Nomi Prins — research hub

Stock & name index

▲ Positive

► Neutral / referenced

▼ Negative

Overall thesis

• A rate headline is not a driver — back-test it against 2022–23 before acting. Her standing answer to "higher rates are bad for gold" (2026-AUG-30): the Fed already ran the experiment — eleven hikes, March 2022 → July 2023, near-zero to 5.25–5.50%, "the fastest spate of tightening since the early 1980s" — and gold "barely budged, trading between about $1,650 and $2,050 the whole time," then, with rates still near the peak of that band, "more than doubled" to a ~$4,360 average and a $5,595 record in January. "If 525 basis points of hikes could not push gold down, a single quarter point, less than a tenth of what gold already absorbed, won't either." What actually sets the price "never comes up at an FOMC meeting": federal debt past $40 trillion, a deficit near $2 trillion, interest on the debt about $1 trillion a year — "more than the entire defense budget and the fastest-growing line in the federal budget" — and the inversion that matters, "higher rates only make that debt service cost worse, because every extra point the Treasury pays lands on the trillions of debt it has to keep refinancing," so the hawkish move is itself gold-supportive. Add the two structural bids: central banks buying "around 1,000 tonnes of gold a year for four straight years… in an effort to de-dollarize and, more broadly, de-fiatize" (the PBoC added for a twenty-first straight month in July, its biggest purchase since 2023), and silver's "supply deficit for five years, of more than 100 million ounces a year, as solar, electronics, and defense consume more than existing mines can produce." Hence the discipline: gold to ~$4,500 and silver to $67 after Jackson Hole was only a repricing of odds — December-hike bets above 70% on "a dull, late-August Friday afternoon, with little trading volume" — "that is all that happened, not any actual movement. Traders sold paper gold and silver in anticipation of a rate hike the Fed has not even made." A hike "could catalyze a down day or week, but not much more," because "a central bank buying gold does not care whether the Fed Funds rate is 3.75 or 4%, and neither does a household in Shanghai or Mumbai buying physical silver."
• The selloffs are paper, not physical. Spring 2026: silver −47% from its $121 January record, gold −24% from $5,595 — while copper held near its record and "nothing in the ground has materially changed." Funds raising cash dump the most-liquid metal positions; algos pile on. Her read: a window for disciplined accumulation, not capitulation (silver/miners cheapest & most volatile, copper the least — "a strategic place to start"). Restated into mid-year (2026-JUN-29): gold ~$4,080 (−27% from $5,595) and silver ~$60 (from $121) sold off on the paper leg (~50t / ~$2.7B of gold-ETF redemptions, trimmed futures) while the physical leg strengthened — CBs bought ~244t in Q1 and a record 43% plan to add — and she frames quarter-end (June 30) + July seasonality + the COMEX July silver delivery month as a snap-back setup. Mid-year update (2026-JUL-06): reading the World Gold Council's H2-2026 outlook, she reaffirms her January $6,000 gold forecast — CBs expected to buy 750–850t in 2026 (a record 95% see reserves rising), Asia-led investment demand holding, and the June GLD paper-selling (~12.6M shares on the Jun-24 selloff) fading toward ~5M by month-end as the July reset plays out; a weak 57k June payrolls cut Fed-hike odds (<30%) with gold firming near $4,190, and OMFIF's survey has 61% of 90 CBs/funds seeing $5,000–$6,000 within a year. The paper-vs-physical break then landed (2026-JUL-15): June CPI fell 0.4% m/m (largest drop since Apr-2020, 3.5% y/y) 90 minutes before Kevin Warsh's first House testimony as Fed chair, and gold reversed intraday +$60 to ~$4,087 (FedWatch a Fed hold at 86%) — "this is that break," with the $6,000 path reaffirmed. Extended to silver (2026-JUL-29): the $121 (Jan) → ~$58 pullback came "largely on the back of CME margin hikes that forced leveraged longs to sell" — mechanics, not fundamentals ("didn't change mine supply… didn't change industrial demand… didn't fill the deficit"), with demand inelastic (no substitute for silver's conductivity — "closer to insulin than coffee") and ~70% of supply a byproduct of copper/lead/zinc mining that can't answer the price. The producer's-eye confirmation followed (2026-AUG-01): First Majestic CEO Keith Neumeier, interviewed live at the Rule Symposium, called the drawdown "just a normal correction, 50% correction" — his 2006 analogy inside the 2002–2012 bull, plus June/July seasonal lows on the 30-year chart — and drew the same paper/physical line from the other side of it: 2010–11 was "very much a paper move" that went parabolic, while the last six months "really been a physical market… we saw it show up at the banks. You see the margins start to increase," including a December 2025 margin call First Majestic paid from cash (>$1B) with ~500,000 oz in its own FirstMint vault. Demand is unchanged with the price ("demand at $120 is exactly the same as it is today… physical demand"), silver is "now deemed a critical metal" for nuclear, AI, robotics and all electronics, and the company is buying back stock into the weakness while brokers report institutions returning.
