Pan American Silver Corp. (NASDAQ: PAAS) — A Gold Miner in Silver Clothing, Priced for Permanent Peak Metals

Independent equity research. Published 2026-06-27.

This article is independent fundamental research and general information, not investment advice. With the single, clearly-labeled exception of the author’s-view block immediately below, it contains no buy/sell recommendation and no price target; the analysis discusses valuation only as embedded expectations and scenarios.

⚡ Claude’s Take

This block is the author’s own subjective opinion. It is general information, not investment advice. Everything below it (the analytical body) is deliberately position-free and carries no price target.

Verdict: HOLD / AVOID adding here / accumulate only on a washout toward book. Not a short. Directional zone: I’d want to be buying a no-moat, peak-priced precious-metals price-taker in the high-$20s to mid-$30s (≈1.6–2.0x book of $17.43, the low end of its own decade-long P/B range), not at $45 (2.6x book, 11.8x peak-cycle EV/EBITDA — both near a ten-year ceiling). Fair-value zone on a metals-blended/normalized basis: ~$32–44. At $45.45 you are paying a peak multiple on a peak-metal-price earnings stream, with the gold/silver tape doing ~80% of the work and no quality or value floor to catch you if it reverts.

PAAS is the higher-quality, less-extreme cousin of Hecla (HL), which I covered today. Both are silver-branded miners re-rated to their richest-ever own-history multiples on a once-in-a-decade metals print. PAAS is the better business — bigger ($19B vs $9B), net-cash, lower-cost (FY25 silver AISC $13.88/oz with by-product credits), genuinely diversified across 12 mines, and only −33.5% off its February-2026 all-time high versus HL’s broken −51%. But “better silver knife” is still a knife: ROIC cleared its cost of capital in only one of the last five years (13.2% in record-2025; negative-to-1.6% in 2022–24), the name is now ~60–65% gold by revenue despite the brand, organic volume is flat-to-declining, and the entire equity is — by the factor model — a ~2.5x-levered bet on the gold price (GoldPrice beta 2.46–2.59, R² 0.81) with negligible idiosyncratic alpha. The market is underwriting realized silver ~$41/gold ~$3,459 — roughly 1.6–1.8x the company’s own ~$25/$2,100 reserve deck — as permanent. The framing is a cooling momentum / commodity-beta trade whose trend just cracked (m3/m6 momentum negative after a +59% year), not a bargain and not yet a fully-abandoned falling knife.

Conviction: medium. What flips me bullish: a reset toward ~1.2–1.5x book (high-$20s) or a structurally widening silver deficit that the reserve deck hasn’t caught up to. What flips me bearish (toward a genuine avoid/possible short for others): silver breaking back below ~$32 while PAAS still trades above ~$40 — peak multiple on collapsing peak earnings. I won’t short it: net cash, real long-life assets, a credible $1B-ish 2026 return program, and a metal that can stay irrational longer than a short can stay solvent. Tag: “A gold miner in silver clothing, priced for the metal never to mean-revert.”

📈 Stock Price Action — Five-Year Event Map

Factual price history, not a recommendation. Price moves are FACT (AZI five-year daily CSV); attributed drivers are INTERPRETATION. No price target, no support/resistance, no chart-pattern reading.

The arc. Over five years PAAS round-tripped from precious-metals darling to left-for-dead miner and back to an all-time record. It bottomed at a ~$11.68 close (2024-02-28) — the nadir of the 2021–2024 silver bear market — then rode the gold-and-silver melt-up to an all-time-high $68.30 close (2026-02-27), a ~5.8x advance in two years. It now trades $45.45 (2026-06-26), −33.5% off that high, inside a 52-week range of $26.66 (Jul-2025) → $68.30 (Feb-2026). The stock sits roughly at its still-rising 200-day EMA ($48.73), with multiples (P/B 2.6–2.85x, EV/EBITDA 11.8x) near the top of their decade — record-price context for a record-earnings, peak-metals year.

Cycle narrative. (1) From mid-2021 PAAS slid with the entire precious-metals complex as real rates rose and the dollar strengthened — a ~56% drawdown that had nothing to do with the company and everything to do with silver. (2) The March-2023 Yamana close (split with Agnico Eagle) re-shaped PAAS toward gold and lifted the share count ~55%; the market digested it as overhang, not upside. (3) Through 2023 into early 2024 the stock kept grinding lower on soft metals and integration/impairment worries, printing its five-year low of $11.68 on 28-Feb-2024 — the genuine value point in hindsight. (4) As gold broke to fresh records in 2024 the trade turned violently; the December-2024 La Arena sale tidied the portfolio while metals did the heavy lifting (~+70% off the low). (5) The 2025 leg paired a sustained metals rally with the MAG Silver deal (closed Sep-2025), adding the high-grade Juanicipio interest and nearly doubling the stock again. (6) The blow-off into February 2026 was a metals super-spike: silver/gold to historic highs drove record FY25 EBITDA and an ATH $68.30 close (27-Feb-2026), momentum and ETF flows amplifying. (7) Since the spike, metals have reversed and the stock is −33.5% off the high; recent daily action is pure metals beta — down on US–Iran de-escalation, up on renewed tension — with 3- and 6-month momentum now negative, a cooling of the trade rather than a company-specific break.

1. Executive Summary

Pan American Silver is the largest publicly-traded “silver” company by name and one of the largest precious-metals producers in the Americas — a Vancouver-headquartered, NASDAQ/TSX-listed operator of twelve mines across Canada, Mexico, Peru, Bolivia, Argentina, Chile and Brazil. FY2025 was a record on every line: revenue $3,619M (+28%), EBITDA $1,635M (45.2% margin), net income $978M, EPS $2.56, attributable free cash flow ~$1.15B, and a net-cash balance sheet. The stock has compounded ~5.8x off its February-2024 low.

Strip the headline and three facts govern the investment case. First, this is a no-moat commodity price-taker. PAAS sells bullion into a global spot market at a price it does not influence; its FY2025 record is a price event, not a franchise outcome. ROIC was negative in 2022, ~1.6% in 2023 and 2024, and reached 13.2% only at a once-in-a-decade silver/gold print — a business that clears its ~9–11% cost of capital exclusively at the top of the cycle. Second, the brand is a misnomer: roughly 60–65% of revenue is now gold, ~25–30% silver, the rest base metals. The “silver leverage” investors think they own is, in practice, majority gold leverage, with the marquee silver-growth story (the La Colorada Skarn) a 2030s, $1.9B, not-yet-sanctioned call option and the other silver optionality (Escobal) a stranded asset whose Indigenous consultation produced a formal denial of consent in May 2025. Third, the valuation prices peak metals as permanent. At $45.45 the stock trades at 2.6x book and 11.8x trailing EV/EBITDA — both near the top of their ten-year ranges (P/B 89.5th percentile, P/S 87.5th of the stock’s own history) — on an EBITDA earned at realized prices ~1.6–1.8x the company’s own reserve assumptions.

