# U.S. Antimony Supply Resilience, 2027–2031

**Evidence, assumptions and calculation register**  
**Research cutoff / source retrieval:** 4 November 2026  
**Research foundation:** [Strategic Materials Policy Tracker — Antimony](https://strategic-materials-policy-tracker.vercel.app/materials/antimony)  
**Decision perspective:** Chief Procurement Officer, Chief Strategy Officer and CFO of a modeled U.S. defense manufacturer.

## 1. Executive analytical conclusion

The qualified, balanced sourcing policy remains the leading *illustrative* response: develop an independently traceable alternative import lane targeting 45% of annual volume, maintain a **provisional five-month reserve of buyer-qualified input**, and negotiate a **contingent 30% domestic offtake target** tied to delivered, usable material. The 30% figure is a future sourcing objective; the analysis assumes zero deliverable domestic volume from that target in 2027–2029 and 15% effective volume in 2030. The main annual cases assume the alternative 45% lane is **already qualified and deliverable**; a separate 2027 downside test removes that assumption.

Under the modeled 2030 eight-month severe interruption, five months of initial stock absorbs the full 39.00 t disrupted volume. At 100 t annual demand and the USGS 2025 metal price benchmark, five months of inventory costs **$2,296,482** and the sum of planned sourcing premiums, one-year inventory carry and the event's assumed spot markup is **$812,955**. This comparison uses assumptions, not actual supplier quotes, customer volumes or company accounts.

**Conditionality:** With the 45% alternative channel already qualified, the severe scenario before domestic qualification leaves **4.35 t** unfilled with five months of inventory. **If no alternative channel qualifies in 2027, 16.50 t remains unfilled** as modeled volumes fall back to the incumbent channel. If the alternative-source lane is curtailed 75% rather than 45%, the 2030 gap rises to **5.70 t**. A correlated domestic delay and 75% alternative curtailment raises it to **12.45 t**. The five-month policy is a procurement design to be tested, not an unconditional purchasing instruction. Reserve a sixth-month purchase option where supplier traceability or continuity criticality warrants it.