• Supply can't respond in time. New copper or silver mines take 7–12 years to permit and build, so any near-term shortfall is locked in regardless of price; silver is in its 6th straight annual deficit (95M oz short in 2025, 820M since 2021) and platinum just posted a record 1.082M-oz deficit.
• The base is shrinking. First Quantum's Cobre Panama has been shut since late 2023 (~350k tons / ~2% of global output), Anglo American has cut guidance on permitting/water, and Chilean copper grades are down 25–30% since the early 2000s — supply is being lost, not added.
• M&A is the tell. Miners are buying copper they can't build, at premiums — Eldorado/Foran (~C$3.8B), BHP+Lundin/Filo (~C$4.1B), Hudbay/Arizona Sonoran (~30% premium). The strongest producers stay profitable through the rout (PAAS at a $6.63/oz silver AISC, AG at $29.76/AgEq oz — wide margins even at $70 silver).
• The Fed is the swing factor — and it's "QE by another name." QT ended Dec 1, 2025; 11 days later the Fed restarted Treasury-bill buying ("reserve management purchases") — "I call it QE" — pushing holdings back to ~$4.48T (+$200B+). But the buying was front-loaded and is now tapering ($40B → $25B → ~$10B/mo) right as Warsh takes the chair — his "first real balance-sheet test." Her read (2026-JUN-18): the spring hard-asset selloff was liquidity / redemptions / algos / paper-market stress, not a physical break, and metals are already bouncing — because the Fed "cannot print commodities, fix supply chains, or dictate geopolitical outcomes."
• Supply-chain warfare is the policy accelerant. The fight for critical minerals is escalating along the whole value chain (mine → processing → magnets): China blacklisted (2026-JUN-22) the very U.S. rare-earth builders Washington funds (MP, USAR) — every escalation raises the value of ex-China supply, with Nov 10 (the truce expiry) the date to watch. In tungsten (2026-JUN-24), Japan's supply collapsed (Samsung/SK Hynix/TSMC cut off), APT trades ~9× year-ago, and DC is bankrolling an ex-China chain (Mactung + two Kazakh projects) into a Jan-1-2027 Pentagon China-sourcing ban — the processing bottleneck, not mining, is the binding constraint. The June Pulse Premium issue (2026-JUN-25) then names the play: Almonty (ALM), the only major Western producer that mines and processes outside China, with a hard-floor U.S.-defense offtake (buy up to $22). The on-the-ground silver thesis holds too: a Zgounder field visit (2026-JUN-05) reaffirmed Aya (AYA) as a pure-play, low-cost, newly-NASDAQ holding. The resource-nationalism thread extends to uranium (2026-JUL-13): India's three-nation blitz — Australia uranium exports (for a 100-GW-by-2047 nuclear goal), Indonesia nickel + rare-earth magnets, a $10B+ Japan metals/AI pact — locks open-market tonnage into captive deals, re-affirming Uranium Energy Corp (UEC), a permitted Texas producer, as the domestic-Western play squeezed by such pacts. The playbook then reached silver (2026-JUL-29): Beijing reclassified it under dual-use export controls with only 44 approved exporters for 2026–27 — "the same playbook China ran with rare earths. And with tungsten. And with antimony" — and it binds even on Mexican/Peruvian/Australian ore, since much of it still transits Chinese smelters. Simultaneously China closed the retail demand channel: ICBC and the big banks ended retail leveraged metals trading on the physically-settled Shanghai Gold Exchange (final cutoff Jul 24, margin pushed to 140%, a teardown running since 2020), forcing Chinese buyers into bars, coins and physically-backed ETFs. Washington's counter-move then landed (2026-AUG-03): two July directives turn the campaign into enforceable procurement — EO 14415 (Jul 20) orders the Pentagon to trace every critical material in every weapon "back to the mine it came from," gives it 180 days to require a full bill of materials from every prime and subcontractor at every tier, and from Jan 1, 2027 closes the nonavailability waiver loophole contractors had used to keep buying Chinese magnets, tungsten, tantalum and molybdenum ("having