Against that, PAAS is unambiguously the higher-quality member of the silver-miner cohort: larger, lower-cost, net-cash ($363–500M), genuinely diversified (smoothing single-asset risk), and run by capable operators who pruned La Arena at a good price and funded Yamana counter-cyclically. It is not a fraud, not over-levered, and not a falling knife in the Hecla mold (−33.5% off its high, not −51%). The honest verdict is a quality-of-the-cohort name trading at a cyclical-high price: the asymmetry from here is unfavorable (base case — metals merely holding today’s historic levels — implies the enterprise is ~35% rich), but the downside is a metal-price call, not a balance-sheet one. This is a HOLD/avoid-adding situation for a fundamental investor, attractive only on a reset toward book or a structural deepening of the silver deficit.

2. Business Overview

What PAAS is. Pan American Silver Corp. (founded 1979; renamed from Pan American Minerals in April 1995) is a precious- and base-metals mining company headquartered in Vancouver, British Columbia. It explores for, develops, mines, processes and refines silver, gold, zinc, lead and copper across seven jurisdictions: Canada, Mexico, Peru, Bolivia, Argentina, Chile and Brazil (FACT — FactorsToday profile; 40-F FY2025, filed 2026-02-18). It is dual-listed (NASDAQ/TSX: PAAS) and a Canadian MJDS foreign private issuer, reporting in U.S. dollars and filing 40-F/6-K with the SEC.

Segments and assets. PAAS reports two operating segments plus a stranded asset:

• Silver Segment — La Colorada (Mexico), Huaron (Peru), Cerro Moro (Argentina, silver-gold), San Vicente (Bolivia, 95%/55% economic via a JV), and a 44% joint-venture interest in Juanicipio (Mexico), operated by Fresnillo and acquired through the ~$2.1B MAG Silver acquisition that closed September 2025. Juanicipio is the portfolio’s best single silver asset — high-grade, low-cost — and contributed equity-method income immediately ($61M in Q4-2025 alone, plus a $44M cash dividend) (FACT — Q4/FY2025 release).
• Gold Segment — Jacobina (Brazil), El Peñón and Minera Florida (Chile), Timmins/Bell Creek (Ontario), and Shahuindo (Peru). This segment is overwhelmingly the Yamana inheritance (closed March 2023) and is now the company’s larger cash engine.
• Escobal (Guatemala) — one of the world’s largest primary silver mines (~20 Moz/yr when operating), idle since 2017 pending a court-ordered ILO 169 Indigenous consultation (see ).

Production and mix. FY2025 attributable production was 22.8 Moz silver and 742.2 koz gold, plus 55.9kt zinc, 27.0kt lead and 3.0kt copper (FACT — FY2025 release). The economically decisive point is the revenue mix: at FY2025 realized prices (silver $40.78/oz, gold $3,459/oz), 742 koz of gold generates on the order of $2.5–2.6B versus roughly $0.9–1.1B from 22.8 Moz of silver — implying gold ~60–65% of revenue, silver ~25–30%, base metals the remainder (INTERPRETATION — production × realized price; reconcile to the segment note; total FY2025 revenue $3,619M). PAAS is, on the numbers, more a mid-tier gold miner than a silver pure-play — a distinction the name obscures and one that matters for how an investor should think about the underlying commodity exposure.

How it makes money. PAAS is a pure price-taker. It sells refined silver and gold bullion into the LBMA/COMEX spot market and sells base-metal concentrates to third-party smelters net of treatment and refining charges. There is no recurring revenue in the software sense; the only recurrence is geological — each mine produces until its orebody depletes, at which point reserves must be replaced through exploration or acquisition. Revenue is the product of two variables management does not control (metal price) and one it partly does (volume), against a cost base (AISC) that determines the margin. FY2025’s +28% revenue jump was almost entirely price: gold output actually fell ~17% year-over-year (892→742 koz, partly the La Arena sale), and silver volume rose only modestly (FACT — production releases).

The silver-growth pitch — real assets, distant cash. Two long-dated projects underpin the “silver growth” narrative. The La Colorada Skarn (revised PEA, March 2026) is a ~$1.9B-initial-capex, 37-year-life project averaging ~19.1 Moz/yr silver over its peak years — but the Board has approved only $265M of access-decline development (April 2026), not the project itself; a prefeasibility study is “a couple of years” away and first meaningful Skarn silver is a 2030s event (FACT — La Colorada Skarn PEA NR; Q1-2026 6-K). Escobal is binary and currently opposed. Neither contributes near-term cash.

Verdict: A large, diversified, currently very cash-generative commodity producer — but a gold-dominant, price-taking one whose silver-growth thesis is a 2030s option plus a stranded asset. The business is genuine and well-run; the brand oversells the silver exposure and the current results oversell the through-cycle earning power.

3. Industry Dynamics

Silver/gold mining is a structurally poor industry, and PAAS sits in a worse-than-average corner of it on jurisdiction. Three structural features define the silver market and frame the whole investment case.

(1) Supply is price-inelastic because most silver is a by-product. Roughly 70–75% of mined silver comes as a by-product of gold, copper and lead-zinc mining, where the silver credit is incidental to the economics of the primary metal (FACT — Silver Institute / World Silver Survey). A higher silver price therefore does not readily call forth new mine supply — primary silver mines are scarce, and by-product output responds to copper/zinc/gold economics, not silver. This is the bull’s strongest structural point: supply cannot easily flood in to cap the price.

(2) Demand is majority-industrial and partly elastic. Roughly 55–60% of silver demand is industrial (solar photovoltaics, electronics, brazing), with the rest investment (coins, bars, ETFs) and jewelry/silverware. The bull thesis leans on the deficit: the World Silver Survey 2026 (April 2026) reports a sixth consecutive structural deficit of ~46 Moz, widening from ~40 Moz the prior year (FACT). The skeptical read is essential: on ~1.1 billion ounces of annual demand, a ~46 Moz deficit is only ~4%, drawn down from very large above-ground ETF/vault inventories — a price-volatility/squeeze setup, not a guaranteed permanent floor. And it is partly self-correcting: 2026 industrial demand is forecast down ~3%, with solar-PV silver use cut ~19% as manufacturers thrift and substitute under margin pressure — direct evidence that “silver demand is inelastic” is false at the margin. What keeps the market tight is physical investment, up ~20% — i.e., reflexive, sentiment-driven hoarding that can reverse as quickly as it arrived (FACT/INTERPRETATION — World Silver Survey 2026).

(3) The capital cycle is at a late-stage high (Marathon lens). Record metal prices have produced record industry margins, record cash flows, and a wave of M&A and exploration spending — every marker of a late-cycle top. PAAS itself epitomizes the cycle: it acquired Yamana’s assets (2023) and MAG Silver (2025), funded with stock, into a rising metal price. The price-inelastic by-product supply dampens the eventual supply response (a genuine difference from, say, the shale oil cycle), but the swing factor — investment demand — is reflexive and two-sided. The honest base case is that current economics are peak, not normal, and that capital flooding into the sector at the top is precisely the condition that precedes mean reversion (INTERPRETATION).

Jurisdictional risk is PAAS’s distinguishing industry feature — and it is unfavorable. Unlike Hecla (US/Canada only) or the North American gold majors, PAAS concentrates assets in higher-risk jurisdictions: Mexico (2023 mining-law reform restricting new concessions, water permits and open-pit mining), Bolivia (resource nationalism, chronic FX scarcity), Argentina (capital controls, export taxes, peso instability — though the Milei reforms are easing this), Peru (community conflict, recurring political instability), and Guatemala (Escobal stranded by ILO 169). These are not tail risks; they are recurring, value-impairing realities embedded in the asset base (FACT — 40-F risk factors).