## 2. Original-source register

| ID | Original publication and URL | Publication / event basis | Observed fact and treatment | Accessed |
|---|---|---|---|---|
| S01 | [USGS, Mineral Commodity Summaries 2026, antimony pp. 42–43](https://pubs.usgs.gov/periodicals/mcs2026/mcs2026.pdf) | February 2026; 2025 estimates | 2025 U.S. apparent consumption 45,000 t contained Sb; import reliance 91%; metal reference price $25/lb. 2021–24 metal + oxide import origins: China 55%. Historical prices and mine production by country. | 2026-11-04 |
| S02 | [USGS MCS 2026 associated data](https://www.usgs.gov/data/mineral-commodity-summaries-2026-data-release) | 2026 data release | Supporting official series; the antimony chapter controls displayed values and units. | 2026-11-04 |
| S03 | [SMPT antimony evidence register](https://strategic-materials-policy-tracker.vercel.app/materials/antimony) | Site register consulted 2026-11-04 | Financing/status taxonomy, associated Stibnite, U.S. Antimony, Alaska and export-control project records. Multiple lifecycle stages reflect the same project; never sum all stage entries. | 2026-11-04 |
| S04 | [EXIM project-finance transaction register](https://www.exim.gov/solutions/project-and-structured-finance/transactions) | Table updated 2026-07-02 | Perpetua project: **$2,906 million**, **Authorized**. Board authorization is separate from definitive loan documents, drawing and commercial output. | 2026-11-04 |
| S05 | [EXIM May 21, 2026 board minutes](https://www.exim.gov/news/minutes/board-meeting-minutes-2026-05-21) | Decision 2026-05-21 | Loan board decision. Model excludes the public project loan from manufacturer funding. | 2026-11-04 |
| S06 | [Perpetua Form 10-Q, quarter ended June 30, 2026](https://www.sec.gov/Archives/edgar/data/1526243/000110465926097134/ppta-20260630x10q.htm) | Reported Q2 2026; published August 2026 | EXIM closing contingent on definitive documents and conditions; planned full construction depends on finance; $59.2m DPA TIA fully disbursed and expired; prototype OTIA current estimate ~$27.1m, with partial reimbursements. | 2026-11-04 |
| S07 | [Idaho National Laboratory, antimony processing pilot](https://inl.gov/news-release/inl-hosts-ribbon-cutting-for-pilot-plant-supporting-domestic-antimony-production/) | 2026-07-30 | Commissioned modular pilot; **approximately one ton of ore/material per day** planned processing. This is pilot *feed*, not antimony metal, trisulfide or qualified output. | 2026-11-04 |
| S08 | [U.S. DOE, pilot process categorical exclusion](https://www.energy.gov/sites/default/files/2025-09/CX-034364.pdf) | 2025-09 | Government documentation of process-development project. | 2026-11-04 |
| S09 | [MOFCOM/GACC Announcement No. 33 (2024)](https://www.mofcom.gov.cn/zfxxgk/fdzdgknr/ztfl/dwmygl/art/2024/art_c37e7a6e7438428c951e2b1e488dc47e.html) | Issued 2024-08-15; effective 2024-09-15 | Export licensing of specified antimony goods, materials, compounds and technologies; applicable control schedule and licensing documentation depend on the exact item. | 2026-11-04 |
| S10 | [MOFCOM Announcement No. 46 (2024)](https://www.mofcom.gov.cn/zfxxgk/gkml/art/2024/art_99f5d51bbe6a412ca309d5818c4dee52.html) | 2024-12-03 | Clause 1 bars export of dual-use items to U.S. military users / military uses. Clause 2 establishes separate U.S.-specific restrictions for named commodities. | 2026-11-04 |
| S11 | [MOFCOM Announcement No. 72 (2025)](https://www.mofcom.gov.cn/zfxxgk/gkml/art/2025/art_43489ca6f2094b70954c25249702b4cc.html) | 2025-11-09; suspension to 2026-11-27 | Suspends **Clause 2** of No. 46 only through **November 27, 2026**. Clause 1 and the general licensing framework remain separate. | 2026-11-04 |
| S12 | [Perpetua, June 1, 2026 construction update](https://perpetuaresources.com/perpetua-resources-advances-construction-of-the-stibnite-gold-project/) | 2026-06-01 | Early works, infrastructure, and producer's **target** of operations in 2029. This is a management schedule and does not establish purchased defense-grade volume. | 2026-11-04 |
| S13 | [USGS antimony commodity portal](https://www.usgs.gov/centers/national-minerals-information-center/antimony-statistics-and-information) | Federal commodity portal | Supporting background / series. | 2026-11-04 |
| S14 | [United States Antimony, Form 10-Q, quarter ended June 30, 2026](https://www.sec.gov/Archives/edgar/data/101538/000110465926094035/uamy-20260630x10q.htm) | Filed 2026-08-11 | DPA award headline **$27.0m**; **$16.2m obligated**, **$10.8m later-phase authority pending**, and **$12.8m cash received by June 30**; 2026 Thompson Falls processing expansion milestones. | 2026-11-04 |
| S15 | [Department of War, FY2026 Mandatory Funding Allocation Plan](https://comptroller.war.gov/Portals/45/Documents/news/FY2026_Mandatory_Funding_Allocation_Plan.pdf) | Dated 2026-05-04 | **$75m antimony midstream processing** in allocation plan. Budget allocation, not a recipient-level executed contract or completed processing plant. | 2026-11-04 |
| S16 | [Perpetua Form 8-K, October 26, 2026](https://www.sec.gov/Archives/edgar/data/1526243/000110465926118875/tm2628317d1_8k.htm) | 2026-10-26 | Separate **tungsten** prototype initiative; excluded from antimony financing and modeled purchases. | 2026-11-04 |
| S17 | [DLA Strategic Materials](https://www.dla.mil/Strategic-Materials/) | Agency program page | DLA operates the National Defense Stockpile. A hypothetical private manufacturer has no assumed access to Government inventory. | 2026-11-04 |

