50% lithium / ~23% iodine / ~22% specialty-fertilizer revenue, and the lithium cycle has clearly turned — Q1 2026 revenue was $1.76B, +70% YoY, gross margin inflected to 44% (from 29% in FY25), and consensus sees 2026 revenue nearly doubling to $8.1B — but the company missed the Street's EPS number in each of the last five quarters, a new $2.36B minority interest now carves ~10% off quarterly earnings, Chile's fiscal grip already flipped 2024 to a −$404M loss, and after a −23% drawdown the price action is broken — so this is a Watch: cheap at ~11× forward if the recovery holds, but you are underwriting a commodity price, not a compounder, and the entry is not yet earned.

1. What it is

Sociedad Química y Minera de Chile (NYSE: SQM, ADR) is one of the world's largest producers of lithium and derivatives (carbonate and hydroxide for battery cathodes), iodine and derivatives (X-ray contrast media, polarizing films, biocides, pharma synthesis), specialty plant nutrition (potassium nitrate and premium blends), plus potassium and industrial chemicals (including solar salts for thermal storage). Its assets center on the caliche and brine deposits of northern Chile. Incorporated 1968, headquartered in Santiago; CEO Ricardo Ramos Rodríguez; ~8,344 employees; NYSE-listed since 1993.

Revenue mix (FY2025 segment data, from filings):

• By product: Lithium & derivatives $2.29B (50%) · Iodine & derivatives $1.04B (23%) · Specialty plant nutrition $0.98B (21%) · Potassium $156M (3%) · Industrial chemicals $75M (2%) · Others $32M. Iodine is the quiet stabilizer — it grew through the lithium bust ($968M FY24 → $1.04B FY25) while lithium collapsed from its $8.15B FY2022 peak to $2.24B (FY24). That 2022 number is the whole cautionary tale in one line.
• By geography (FY2025): Asia & others $2.88B (63%) · North America $689M (15%) · Europe $655M (14%) · Latin America & Caribbean $188M (4%) · Chile $162M (4%). Demand is global (battery supply chains run through Asia), but production and the fiscal regime are ~100% Chile — the concentration that matters is jurisdictional, not commercial.
• Ownership structure changed in FY2025: minority interest on the balance sheet jumped from $37M to $2.36B, and intangible assets from $35M to $2.56B. This is consistent with the SQM–Codelco lithium partnership taking effect (the Chilean state miner taking a stake in the Atacama lithium operation — background context; the JV itself is public record, the balance-sheet figures are from the data file). The practical effect is visible in Q1 2026: $38.2M of the quarter's $402.9M continuing-ops income (≈9.5%) went to minority holders.

2. The expert thesis — what the panel actually says (traceable)

No expert-panel coverage — this note is fundamentals-driven. The Synthos KB contains zero claims on SQM. (A literal grep for "SQM" returns two Andreas Steno claims about Chinese real-estate vacancy measured in square meters — false positives, disclosed here so the audit trail is honest.)

That means: no bullish expert underwrites this name, and no bearish expert warns on it. Everything in this note — the bull case, the bear case, and the verdict — is built from the FMP data file (financials, segments, analyst estimates, technicals). Per house standard for screen-surfaced tickers, conviction is recorded as None and the note carries no borrowed authority.

What stands in for a thesis network: the Street's own distribution — 5 Buy / 8 Hold / 3 Sell (consensus Hold), targets from $53 to $106 with a $91 consensus — an unusually wide, genuinely divided book that mirrors the commodity-price dependence at the heart of the story.

3. Synthos scores & the Bull / Base / Bear cases

The one-glance judgment — three scores, 0–10, each anchored to real metrics (not probabilities we can't honestly calibrate):

The three cases (our own scenario model — assumptions shown; each target is a ~12–18-month fair value). We deliberately do not attach probabilities: the base case is the expected path, so a weighted blend would just restate it with false precision. The cases bound the range; the scores summarize them.

Synthos fair value = the base case, ~$90 (+22%), anchored deliberately on the Street's $91 consensus — with zero expert coverage and a commodity-price-driven model, our scenario math earns no premium over the 16-analyst book, and pretending otherwise would be false precision. The honest signal is the asymmetry: +50% bull vs −25% bear looks favorable, but the bear case is the recent lived history (2023–24), not a tail scenario. This is a tracked call — the Forecaster Scorecard grades it once it matures.