Profit pools and value chain. The precious-metals value chain runs explorers → developers → producers (PAAS) → refiners → bullion banks/end-users, with royalty/streaming companies (Franco-Nevada, Wheaton) sitting above the producers and capturing a structurally superior share of the economics — no operating cost, no sustaining capex, no depletion-capital burden, and durably high through-cycle ROIC. That contrast is the cleanest illustration of why operating mining is a bad business: the part of the chain with a moat is the part PAAS is not in (INTERPRETATION — FNV/WPM peer reports).

Verdict: Structurally mediocre-to-bad industry, currently at a cyclical high. No pricing power, capital-intensive, reserve-depleting, and reliant on a deficit narrative that is real but small (~4% of demand) and partly self-correcting. PAAS carries above-average jurisdictional risk versus its North American peers. Current margins and cash flows should be treated as peak-of-cycle, not normalized.

4. Competitive Position

Start from Greenwald’s null hypothesis: a commodity miner has no moat. There is no demand captivity (every seller receives the identical LBMA/COMEX price), no switching costs, no network effects, no brand premium on a bar of bullion. The entire competitive-advantage question reduces to one thing: does PAAS possess a durable cost advantage — an orebody or a scale position that lets it earn above its cost of capital through the full cycle? The honest answer is no franchise, only scale.

Cost position — competitive, but a portfolio average, not a crown jewel. PAAS’s FY2025 Silver Segment AISC of $13.88/oz is genuinely good — well below First Majestic’s ~$20+/oz (AgEq) and respectable against the better primary producers — helped by rising gold/zinc by-product credits and Yamana synergies (FACT — FY2025 release). But this is a blended portfolio average across twelve mines, not a single bottom-of-the-cost-curve asset. The contrast with Hecla is the sharpest competitive point in the sector: HL owns one world-class asset, Greens Creek, whose by-product credits drove a negative silver AISC (−$2.36/oz) in 2025 — a genuine cost-curve outlier. PAAS has no equivalent crown jewel. Juanicipio (44%) is the closest, and it is excellent, but PAAS owns less than half of it, does not operate it, and management itself warns the silver grades zone downward into base metals over time. PAAS’s edge is scale and geographic diversification, which smooths single-asset and single-jurisdiction risk — real and valuable for survivability — but is, in Greenwald’s terms, closer to diversification than to advantage: a collection of average-cost, depleting, price-taking assets, none individually exceptional, none driving consolidated returns above the cost of capital through the cycle (INTERPRETATION).

The acid test — does the claimed advantage show up in financial outcomes through the cycle? It does not. PAAS’s ROIC was negative in FY2022, ~1.6% in FY2023 and FY2024, and reached 13.2% only in record-price FY2025 (ROE 16.9%) (FACT — ROIC profitability ratios). A business that clears its ~9–11% cost of capital only at the very top of the commodity cycle is the textbook signature of a no-moat cyclical, not a compounder. Market-share stability — another Greenwald test — is irrelevant in a market where “share” confers no pricing power and is bought, not earned. Scale gives PAAS lower single-asset risk and an organic growth pipeline (the Skarn) that sub-scale peers cannot fund — but scale in a price-taking business is not a moat; it is simply more of an undifferentiated thing.

Direct peer map.

• vs. Hecla (HL): HL has the crown-jewel low-cost asset (Greens Creek) and clean US/Canada jurisdiction; PAAS has neither, but offers materially more scale/diversification, a net-cash balance sheet, and a real (if distant) silver-growth pipeline. On valuation PAAS is the less-extreme of the two (2.6–2.85x P/B vs HL ~3.9x P/TBV; −33.5% off the high vs HL −51%).
• vs. gold majors (Newmont, Agnico): PAAS is smaller, shorter-life, lower-grade, and carries higher jurisdiction risk; Agnico in particular runs a far higher-quality, lower-risk (Canada/Finland/Australia) portfolio.
• vs. Franco-Nevada / Wheaton (royalty/streaming): these are the closest thing in the sector to a genuine moat — asset-light, no operating cost, no capex, no depletion-capital burden, durably high ROIC — and PAAS has none of those structural features. An investor wanting precious-metals exposure with a moat buys the royalty model, not the miner (INTERPRETATION — HL/NEM/AEM/FNV/WPM reports).

Verdict: No durable franchise moat. PAAS is a no-moat, price-taking commodity producer whose only honest advantage is scale plus geographic diversification. It is larger and better-spread than HL, CDE or AG, but it lacks a single low-cost crown-jewel asset and carries worse jurisdiction. The “premier silver company” branding is marketing; the financial reality is a diversified collection of depleting, roughly average-cost, price-taking assets that clear WACC only at peak metal prices — currently flattered by a record price environment and by a gold-weighted mix the “silver” name conceals.

5. Growth History and Forward Opportunities

The headline growth is real in dollars and almost entirely hollow in substance. Revenue compounded from $1,633M (FY21) to $3,619M (FY25) — a ~2.2x jump — but a decomposition shows essentially none of it is organic volume. Two forces explain the curve: acquired production and metal price.

Volume by year — the decisive tell. Strip out the acquisitions and organic output has been flat-to-declining:

• Silver: ~18.5 Moz (FY22) → ~20.4 Moz (FY23, already including nine months of Yamana) → ~21.1 Moz (FY24) → 22.8 Moz (FY25, including only ~3.5 months of Juanicipio post-MAG). FY26 guidance is 25–27 Moz. The FY25→FY26 “+14% silver” step-up management touts is the full-year run-in of an acquired 44% Juanicipio stake, not mine-level organic growth (FACT — production releases; FY26 guidance).
• Gold: ~552 koz (FY22) → 882.9 koz (FY23, +60% — the entire gold base is Yamana: Jacobina, El Peñón, Cerro Moro, Minera Florida) → record 892.5 koz (FY24) → 742.2 koz (FY25, DOWN ~17%) as El Peñón low-grade stockpiles exhausted, Dolores residual leaching declined, and La Arena was sold. FY26 guidance 700–750 koz — gold peaked in FY24 and is now structurally lower (FACT).

The acquisitions were paid largely in stock: diluted shares went from 210M (2021) to 381M (2025) and 421.8M outstanding at year-end 2025 post-MAG — roughly +100% in four years. Per-share volume growth is therefore materially worse than the already-weak headline. The realized-price tailwind did the rest: FY25 realized silver ~$40.78/oz (Q4 averaging ~$58) against an ~$25/oz reserve deck — peak-of-cycle, not a structural earning-power improvement.

Forward pipeline — long-dated and uncertain.