### Historical market dataset and denominators

| Metric | Year / period | Value | Units and scope |
|---|---|---:|---|
| U.S. apparent consumption | 2025 estimate | 45,000 | metric tons **antimony content**; USGS formula includes primary and secondary supply and net metal/oxide imports |
| U.S. net import reliance | 2025 estimate | 91% | share of apparent consumption; *distinct from* country-of-import breakdown |
| China's U.S. metal + oxide import share | 2021–2024 combined | 55% | U.S. import-origin distribution for metal and oxide combined; not supplier-tier ownership |
| Antimony metal average benchmark | 2024 | $10.24 | USD per pound, minimum 99.65% metal, CIF |
| Antimony metal average benchmark | 2025 estimate | $25.00 | USD per pound, minimum 99.65% metal, CIF |
| World mine production | 2025 estimate | 110,000 | metric tons **contained Sb**, rounded |
| China / Russia / Tajikistan mine output | 2025 estimate | 40,000 / 32,000 / 22,000 | metric tons contained Sb; combined approximately 85.5% of rounded world figure |

**Technical unit discipline.** The model expresses all demand, physical stocks and disrupted deliveries in **metric tons of contained Sb-equivalent**. Actual products include concentrate, metal, antimony trioxide and military-specified antimony trisulfide, with different Sb content, prices, quality requirements, production steps and storage properties. The USGS metal benchmark provides an **analytical cost proxy**; a purchasing model would replace it with prices, mass-balance conversions and qualified material specifications by SKU.

## 3. Assumptions entered in editable workbook cells

| Variable | Default | Location / evidence class |
|---|---:|---|
| Buyer annual demand | 100 t Sb-equivalent / year | `Inputs!B6`; illustrative |
| lb per metric ton | 2,204.62262185 | `Inputs!B7`; unit conversion |
| Metal price benchmark | $25/lb; $55,115.57/t | `History!C10` → `Inputs!B8`; observed historical benchmark |
| Alternative import premium | 15% | `Inputs!B9`; illustrative |
| Qualified domestic supply premium | 20% | `Inputs!B10`; illustrative |
| Annual inventory carrying rate | 12% | `Inputs!B11`; illustrative |
| Chosen comparison year | 2030 | `Inputs!B12`; editable |
| Maximum inventory cash | $3,000,000 | `Inputs!B13`; illustrative buyer constraint |
| Maximum annual unfilled volume | 5 t Sb | `Inputs!B14`; illustrative buyer service constraint |
| Balanced policy target | 25% incumbent / 45% alternative / 30% future domestic | `Options!B9:D9`; scenario shares |
| Chosen balanced stock | **5 months** | `Options!E9`; reviewed proposal, editable |
| Severe interruption | 8 months | `Scenarios!B8`; illustrative |
| Severe curtailed share, incumbent / alt / domestic | 90% / 45% / 15% | `Scenarios!C8:E8`; illustrative |
| Emergency sourcing fraction | 10% of uncovered deficit | `Scenarios!F8`; illustrative |
| Emergency spot-price markup | 150% of reference price | `Scenarios!G8`; illustrative |
| Domestic supply availability | 0 / 0 / 0 / 50% / 100% of domestic target in 2027–31 | `Ramp!B6:B10`; illustrative; subject to buyer acceptance |

Alternative strategies remain editable: **domestic-led** (30% incumbent / 25% alternative / 45% conditional domestic, 2 months stock); **diversification-led** (45/55/0, 3 months); **inventory-led** (75/25/0, 9 months); **balanced** (25/45/30, 5 months). Before domestic qualification, its share remains assigned to the incumbent import lane. This prevents assumed offtake from appearing as current deliveries.