4. Exponential Potential

Synthos separates compounders (durable high returns on capital) from exponentials (accelerating, multi-baggers-from-here). SQM is neither — it is a cyclical, and the estimates say so plainly:

• Forward revenue: $4.57B (FY25) → $8.08B (2026E, +77%) → $7.90B (2027E, −2%) → $8.24B (2028E, +4%) → $9.33B (2029E) → $7.88B (2030E). One violent up-leg, then a plateau that never re-approaches the $10.7B 2022 peak.
• Forward EPS: $2.06 (FY25) → $6.78 (2026E) → $6.74 → $7.30 → $4.91 → $6.41. Flat-to-choppy for four years after the snap-back — consensus itself models a 2029 down-cycle inside the forecast window.
• The near-term second derivative is real: quarterly revenue $1.04B → $1.17B → $1.34B → $1.76B (Q1 2026), four straight sequential increases with YoY growth accelerating from +0.6% (Q2 2025) to +69.8% (Q1 2026). This is what the momentum screen caught.
• Honest caps: (1) estimate breadth thins from 11 EPS analysts (2026–27E) to 4 by 2029–30E, and the out-year ranges are enormous (2030E revenue $6.0–10.3B); (2) the 2026 consensus itself demands a steep H2 — Q1 actual ($1.76B) plus the Q2 estimate ($2.17B) is $3.93B, so H2 must deliver ~$4.15B; (3) capex runs heavy (16.5% of TTM revenue; $877M in FY25) — brine expansion is bought with capital, not code.

Exponential Potential: Moderate-Low (4/10). Own it (if at all) as a cyclical recovery trade with a defined exit, not as a hold-forever exponential.

5. Financials (real numbers — FMP annual/quarterly)

• Revenue: FY2025 $4.57B, +0.9% (FY24 $4.53B) — the trough flattening out. The full cycle: 2020 $1.82B → 2021 $2.86B → 2022 $10.71B (lithium supercycle peak) → 2023 $7.47B → 2024 $4.53B. TTM revenue is now $5.31B as the recovery quarters roll in.
• Quarterly trajectory (the turn): Q1 2025 $1.04B → Q2 $1.04B → Q3 $1.17B → Q4 $1.34B → Q1 2026 $1.76B (+69.8% YoY). Gross margin over the same span: 29.4% → 24.3% → 29.5% → 34.1% → 44.2%. This is a textbook commodity-price inflection hitting the P&L.
• Earnings: FY2025 net income $587M / EPS $2.06, versus FY2024's −$404M / −$1.42 — a loss created almost entirely by a $1.37B income-tax expense against $974M of pretax income (the Q1 2024 quarter alone booked $1.17B of tax, consistent with the resolution of SQM's long-running Chilean tax dispute — background context, figures from the file). FY2022 peak EPS: $13.68. Q1 2026: net income $364.7M / EPS $1.28, after $38.2M carved out to the new minority interest.
• The miss streak (from the earnings calendar): last five prints all missed Street EPS — $0.48 vs $0.63 · $0.31 vs $0.52 · $0.625 vs $0.69 · $0.64 vs $0.75 · $1.28 vs $1.78 — even as the last two quarters beat on revenue. Costs, tax rate (Q1 2026 effective rate ~41%) and the minority carve are eating the top-line upside; the 2026E consensus of $6.78 has to be read against this pattern.
• Cash flow: FY2025 operating CF $1.31B, capex −$877M, FCF $438M. TTM FCF yield 5.1%. FY2023 shows the cycle's violence: operating CF was negative ($-178M) on working-capital unwind and FCF was −$1.28B. Dividends: $2.24B paid in 2022, $1.47B in 2023, $67M in 2024, $4.3M in FY2025 — the payout follows the cycle down (TTM ratios still show a $0.99/sh dividend, ~1.3% yield; the FY25 cash-flow line and the TTM figure don't reconcile cleanly — a data-timing artifact we flag rather than smooth).
• Balance sheet: cash + short-term investments $2.73B vs total debt $4.82B → net debt $3.07B, 1.22× EBITDA. Current ratio 3.3× at FY-end (2.76× TTM), interest coverage 8.6×. Total assets $14.5B; parent equity $5.70B plus the new $2.36B minority interest. Tangible book value ~$19.74/sh.
• Data-quality caveats (honest): the FY2025 cash-flow statement lumps everything into a single "other non-cash items" line (net income shows as 0), which also breaks the TTM income-quality ratio (0.008 — meaningless as computed). R&D shows as $0 across all years (SQM expenses it within other lines). Neither changes the picture; both are flagged.

6. Valuation — priced in or room?

On trailing numbers SQM is unremarkable: 25.6× TTM EPS, 12.5× EV/EBITDA, 4.4× EV/sales, 3.6× book — a middling multiple on trough-to-recovering earnings. The case lives entirely in the forward numbers: ~10.9× 2026E EPS ($6.78), ~10.9× 2027E ($6.74), ~10.1× 2028E ($7.30). If consensus is right, you are paying single-digit-teens multiples for a recovered earner with a 5.1% FCF yield — genuinely cheap. The two honest asterisks: (1) those estimates come from a company that has missed EPS five quarters running, and (2) consensus itself pencils EPS back down to $4.91 in 2029 — the "E" in this P/E is a cycle, not a run-rate. FMP's letter rating is B (overall 3/5; DCF score 4/5 and ROE 5/5 vs P/E 1/5 and debt/equity 1/5 — the machine sees the same split we do). Street targets (context): consensus $91, median $102.50, high $106, low $53 — a 2× high-to-low spread with a Hold consensus (5 Buy · 8 Hold · 3 Sell). Cheap-if-right, fully-priced-if-the-cycle-rolls: exactly what val_desc says.