1. Juanicipio (44%) — the one genuinely high-quality, low-cost asset; FY26 is its first full contributing year. But management itself warns the silver grades decline into base metals over time as the geology zones downward; this is a windfall to be enjoyed, not a growth ramp to extrapolate.
2. La Colorada Skarn — the marquee silver-growth story, and a 2030s call option. The revised PEA (March 2026) cut the design to a 15,000 tpd long-hole-stope operation, lowering capex ~$1B to ~$1.9B, with a 37-year life and ~19.1 Moz/yr over the peak five years. But the Board approved only $265M — a 12.4 km access decline over five years — not the project; a PFS is years away, and surface/plant infrastructure is deferred. First meaningful Skarn silver is a 2030s, $1.9B, not-yet-sanctioned event (FACT — Skarn PEA NR).
3. Jacobina optimization (flowsheet simplification, paste backfill) — incremental, engineering-stage.
4. Timmins/Bell Creek shaft extension (~$131M, extends life to ~2040–46) — a life extension, not growth.
5. Escobal — binary and currently opposed.

Verdict: Low-quality growth. The revenue ramp is acquired plus price-driven; organic silver volume is flat-to-down, gold has peaked and is declining, and per-share volume has been diluted by ~100% more shares. The forward “silver growth” rests overwhelmingly on a 2030s, $1.9B, partially-sanctioned project, with near-term “growth” being the acquired Juanicipio full-year dressed as momentum. This is acquisition-and-price growth wearing the costume of organic growth.

6. Financial Quality

FY2025 is, on the surface, a triumph — and underneath, a metal-price artifact. Revenue $3,619M (+28%), EBITDA $1,635M (45.2% margin), net income $978M, EPS $2.56, ROE 16.9%, ROIC 13.2%. Strip the surface and almost the entire improvement is price, not the business. Attributable silver production (22.8 Moz) and gold (742 koz) were flat-to-down; what moved was the realized price: silver $40.78/oz (+45% YoY), gold $3,459/oz (+45%). Against Silver-Segment AISC of $13.88/oz, the FY25 cash margin on silver is roughly $27/oz (~66%). This is a price-driven peak in unit economics, full stop (FACT — Q4/FY2025 release).

The margin trajectory makes the cyclicality unmistakable. EBITDA margin ran 16.7% → 24.5% → 32.7% → 45.2% across FY22–25, tracking the silver tape almost one-for-one. The decisive tell is ROIC across the cycle: negative in 2022, ~1.6% in both 2023 and 2024, and 13.2% at the FY25 record. For a producer with a WACC of roughly 9–11%, PAAS earned below its cost of capital in three of the prior five years and clears it by only 2–4 points even at a once-in-a-decade silver price. A business with a genuine cost-curve advantage shows positive ROIC through the trough; PAAS does not. Returns here are a derivative of the metal price (INTERPRETATION — ROIC ratios).

Quality of earnings — clean this year, but do not extrapolate the series. FY25 operating cash flow was ~$1,333M against net income $980M (CFO/NI ~1.36x), which looks healthy and largely is — the gap is D&A ($497M) plus deferred tax, netted against a “other non-cash” line that reverses the $137M La Arena disposal gain and the not-yet-fully-cash equity-method Juanicipio income, with a small working-capital drag. The critical caveat is multi-year: PAAS’s OCF is inversely cyclical through precious-metal inventory and receivable swings — FY22 OCF collapsed to $31.8M on a $340M loss as inventory built into the price decline, while FY20 enjoyed a working-capital release. FY25 is a “clean” year, but the series must be normalized; do not extrapolate FY25 OCF. SBC is trivial (~$2M) — the dilution here is all M&A, not equity comp. FCF was ~$1,019M reported-to-firm (company-reported attributable FCF $1,151M), or ~$2.67/share (FACT — cash flow statement; reconcile to 40-F).

Balance sheet — the genuine quality. This is a fortress: cash and short-term investments ~$1,215–1,319M, total debt $852M (two senior notes plus leases), net cash ~$363–500M, net-debt/EBITDA negative, interest coverage ~30x, current ratio ~2.7x, and total liquidity ~$2.07B including a $750M undrawn facility (FACT — balance sheet; enterprise-value tool). There is no maturity stress and ample capacity to fund the Skarn decline and the 2026 return program. One subtlety for the risk section: IFRS purchase accounting allocated the Yamana/MAG consideration largely to mining property rather than goodwill (ROIC shows ~$0 goodwill), so there is no goodwill cushion — a future write-down on a silver reversion toward the ~$25/oz reserve deck would hit PP&E directly.

Unit economics. The entire model reduces to AISC vs. realized price. FY25: silver AISC $13.88 vs. realized $40.78 → ~$27 margin; gold AISC $1,621 vs. realized $3,459 → ~$1,838 margin. FY26 guidance already signals cost creep (silver AISC $15.75–18.25, gold AISC $1,700–1,850) even before any price reversion — meaning the margin is squeezed from both ends if metals soften (FACT — FY26 guidance).

Verdict: Economics do not improve with scale in the way that matters. Yamana and MAG bought diversification, a far larger absolute FCF base, and lower single-asset risk — genuinely valuable — but they did not lift through-cycle returns. ROIC remains a metal-price function: sub-WACC in the trough, ~13% at the peak. The quality of this company is its balance sheet and asset diversity, not its returns on capital. FY2025 is a cyclical high printed at a cyclical-high metal price.

7. Capital Allocation

This is the crux for a serial miner, and the record is a genuine barbell — counter-cyclical discipline next to top-of-cycle dilution, bound together by an incentive system that rewards the dilution. PAAS is a serial, stock-funded acquirer: diluted share count went from 210M (2021) to ~326M (Yamana close, 2023) to a 381M FY25 average and 421.8M outstanding at year-end post-MAG — roughly +100% in four years, almost entirely M&A scrip.

(a) Yamana Gold (closed Mar-31-2023) — the favorable, counter-cyclical leg. A ~US$4.8B transaction executed with Agnico Eagle (Agnico took the Canadian assets including Canadian Malartic). PAAS issued ~153.8M shares plus ~$1B cash (the cash funded by Agnico’s portion) for Jacobina, El Peñón, Minera Florida, Cerro Moro and the MARA development project. The key fact is timing: struck Nov-2022, closed Mar-2023, with silver near $20–23 and PAAS stock near $15–17 — PAAS paid in cheap scrip near a cyclical low. That is the right side of the Marathon capital cycle. Per-share dilution was severe (~+55% shares) and the deal dragged ROIC through the FY23–24 trough, but Jacobina’s gold is now a core engine of FY25 cash flow. Ugly near-term per-share optics; defensible long-term logic (FACT — transaction disclosures).

(b) MAG Silver (announced May-2025, closed Sep-2025) — the unfavorable, pro-cyclical leg. ~US$2.1B / $20.54 per MAG share, a mix of $500M cash + 0.755 PAAS shares, ~21% prorated premium (27% to the 20-day VWAP); MAG holders ended ~14% of PAAS. PAAS acquired 44% of Juanicipio, a premier low-cost, high-grade silver mine. The asset is genuinely top-tier — but the structure is the textbook value-destroyer: paying a premium for silver beta near the silver high, largely in stock that was itself elevated. The two deals bracket the cycle perfectly: cheap scrip at the low (Yamana), pricey scrip at the high (MAG) (FACT).

© Divestiture discipline — the best part of the record. La Arena (Peru gold mine + La Arena II Cu-Au project) was sold to Zijin (completed Dec-2024) for $245M cash + $50M contingent + a retained 1.5% gold NSR royalty, booking a ~$137M gain. Morococha and Manantial Espejo were wound down as non-core. Management does prune higher-jurisdiction-risk assets at decent prices and retains royalty optionality — a discipline many peers lack (FACT).