## 4. Financial and physical calculations

For each year / strategy / scenario:

1. **Effective domestic share** = target domestic share × qualified ramp fraction. The remaining target is reallocated to the incumbent lane. Alternative import share remains fixed as a procurement design assumption.
2. **Gross disrupted quantity (t)** = buyer annual demand × scenario months / 12 × sum of each lane's effective share × its scenario curtailed fraction.
3. **Beginning reserve (t)** = annual demand × stock months / 12. **Stock drawn (t)** = minimum of gross disruption and reserve.
4. **Emergency fill (t)** = uncovered disruption after stock × modeled emergency-fill fraction. **Residual unfilled (t)** = gross disruption − stock drawn − emergency fill.
5. **Planned annual sourcing premium** = demand × USGS metal price proxy per t × (alternative share × alternative premium + effective domestic share × domestic premium).
6. **Annual stock carry** = initial reserve × price proxy × annual carry rate. **Spot markup** = emergency fill × price proxy × emergency-price markup. **Modeled annual incremental expense** = planned sourcing premium + stock carry + spot markup.
7. **Initial cash tied up in inventory** = starting reserve × metal price proxy; shown separately from incremental expense. It is working capital and an inventory asset, distinct from the annual expense projection.

The model **does not assert full transaction-level cash outlays**: actual source quantities bought during an interrupted year, base-price principal on replacement procurement, stock replenishment, carrying costs for partial-year holdings, duty and freight, storage capacity, throughput and commercial losses need contract-specific data. Every annual event independently begins with the stated inventory. The model is an **episodic stress test**, not a five-year stock-flow forecast. No frequency-weighted expectation or NPV is calculated.

### Decision-year comparison: 2030 severe case

| Strategy | Unfilled, t Sb | Inventory capital | Modeled annual increment | Continuity ≤5 t | Stock ≤$3m |
|---|---:|---:|---:|---|---|
| Domestic-led | 22.125 | $918,593 | $768,173 | Fails | Passes |
| Diversification-led | 16.650 | $1,377,889 | $772,996 | Fails | Passes |
| Inventory-led | 0 | $4,133,667 | $702,723 | Passes | Fails |
| **Balanced, 5 months** | **0** | **$2,296,482** | **$812,955** | **Passes** | **Passes** |

Compared with a six-month balanced buffer, the revised five-month policy frees approximately **$459,296** of working capital and reduces annual carrying expense by **$55,116** under the benchmark. Compared with the inventory-led policy, it frees **$1,837,186** in stock capital while costing approximately **$110,231 more** annually in the chosen 2030 scenario.

### Derived policy requirement and reversal conditions

| Analytical test | Calculated outcome | Decision meaning |
|---|---:|---|
| 2027–29 severe gross loss | 46.50 t | With zero domestic access, balance relies on incumbent and alternative imports |
| 2027 with no alternative supplier qualified | **16.50 t unfilled** | All sourcing volume reverts to the incumbent import lane; the five-month policy fails the continuity threshold |
| 2027–29 minimum inventory for ≤5 t residual | **4.913 months** | Five months passes by a small margin; service-critical inputs may require tighter tolerance |
| 2027–29 residual with five months | **4.35 t** | Material continuity at risk despite an illustrative threshold pass |
| 2030 severe gross loss | **39.00 t** | Assumes 50% of domestic target qualifies, i.e. 15% actual share |
| 2030 minimum inventory for ≤5 t residual | **4.013 months** | Algebraic minimum for buyer's modeled tolerance |
| 2030 minimum inventory for zero gap | **4.680 months** | Five months fully covers the selected severe event |
| 2030 five-month cash headroom | **$703,518** | Cash ceiling remains above modeled inventory purchase |
| USGS-metal proxy price reaching $3m cash ceiling | **$32.66/lb** | Approx. **30.6%** above 2025 benchmark; actual grade prices may differ |
| 2030 alternative curtailed 75%, domestic ramp available | **5.70 t unfilled** | Breaches 5 t tolerance; diversify upstream refining or increase stock |
| 2030 75% alternative curtailment + domestic delay | **12.45 t unfilled** | Correlated sourcing weakness is the most material tested near-term downside |
| 2030 severe interruption lengthened to 12 months | **15.15 t unfilled** | Long-duration disruption requires replenishment and additional sources |