7. Technicals (from the tech block)

• Trend: damaged mid-term, intact long-term. $73.52 is 11% below the 50-DMA ($82.61) but still above the 200-DMA ($70.09) — the uptrend's last line of defense sits ~5% below the price.
• Location: −22.9% off the 52-week high ($95.31); max drawdown from peak −35.2% at the worst; still +105% above the 52-week low ($35.80).
• Momentum: RSI(14) 24.7 — deeply oversold (below 30); MACD −2.58 (negative and falling). The sellers are in control near-term, though this is statistically stretched territory where bounces start.
• Relative strength: −11.6% over 3 months vs SPY +14.6% / QQQ +23.6% — a ~26-point underperformance gap; +6.9% over 6 months vs SPY +10.2%; but +96.4% over 12 months vs SPY +21.1% — the screen's momentum is a year old, and the recent tape has broken it.
• Read: this is a falling-knife setup inside a longer uptrend — the worst combination to chase and a reasonable one to stalk. The 200-DMA (~$70) is the natural line: a hold-and-turn there with RSI recovering would repair the setup; a decisive break of it, and the next reference is far lower. Consistent with Watch, trigger ~$65.

8. Moat & competitive position

SQM's advantage is resource quality, not technology: the Atacama brine is among the lowest-cost lithium sources on earth, and its caliche ores underpin a leading global iodine position (a $1.04B segment that grew straight through the lithium bust — the closest thing to a moat-proof in the file). But three structural qualifiers keep this from being a fortress: (1) price-taking — SQM sets neither lithium nor iodine prices, and FY22→FY24 showed a −58% revenue round-trip with no defense; (2) the state is now inside the tent — the fiscal regime already produced the 2024 tax loss, and the new minority interest (consistent with the Codelco partnership) permanently redirects ~10% of earnings; (3) concession, not ownership — the core asset is a Chilean concession whose terms are politically renegotiable, as the last three years demonstrated.

Peer set (FMP-supplied, market cap): Alamos Gold $13.3B, Albemarle $15.8B, CF Industries $17.4B, Cemex $17.9B, DuPont $19.0B, IFF $21.1B, LyondellBasell $17.1B, Pan American Silver $19.0B, RPM $14.1B, Reliance Steel $19.3B. Only Albemarle is a true comp (the other Atacama lithium major); the rest are generic materials names of similar size — judge SQM against the lithium cohort (ALB, and non-listed-here peers), not this list. Against ALB specifically, SQM's iodine/SPN legs are a genuine diversification edge.

9. Management, capital allocation & guidance

• Capital allocation: counter-cyclically sane, cyclically painful. Capex held at $877M–$1.1B/yr through the bust (expansion continued into the trough); the dividend went from $2.24B (2022) to effectively zero (FY25: $4.3M paid) — following SQM's earnings-linked payout policy rather than a fixed commitment. FY25 also saw modest deleveraging (net debt issuance −$118M). No buybacks in any year on file — this is a dividend-when-earned, not a repurchase, story.
• Insider activity: the file contains no meaningful insider data — the single record is a 2016 institutional Form 3. As a foreign private issuer/ADR, SQM's insiders don't file US Form 4s; we flag the blind spot rather than pretend it's clean.
• Guidance: no management guidance is ingested in the KB for SQM (zero claims of any kind). The de facto bar is the Street's Q2 2026 estimate — EPS $1.89 on ~$2.17B revenue (2026-08-18) — a steep sequential step-up that management must validate for the 2026 consensus ($6.78 EPS) to survive contact with the miss streak.

10. Catalysts & what to watch

• Next earnings: 2026-08-18 (Q2 2026; Street EPS $1.89, revenue ~$2.17B). The two lines that matter: does revenue hold the ~$2B+ run-rate consensus H2 requires, and does the EPS-miss streak finally break (watch the effective tax rate and the minority-interest line — Q1's 41% tax rate and $38M carve are where the misses live).
• Lithium price prints: the single variable that moves everything — Q1's 44% gross margin is the realized-price signal to track quarter by quarter.
• Iodine segment: the stabilizer — FY25 $1.04B (+7.7%); continued strength cushions any lithium wobble.
• Chile fiscal/regulatory headlines: royalty terms, the operationalization of the state lithium partnership, and anything touching the Atacama concession — the 2024 tax year proved these are P&L events, not headlines.
• The 200-DMA (~$70): the technical line between "pullback in an uptrend" and "trend change."