(d) Dividend / (e) buyback. The dividend is variable and net-cash-linked, raised three times through 2025 (FY25 $0.459/sh); the new 2026 framework targets returning 35–40% of attributable FCF via dividends + buybacks, with “up to $1 billion” of total 2026 returns and a $0.18/quarter dividend (~$305M). Shareholder-friendly in intent — but an FCF-linked payout rises pro-cyclically, returning the most cash at the peak (income-friendly, not counter-cyclical discipline). The NCIB so far is cosmetic: FY25 repurchases were just ~1.65M shares at ~$27.92 ($46M) and Q1-26 only ~460k — a rounding error against $1.0B+ of FCF and against the ~$1.5B of new shares issued for MAG the same year. PAAS is a massive net issuer, not a repurchaser, and the 2026 buyback is back-end-loaded and execution-unproven, now running into a falling metal price (FACT — NCIB disclosures; 2026 return framework 6-K, May-2026).

(f) Impairments. A modest serial-acquirer tally: ~$99M (FY22) and ~$79M (FY23) of asset impairments clustering at the trough; no impairment in FY25. With no goodwill cushion, any future write-down on a metal reversion would hit mining property directly.

(g) Incentive alignment — no return hurdle, and that is the structural problem. The long-term incentive is PSUs vesting on relative TSR against an internally-selected peer group (0–200%, capped); the annual incentive is a scorecard of production, AISC/cost, ESG, safety and diversity goals. There is no ROIC, return-on-capital, or per-share-value metric anywhere in the plan (FACT — management circular). This is precisely the incentive structure that produces dilutive, top-of-cycle M&A: management is paid for growth and relative TSR, not for capital efficiency or per-share value. A shareholder formally complained that pay did not align with TSR. There is no controlling shareholder; CEO Michael Steinmann’s direct stake is modest (~145k shares) against a PSU/RSU-heavy package — low absolute insider skin-in-the-game for a 422M-share company.

(h) The decisive per-share test — failed through FY24. Revenue/share ran $6.37 → $7.76 → $7.10 → $7.09 → $7.76 → $9.49 (FY20–25): essentially flat 2021–2024 despite doubling both the share count and the asset base, only breaking out in FY25 on price. FCF/share: $1.35 / $0.71 / −$1.15 / $0.22 / $1.10 / $2.67 — no durable per-share FCF growth until the FY25 price spike. Through FY24, the serial dilution roughly offset the asset growth on a per-share basis — growth was swallowed by scrip. The FY25 per-share breakout is the silver price, not allocation alpha (INTERPRETATION — per-share data).

Verdict: Adequate, not intelligent in the compounding sense. The good (counter-cyclical Yamana scrip, disciplined La Arena divestiture with retained royalty, fortress balance sheet, a sensible FCF-linked return framework) sits beside the textbook bad (pro-cyclical MAG premium for silver beta at the top) and an incentive plan with no capital-return hurdle that structurally encourages exactly that behavior. On the decisive per-share test, management did not compound per-share value through FY24, and FY25’s apparent success is metal price. These are capable operators and opportunistic dealmakers — but they are price-takers running a production-and-TSR incentive, not per-share-value compounders.

8. Changes and Headwinds — Last Two Years

The two-year change set skews toward pro-cyclical and binary-overhang items, wrapped in record (price-driven) results.

• Yamana Gold (closed Mar-2023, ~$4.8B PAAS share, with Agnico) — transformative; created the gold segment, acquired counter-cyclically. Strengthens the asset base; dilutes per share.
• La Arena sold to Zijin (Dec-2024, $245M + contingent + gold NSR) — disciplined portfolio pruning. Strengthens.
• MAG Silver (closed Sep-2025, ~$2.1B, cash+stock) → 44% of Juanicipio — high-quality asset bought pro-cyclically with silver near record highs; the bear’s “top-of-cycle M&A” hook. Mixed.
• La Colorada Skarn revised PEA (Mar-2026) + $265M first-tranche Board approval (May-2026) — this is the decline only, not sanction of the $1.9B project; partner discussions disclosed but unpriced. Neutral-to-positive optionality, uncommitted.
• Escobal — Xinka Parliament DENIED consent (May-2025) after an ~8-year ILO-169 process, demanding permanent closure (per Earthworks/Northern Miner reporting). Management’s Q4-2025 framing (“critical minerals…will accelerate”) is contradicted by the Xinka denial; there is no restart timeline. Treat management optimism as hypothesis, not evidence. Weakens / binary overhang.
• 2026 Shareholder Return Framework (May-2026) — 35–40% of attributable FCF, up to $1B in 2026; dividend raised three straight quarters. But execution is back-end-loaded and unproven, now into a falling metal price. Positive intent, unproven.
• Record results that are peak-price artifacts — FY25 NI $980M/EPS $2.56, attributable FCF $1.2B; Q1-26 NI $456M/$1.08, FCF $488M. PAAS is unhedged on both metals — full leverage in both directions.
• Management — Michael Steinmann remains President & CEO; no succession announced.
• The price round-trip itself — ATH $68.30 (Feb-2026) → $45.45 (Jun-2026), −33.5%, as the metals super-spike reversed (see the Five-Year Event Map above).

Verdict: Mixed, net slightly negative for thesis quality. The disciplined moves (Yamana, La Arena) are real, but the recent set skews toward pro-cyclical M&A (MAG), a long-dated/uncommitted growth project (Skarn), an opposed/binary asset (Escobal), and a back-end-loaded, unproven return program — all wrapped in record results that are price-driven, not structural.

9. Risk Analysis

The dominant risk by an order of magnitude is metal price; everything else is secondary. Because PAAS is unhedged with a GoldPrice beta of ~2.5, a metal reversion drives the equity directly and with leverage.

Catastrophic-loss assessment: very low probability. The diversification that is not a moat is genuinely valuable here — no single mine or jurisdiction can sink the company, and the net-cash balance sheet removes solvency risk. The realistic bad outcome is a large drawdown on metal reversion (the −85% lifetime max drawdown shows the magnitude possible), not a permanent capital impairment.

10. Valuation Discussion (Embedded Expectations)

The single variable that matters is the metal price. At $45.45 (2026-06-26), market cap ~$19.8B and EV ~$19.3B, PAAS trades at 11.8x trailing EV/EBITDA and 2.6–2.85x book — and every valuation judgment reduces to one question: will silver near ~$41/oz and gold near ~$3,459/oz — the realized prices behind FY25 — persist? The market’s answer, at this price, is “yes, indefinitely.”

Current multiples sit at the top of PAAS’s own decade (FACT — ROIC get_valuation_multiples, 10-yr, accessed 2026-06-27):

The AZI valuation_index reconciles: P/B 89.5th and P/S 87.5th percentiles of the stock’s own history are the honest tell — PAAS trades near its own ceiling. The P/E 4.4th percentile is an artifact and should be ignored: FY25 GAAP EPS ($2.56) is a fresh peak after near-zero/negative earnings in FY22–24, so the trailing-P/E percentile mechanically reads “cheap” off a denominator that just inflected from a loss. Use P/B, P/S and EV/EBITDA — all of which say expensive-vs-own-history.