**Control logic:** The main `Cases` grid assumes the modeled 45% alternative sourcing lane is qualified from 2027. `Policy Tests` row 20 tests the absence of any qualified alternative, with all demand assigned to the incumbent lane. The `Policy Tests` sheet derives minimum months for the 5 t gap hurdle and zero modeled gap; the six adverse cases are formula-driven. `Sensitivity` links its central axes to editable scenario values and stock policy. Dashboard metrics link to the 60 year×strategy×scenario calculations. `Checks` includes 21 pass/fail formula validations, including unit conversion and source-policy reconciliation.

## 5. Project-financing classifications and overlap controls

- **Stibnite EXIM: $2.906bn.** Board authorization / EXIM table status **Authorized**. Financing documentation, closing and disbursement remain distinct decisions. Project financing is not buyer funding or production inventory.
- **Perpetua DPA TIA: $59.2m.** Company reported full disbursement and expiry in 2025; historical realized government funding, separate from subsequent processing agreements.
- **Perpetua prototype OTIA: ~$27.1m current estimate.** Separate procurement/research instrument for trisulfide pilot and development work, with partial reimbursements. Pilot feed capacity ≠ qualified commercial output.
- **U.S. Antimony DPA grant: $27m total award.** SEC Q2 2026: $16.2m obligated initially, $10.8m later authorization required, $12.8m cash received. SMPT's $7m Alaska and $20m Thompson Falls subdivisions sit within the $27m total; these scope allocations must not be added on top. Phase obligations / cash receipts form another reporting dimension.
- **Alaska Range Resources / Estelle: $43.4m.** SMPT classifies the contract as a separate recipient/project; the existence of funding does not demonstrate qualified buyer deliveries.
- **FY2026 DoW $75m antimony midstream processing allocation.** Policy budget intent, separate from executed supplier commitments.
- **October 2026 Perpetua tungsten OTIA.** Separate commodity, excluded from antimony funding totals.
- **DLA National Defense Stockpile.** Federal government inventory, distinct from a manufacturer's working capital or guaranteed private-market material.

## 6. Material open questions and decision gates

**Procurement + Engineering:** bill of materials by SKU, contained-Sb conversion factors, qualification protocols, test acceptance, storage requirements and shelf life; refinery and smelter location; chain-of-custody, permitted end use and export-licensing legal advice; achievable substitute or redesign options; lead-time / pipeline inventory.

**Finance:** supplier quotes by form, grade and currency; freight, duties, financing rates, carrying cost and insured storage; actual cash cap; production interruption cost per day; replenishment and reverse logistics; customer backlog and penalty exposure.

**Strategy + Legal:** conditional Stibnite offtake options tied to final financing, project completion, processing agreements, qualified grade, guaranteed deliveries and enforceable remedies. Any domestic-production assumption is gated by documentary progress and accepted shipments.

**Decision reversal:** with zero qualified alternative access in 2027 the five-month case leaves 16.50 t uncovered; qualify the channel before relying on the policy. Raise buffer toward six months, accelerate a distinct refining route or renegotiate delivery options if the alternative source shares origin/refinery exposure, the input's 5 t tolerance is too high, the disruption lasts beyond the selected eight-month event, or the metal-equivalent procurement price breaches the working-capital ceiling. A live decision requires grade-specific contracts and inventory replenishment economics.

---

**Portfolio methodology note:** Manufacturer demand, supplier lanes, contract terms, service thresholds, price premiums, shock rates and ramp timing are illustrative. The sourcing comparison is research-based and developed independently of any commissioned client work. Source links are primary government, laboratory or SEC/company records wherever available.