Thesis tripwires (what would change the call): a sixth consecutive EPS miss with revenue also light; lithium-driven gross margin rolling back below ~35%; a break and hold below the 200-DMA; any new Chilean fiscal demand. On the upside: a clean beat-and-raise at $65-ish would flip this to a Buy — Tactical.

11. Key risks

• Commodity price (dominant): ~50% of revenue is lithium at market prices; the 2022→2024 round-trip (revenue −58%, EPS $13.68 → −$1.42) is the in-sample worst case, and consensus itself models a 2029 down-leg.
• Single-jurisdiction / fiscal: production, concessions and the tax regime are all Chile. 2024's $1.37B tax expense exceeding pretax income is the demonstrated, not hypothetical, version of this risk.
• Minority-interest drag: the new $2.36B minority stake took ~9.5% of Q1 2026 income; EPS models that ignore the carve will keep over-forecasting — plausibly part of the miss streak itself.
• Estimate risk: five straight EPS misses; 2026 consensus requires a steep H2 ramp; out-year estimate breadth thins to 4 analysts with huge ranges.
• Working-capital cyclicality: 218-day cash-conversion cycle and 196 days of inventory mean price swings hit cash flow with a lag and a whip (2023's negative operating CF is the exhibit).
• Demand-side: EV/battery demand growth decelerating (or chemistry shifts) would cap the lithium leg; iodine and SPN only partially offset.
• Data blind spots (honest): no insider visibility (ADR), no KB expert coverage, and an FY25 cash-flow statement reported in lump form — conviction is capped by what we can verify.

12. Verdict, position sizing & monitoring

Watch. The lithium recovery is real and visible in the file — Q1 2026 revenue +70%, gross margin 44%, four straight sequential revenue increases — and at ~10.9× 2026–27E EPS with a 5.1% FCF yield and a $90 base-case fair value (+22%), the stock is cheap if consensus holds. But everything that keeps this from being a Buy is also in the file: a five-quarter EPS-miss streak, a fiscal regime that already flipped a year to a loss, a new minority interest skimming ~10% of earnings, estimates that flatline after the snap-back, and a tape that is 11% below its 50-DMA with RSI at 24.7 while consensus sits at Hold. With zero expert coverage, no borrowed conviction can bridge that gap. Earn the entry.

• Sizing: 0% today. If the ~$65 trigger hits (<10× 2027E EPS, just under the 200-DMA) with lithium pricing intact — or a clean beat-and-raise breaks the miss streak — this becomes a Buy — Tactical at ≤1–2% in the commodity-cyclical sleeve, with an explicit exit on the §10 tripwires. This is a rent-the-cycle position, never a core anchor.
• Monitoring: re-underwrite at each print (next 2026-08-18); track gross margin, the tax/minority lines, and the 200-DMA. This verdict is logged as a tracked Synthos call as of 2026-07-06 at $73.52.
• Single biggest risk: the lithium price rolling over again — a price-taker in one country, with the state as a partner, and the last downcycle still visible in the rear-view mirror.

Provenance & disclosures

• Traceability: 0 KB claims on SQM — no expert-panel coverage; this note is fundamentals-driven per house standard for screen-surfaced names. A literal grep for "SQM" returns only square-meter (real-estate) false positives, disclosed in §2. kb_net_conviction is null because nothing exists to aggregate; fabricated conviction is structurally impossible (claim-ID reconciliation).
• Data as-of: fundamentals 2026-03-31 (Q1 2026, filed 2026-05-27) · estimates & prices 2026-07-06 (FMP data file) · KB claims: none. Forward figures are analyst consensus (FMP), labeled as estimates; out-year breadth thins to 4 analysts by 2029–30E.
• Valuation anchoring: base-case fair value (~$90) is deliberately anchored on the Street's $91 consensus — with no expert coverage, our scenario multiples (13× 2027E base, 15× 2028E bull, 11× trough-EPS bear) are labeled assumptions, not private information.
• Background-context labels: the Codelco lithium partnership and the Chilean tax-dispute resolution are public-record context used to explain balance-sheet and tax figures that are themselves from the data file (minority interest $37M→$2.36B; FY24 tax expense $1.37B); no figures were sourced outside the file.
• Data-quality caveats: FY2025 cash-flow statement is reported in lump form (breaks the TTM income-quality ratio); dividend TTM vs FY25 cash-paid figures don't reconcile cleanly (timing artifact); insider data is effectively absent (ADR); R&D reports as $0 by line-item convention.
• Not investment advice. Independent research, educational and informational only, never personalized. Hypothetical/forward figures are labeled; the only performance numbers Synthos will headline are the live, real-money Flagship's.
• Version: 2026-07-06. Prior versions available via the deep-dive version dropdown ("based on the info at the time").