Embedded-expectations math. FY25 EBITDA of $1,635M (45.2% margin) was earned on realized silver $40.78 and gold $3,459 — roughly 1.6–1.8x the company’s reserve-deck assumptions of ~$25 silver / ~$2,100 gold (the prices at which PAAS books reserves and plans mines). At 11.8x trailing EV/EBITDA on a peak-of-cycle EBITDA, the stock embeds both a peak commodity print and a peak multiple — a double-peak. Stripping either one out collapses the case.

Scenario analysis (metal-price-driven; the EBITDA figures are ASSUMPTIONS built off the ~45% FY25 margin and the gold-weighted mix):

*Equity per share ≈ (implied EV + ~$0.4B net cash) / 421.8M shares. Illustrative; rounds the scenario EV ranges.

The asymmetry from here is unfavorable. Even the base case — metals merely holding near today’s already-historic levels — implies the EV is ~33% rich, because the market is paying a peak EBITDA multiple on a peak-price EBITDA. Only the bull case (metals sustaining and rising from a near-100-year-high real-silver print) roughly justifies the current price, and it leaves no margin of safety. The bear case — a partial reversion toward the reserve deck, which is simply the long-run mean PAAS itself plans around — implies severe downside, with EBITDA potentially halving on operating leverage that cuts both ways.

P/NAV and FCF yield point the same way. At 2.6–2.85x book — well above the ~1.0–1.3x P/NAV at which quality silver miners historically trade through-cycle — PAAS offers no NAV cushion, and the consensus NAV itself is computed on elevated spot decks, compounding the optimism. The ~6% trailing FCF yield is genuine but is a peak-cycle yield; on normalized metals it compresses toward 3–4%, unremarkable for a depleting price-taker.

Peer cross-read. PAAS’s profile mirrors Hecla — both re-rated to richest-ever own-history multiples on peak metals — but PAAS is the less extreme, higher-quality version: 2.6–2.85x P/B vs HL ~3.9x P/TBV; −33.5% off the ATH vs HL −51%; net cash and lower-cost. Against the broader silver set (CDE, AG, EXK, FSM, SSRM) PAAS screens as the scale leader at a full — not insane — multiple.

Verdict: PAAS is priced for the continuation of peak metals. It is not cheap on any normalized basis; it is fairly-to-richly valued on spot and expensive on mid-cycle. No price target — the ranges above are embedded-expectations scenario analysis only.

11. Variant Perception

Consensus view. PAAS is owned as the premier, diversified, lower-risk way to play a structural silver bull market — a scaled operator with a fortress balance sheet, a rising dividend, a $1B return program, and a multi-decade silver-growth pipeline (the Skarn). The sell-side frames it as quality-at-a-fair-price with leverage to a widening silver deficit.

The strongest bull case. Silver is in its sixth consecutive structural deficit; by-product supply is price-inelastic and cannot respond; industrial demand (solar, electrification) plus reflexive investment demand keep the market tight; and a generational monetary/de-dollarization backdrop sustains gold. In that world, PAAS’s unhedged ~2.5x gold beta, net-cash balance sheet, and growth optionality make it a high-quality vehicle that re-rates further, and today’s −33.5% pullback is a buying opportunity in an ongoing bull market.

The strongest bear case. PAAS is a no-moat price-taker trading at richest-ever own-history multiples (P/B 89.5th, EV/EBITDA 11.8x vs 7.0x avg) on a double-peak — peak EBITDA at a peak multiple. ROIC clears WACC only at the top of the cycle; the silver deficit is small (~4% of demand) and partly self-correcting (solar thrifting −19%); the “silver” exposure is really ~60–65% gold; organic volume is flat-to-declining; growth is a 2030s, $1.9B, un-sanctioned option plus a stranded, opposed Escobal; and management dilutes pro-cyclically with no per-share/ROIC incentive hurdle. A partial mean-reversion toward the reserve deck — the company’s own long-run planning assumption — implies severe downside.

The 3–5 assumptions that actually matter:

1. Will realized silver/gold hold ~1.6–1.8x the reserve deck? (The whole valuation rests here.)
2. Is the silver deficit a durable price floor or a small, reflexive, self-correcting imbalance?
3. Does PAAS sanction and build the Skarn on time/on budget — or is the “silver growth” perpetually ~7 years away?
4. Does management refrain from another pro-cyclical, dilutive deal at the top?
5. Does Escobal stay stranded (consensus) or surprise on a restart (deep optionality)?

Falsification evidence. The bull breaks if silver falls back below ~$32 while PAAS still trades above ~$40 (peak multiple on collapsing earnings), or if solar thrifting and investment-demand reversal turn the deficit into balance. The bear breaks if the deficit demonstrably widens beyond above-ground stocks’ ability to fill it (a true squeeze), or if PAAS resets to ~1.2–1.5x book and the metal holds — at which point the quality and net cash make it attractive.

Factor-positioning input. PAAS is, statistically, not a stock but a ~2.5x-levered long on the gold price (GoldPrice beta 2.46–2.59, All-Factors R² 0.81), with essentially zero style loading (Value ~0.05, Momentum ~−0.05, Quality negative) — almost all return is the precious-metals complex, negligible idiosyncratic alpha. The trend has cracked but not fully broken: y1 +59%/Sharpe 1.02 and y3 +49%/Sharpe 0.93 against m3 −31.7%/Sharpe −0.58 and m6 −29.3%/Sharpe −0.49 (annualized), with rs_6m flipping to −15% while rs_12m is still +60.6%. The frame is a cooling momentum / commodity-beta trade whose trend just broke — not yet a fully-abandoned falling knife in the Hecla mold (PAAS sits ~0.93x its still-rising 200-EMA; HL is decisively broken). For variant perception, the positioning risk is that the momentum cohort that drove PAAS up is the same flow that unwinds it, and the factor model says there is no quality or value floor to catch it if metals revert — the downside is the gold beta in reverse, magnified ~2.5x.

12. Fact vs. Interpretation Table

13. Open Questions

1. What is the exact FY25 revenue split by metal and segment? (Reconcile the ~60–65% gold estimate to the 40-F segment note.)
2. What metal-price deck does PAAS actually use for reserves at year-end 2025 — has it been lifted from ~$25/$2,100 toward spot, and if so, how much “reserve growth” is just price?
3. What are the Skarn’s full sanction economics (IRR, payback, NPV) at a normalized vs. spot deck, and what is the realistic FID date?
4. What is the carrying value of Escobal, and what impairment would a formal permanent closure trigger?
5. Will the 2026 buyback actually execute at $1B, and at what average price — counter-cyclically into weakness, or not at all?
6. Insider activity: as a SEDI (not Form 4) filer, are insiders buying or selling at these levels? (EDGAR cannot answer; SEDI required.)
7. What is management’s normalized through-cycle AISC trajectory given FY26 cost creep — is the cost curve drifting up structurally?

14. What Must Be True

Bull case — what must be true:

• Silver and gold sustain near or above today’s realized levels (~$41/~$3,459) for years, i.e., the deficit is a durable floor, not a reflexive spike.
• Industrial demand (solar) does not thrift away the deficit, and investment demand stays elevated.
• PAAS sanctions and builds the Skarn roughly on time/budget, converting the 2030s option into real silver growth.
• Management resists further pro-cyclical dilution and executes the $1B return program counter-cyclically.
• Falsification test: silver closes below ~$32/oz for a sustained period while PAAS still trades above ~$40 — a peak multiple on collapsing earnings — or the silver market moves into balance/surplus on solar thrifting. Either falsifies the bull.

Bear case — what must be true:

• Metals partially revert toward the reserve deck (silver ~$28–32, gold ~$2,800), compressing EBITDA toward ~$650–800M.
• The richest-ever multiple de-rates toward the 10-yr average (EV/EBITDA 11.8x → ~7–8x) as the momentum cohort unwinds.
• Growth stays a perpetual 2030s promise; Escobal stays stranded.
• Falsification test: the silver deficit demonstrably widens beyond above-ground stocks’ capacity to fill it (a genuine squeeze, silver structurally above ~$45) or PAAS resets to ~1.2–1.5x book while metals hold — at which point quality + net cash make it attractive and the bear is wrong.

15. Source Appendix

See the separate Source Appendix (Appendix B in the combined report) for the full citation list. Primary sources: PAAS 40-F (FY2025, filed 2026-02-18) and 6-K filings (SEC EDGAR, CIK 0000771992); PAAS Q4/FY2025 and Q1-2026 news releases and earnings-call transcripts; La Colorada Skarn revised PEA (Mar-2026); 2026 Shareholder Return Framework / Investor Day (May-2026); management information circular. Quantitative data: ROIC.ai (statements, ratios, enterprise value, valuation multiples — reconciled to the 40-F); AZI price CSV and valuation_index percentiles; FactorsToday factor model (loadings, leaderboard, related stocks). Industry: Silver Institute / World Silver Survey 2026. Peer cross-read: prior published research on HL, CDE, AEM, NEM, FNV, WPM. Escobal/Xinka: Northern Miner, Mining.com, Earthworks. All non-obvious facts cited inline with source and date; figures are FACT unless labeled INTERPRETATION or ASSUMPTION.

APPENDIX A — Standard Diligence Questionnaire

Pan American Silver Corp. (NASDAQ: PAAS) — 2026-06-27

Supplemental diligence. Labels: FACT / INTERPRETATION / ASSUMPTION.

General

What thoughtful questions have other investors asked about this company? Chiefly: (1) Is PAAS really a silver play or a gold miner with a silver brand? (INTERPRETATION: the latter — ~60–65% of revenue is gold.) (2) Is the La Colorada Skarn a real growth driver or a perpetual 2030s option? (3) Will Escobal ever restart? (4) Was the MAG Silver deal value-accretive or pro-cyclical dilution at the top? (5) Why does the incentive plan have no return-on-capital hurdle? (6) Is the dividend/buyback program durable through a metal-price reversion?

Cyclicality & Earnings Nature

Cyclical high or low? FACT: Unambiguous cyclical high. FY25 EBITDA margin 45.2% (vs 16.7% in FY22); ROIC 13.2% (vs negative in FY22). Realized silver $40.78/gold $3,459 ≈ 1.6–1.8x reserve deck. External environment or internal action? INTERPRETATION: ~80%+ external (metal price). FactorsToday GoldPrice beta ~2.5, R² 0.81 — the equity is a levered metal-price proxy. Revenue stability? FACT: Highly unstable; revenue $1.5B (FY22) → $3.6B (FY25), driven by acquisitions + price. No recurring revenue. Product/market outlook & size? FACT/INTERPRETATION: Silver market ~1.1B oz demand, ~46 Moz deficit (~4%); growing industrial use (solar) but partly elastic (solar thrifting −19% in 2026). Gold market large and global. Both are commodity markets where PAAS is a price-taker.

Business Quality & Competitive Moat

More or less competitive? INTERPRETATION: Late-cycle capital inflow (M&A, exploration) typically precedes more competition for assets and mean-reverting returns (Marathon lens). ROIC / ROE? FACT: ROIC 13.2% / ROE 16.9% (FY25 peak); negative-to-1.6% ROIC FY22–24. Clears WACC (~9–11%) only at the cycle top. Industry profitability / barriers / competitors? INTERPRETATION: Capital-intensive, reserve-depleting, no pricing power; barriers (permits, orebody scarcity) protect the industry’s high prices but not any single producer’s returns. Peers: HL, CDE, AG, FSM, EXK, SSRM (silver); NEM, AEM (gold); FNV, WPM (royalty — the moat model). Easily understood? FACT: Yes — production × (price − AISC). Undermined by low-cost foreign labor? N/A — orebody-located; cost driven by grade/geology/jurisdiction, not labor arbitrage. Do brands matter? FACT: No — bullion is fungible at the LBMA price. “Pan American Silver” is corporate, not product, brand. Nature of competition? INTERPRETATION: Competition for assets and capital, not customers. No moat. Switching costs? None — there is no customer to retain.

Financial Condition & Balance Sheet

Unrecognized assets? INTERPRETATION: Escobal (idle, ~20 Moz/yr optionality, opposed); La Colorada Skarn resource; retained La Arena 1.5% gold NSR royalty; exploration upside. None reliably cash-generative near-term. Off-balance-sheet liabilities? FACT/OPEN QUESTION: Mine-reclamation/closure provisions (on balance sheet under IFRS); operating-lease and smelter-contract obligations — quantify from 40-F. Escobal closure/social obligations are an overhang. Accounting conservatism? INTERPRETATION: Moderate. IFRS allocated M&A consideration to mining property, ~$0 goodwill — no goodwill cushion, so future write-downs hit PP&E directly. GAAP/IFRS earnings distorted historically by impairments (FY22 ~$99M, FY23 ~$79M) and disposal gains (La Arena +$137M). CapEx-hungry? FACT: Yes — sustaining capex plus a $1.9B Skarn ahead; mining is structurally capital-intensive and depletion requires continuous reinvestment.

Capital Allocation & Management

FCF generation & use & philosophy? FACT: FY25 attributable FCF ~$1.15B. 2026 framework: return 35–40% of attributable FCF (up to $1B) via dividend (~$305M) + buyback. Philosophy: FCF-linked (pro-cyclical) returns + opportunistic M&A. Significant acquisitions? FACT: Yamana (Mar-2023, ~$4.8B, counter-cyclical/cheap scrip — GOOD timing); MAG Silver (Sep-2025, ~$2.1B, pro-cyclical/pricey scrip — POOR timing). La Arena divested to Zijin (Dec-2024) with discipline. Buying back shares? FACT: Cosmetically — ~$46M FY25 (~1.65M sh @ ~$27.92) vs ~$1.5B new shares for MAG. Net issuer, not repurchaser. Issuing shares to insiders? FACT: Trivial SBC (~$2M). Dilution is all M&A scrip (210M → 421.8M shares). Director/management comp? FACT: Relative-TSR PSUs (0–200%) + production/cost/ESG/safety AIP. No ROIC or per-share-value metric. CEO Steinmann direct stake modest (~145k sh). Management motivations? INTERPRETATION: Paid for growth and relative TSR, not per-share value — structurally incentivized toward dilutive, top-of-cycle M&A.

Valuation & Market Data

ADR/MLP/K-1? FACT: No — Canadian MJDS foreign private issuer (40-F/6-K), common shares, USD reporting. No K-1. Dividend policy? FACT: Variable, net-cash-linked; FY25 $0.459/sh; 2026 $0.18/qtr (~$305M). Yield ~1.4%. Profitability? FACT: Peak — see ROIC/margins above. NI vs CFO divergence? FACT: FY25 CFO/NI ~1.36x (clean). But OCF is inversely cyclical multi-year (FY22 OCF $31.8M); do not extrapolate FY25.

Risks & Downside

What causes the stock to decline? FACT/INTERPRETATION: Metal-price reversion (dominant), multiple compression from richest-ever levels, jurisdictional shocks, Escobal closure, dilutive M&A, AISC creep. Catastrophic loss risk? INTERPRETATION: Low — diversified (12 mines/7 countries), net cash, no solvency risk. The realistic bad case is a large drawdown on metal reversion (−85% lifetime max drawdown shows the magnitude), not permanent impairment. Total loss? Highly improbable absent a systemic metals collapse.

Recent News & Events

Environment changed recently? FACT: Yes — metals super-spike to ATH $68.30 (Feb-2026) then −33.5% reversal; momentum rolled over (m3/m6 negative). AZI feed shows only macro precious-metals moves (US–Iran geopolitics) — zero idiosyncratic company news, net-negative skew into late June. Significant acquisitions? FACT: MAG Silver closed Sep-2025 (Juanicipio 44%). Accounting-policy change? OPEN QUESTION: None identified beyond ordinary IFRS. Recent operational changes? FACT: La Colorada Skarn revised PEA (Mar-2026) + $265M decline approval (May-2026); Timmins shaft extension; Jacobina optimization; gold output peaked FY24 and is declining.

APPENDIX B — Source Appendix

Pan American Silver Corp. (NASDAQ: PAAS) — 2026-06-27

Primary sources first. All non-obvious facts cited inline in the memo with source and date. Internal/third-party aggregated data reconciled to primary filings.

Primary — SEC / Regulatory Filings (CIK 0000771992)

1. PAAS 40-F, FY2025 (filed 2026-02-18) — annual report; segment data, reserves/resources, AISC, balance sheet, cash flow, purchase-price allocations, risk factors. https://www.sec.gov/Archives/edgar/data/771992/000077199226000019/paas-20251231.htm
2. PAAS Q4/FY2025 6-K / news release (2026-02-18) — record FY25 results, production, realized prices, AISC, FY26 guidance. https://www.sec.gov/Archives/edgar/data/771992/000077199226000020/paasq42025form6k.htm
3. PAAS Q1-2026 6-K (2026-05-06) — Q1 results, Juanicipio contribution, FCF. https://www.sec.gov/Archives/edgar/data/771992/000077199226000042/paasq12026form6k.htm
4. La Colorada Skarn revised PEA 6-K (2026-03-25) — $1.9B capex, 37-yr life, ~19.1 Moz/yr peak. https://www.sec.gov/Archives/edgar/data/771992/000077199226000031/form6k2026-03x24lacolorada.htm
5. La Colorada decline approval 6-K (2026-05-07) — $265M first-tranche access decline. https://www.sec.gov/Archives/edgar/data/771992/000077199226000047/form6ktr2026-lacolorada.htm
6. Investor Day 6-K (2026-05-13) — 2026 Shareholder Return Framework (up to $1B; 35–40% of attributable FCF). https://www.sec.gov/Archives/edgar/data/771992/000077199226000049/form6k2026-05x13investorday.htm
7. NCIB / shareholder-return 6-Ks (2026-03-04 rexncib; 2026-05-06 rexshareho) — buyback program detail.
8. 2025 reserves & resources 6-K (2025-09-11) — reserve update / price deck.
9. 2025 production & 2026 guidance 6-K (2026-01-21).
10. AGSM / management information circular (2026-03-23 / 2026-05-01) — executive compensation, PSU/TSR metrics, director/officer holdings.
11. Schedule 13G/13G-A filings (2026-01-06, 2026-05-15, 2025-10-17) — institutional ownership.
12. PAAS 40-F corpus FY2021–FY2024 (mirrored locally) for multi-year trend.

Primary — Company IR / Releases

13. PAAS audited FY2024 results (record revenue/OCF/FCF). https://panamericansilver.com/news/pan-american-silver-reports-audited-financial-results-for-2024-…
14. PAAS preliminary 2023 production & 2024 guidance. https://panamericansilver.com/news/pan-american-silver-announces-preliminary-2023-production-results-and-guidance-for-2024/
15. PAAS earnings-call transcripts, Q4-2025 and Q1-2026 (ROIC.ai transcript tools).
16. Yamana Gold acquisition disclosures (closed 2023-03-31; with Agnico Eagle); MAG Silver acquisition (announced May-2025, closed Sep-2025); La Arena sale to Zijin (completed 2024-12-03).

Industry / Market

17. The Silver Institute — World Silver Survey 2026 (Apr-2026) — sixth consecutive deficit ~46 Moz; supply/demand by category; solar PV demand −19%; physical investment +20%.
18. Northern Miner — “Xinka say no to Pan American’s Escobal” (Nov-2025). https://www.northernminer.com/news/xinka-say-no-to-pan-americans-escobal/1003884628/
19. Mining.com — “Indigenous group says no to Pan American’s Escobal mine in Guatemala.” https://www.mining.com/indigenous-group-says-no-to-pan-americans-escobal-mine-in-guatemala/

Quantitative Data (third-party aggregated; reconciled to filings)

20. ROIC.ai — income statement, balance sheet, cash flow, profitability/credit/liquidity ratios, per-share data, enterprise value, valuation multiples (10-yr), company profile, earnings-call transcripts. Accessed 2026-06-27.
21. AZI (azitrading.com) — 5-year daily price CSV (adjusted/unadjusted OHLCV, EMAs, beta/alpha); valuation_index own-history percentile ranks (P/E 4.4th, P/B 89.5th, P/S 87.5th, composite 60.4th); news feed. Accessed 2026-06-27.
22. FactorsToday (factorstoday.com) — stock loadings (GoldPrice beta 2.46–2.59, R² 0.81; Industry Gold Miners 1.86; near-zero style loadings), leaderboard (y1/y3/m3/m6 risk-adjusted returns), stock-info (beta 1.13, rs metrics), related stocks (SILJ/SLVP/SIL/GDXJ/AYA). Accessed 2026-06-27.

Notes / Limitations

• PAAS is a Canadian MJDS foreign private issuer: insiders file to Canadian SEDI, not SEC Form 4 — EDGAR carries no Form 4 corpus, so the discretionary open-market insider-buy/sell signal is unavailable without SEDI. Insider activity is an open question.
• Revenue-by-metal split (~60–65% gold) is an INTERPRETATION from production × realized price; reconcile to the 40-F segment note.
• All scenario EBITDA figures in are ASSUMPTIONS for embedded-expectations analysis, not forecasts. No price target is expressed anywhere outside Claude’s Take.