no U.S. supplier is no longer an excuse") — so "every producer of those metals outside the restricted countries becomes a source the Pentagon's contractors now have to qualify"; a Jul 30 presidential determination hands Commerce Defense Production Act power to block exports of critical minerals including the metal inside dead magnets and spent lithium-ion batteries, keeping U.S. scrap captive. The dependency it's closing: 100% of U.S. gallium and natural graphite imported (and none produced domestically), 99% of uranium, 91% of antimony, 60% of aluminum, against China refining >80% of world tungsten and ~90% of rare earths while the U.S. is just 15% of rare-earth mine output — with Russian enriched-uranium waivers (fuel for the 19% of the grid that is nuclear) also expiring in 2027. The money behind it: the March-2025 DPA order adding copper/uranium/gold/potash and redirecting the DFC to lend to domestic mines; Project Vault (Feb 2026), a $12B reserve on a record $10B EXIM loan with 54 countries convened; $2.9B more for rare earths/magnets in June plus military bases opened to processors; and >$350B of DOE + Office of Strategic Capital lending authority now open to permitted U.S. copper projects (copper ~$6.43/lb vs the May record $6.71 on AI data-center demand). Aluminum is caught without being named — trace-to-smelter plus the scrap-export block in a country down to four smelters, atop a 50% Section 232 tariff (Jun 2025), a Jul-2026 discount for firms investing in domestic smelting, and a DOE-backed 500,000-ton plant, the first new U.S. smelter in ~45 years. Her stated screen is "where defense and energy infrastructure intersect," with copper and aluminum picks teased for this month's gated Pulse Premium and Founders+ issues. Nine days later the same playbook is documented running on copper itself (2026-AUG-12) — and this time on all three sides at once: the DRC (No. 2 producer, ~3.3Mt/yr) made public on Aug 6 a Jun-29 order banning copper and cobalt concentrate exports outright — tiny in tonnage (Q1: 696,725t cathode vs ~54,000t concentrate) but "systemic," copying the Indonesia nickel model of forcing onshore processing (LME jumped 1.8% to $14,369.50/t on the news); China — which refines "nearly half the world's copper" and buys ~two-thirds of the concentrate — named copper a strategic mineral on May 19 under its revised Mineral Resources Law, letting the state hold reserves off-market for at least five years (with the CNIA lobbying for a dedicated copper reserve), and since May 1 has banned sulphuric-acid exports, the reagent behind "about a fifth of the world's refined copper" (Chile alone buys >1Mt/yr of Chinese acid); and the U.S. 50% semi-finished tariff has drained the world into American warehouses (>200,000t landed in July, fastest in 12 years; COMEX + U.S.-held LME >740,000t plus 110,000t off-warrant) while the Jun 30 ruling on tariffing refined copper — "the form American industries actually run on" — sits overdue and in flux, and the Jul 30 DPA determination lets Commerce block copper and scrap exports. Net effect: LME available inventory has collapsed to ~94,200t, "little more than a day of global consumption," from ~400,000t in April, with Trafigura alone pulling >51,000t out of LME warehouses this year (largest since 2013) to sell into the U.S. premium — against a Morgan Stanley 600kt 2026 refined deficit, the widest in 20+ years, and S&P Global's 42Mt-by-2040 demand vs mine supply peaking near 33Mt in 2030. Her inversion of all of it is the screen: look "to countries where mining capital can get in, and copper can get out more freely" — the basis for a gated Founders+ pick (a developer on "one of the largest undeveloped deposits in a jurisdiction that welcomes both investment and exports") dropping the next day, not captured here. The uranium leg then got its operator's-eye mechanism (2026-AUG-18): Scott Melbye — president of the Uranium Producers of America, EVP of UEC and CEO of Uranium Royalty Corp (UROY), interviewed live at the Rule Symposium — argues spot "stuck at an $85 level" is "a bit of a coiled spring" pulled up by the $95 long-term market: utilities' long-term RFPs are drawing "fewer and fewer offers and less quality offers in terms of price and flexibilities," so "if they don't like what they see in the long-term market, their only option is to come to the spot" — >$100/lb in 2H26. Prins concurs (catch-up to $95–100, a Prinsights $110 target for the year) and flags the anomaly that the uranium equities have underperformed the commodity, "which is rare throughout the commodity space" — Melbye's version: "UEC, URC are all trading well below where they should be given the fundamentals," against nuclear power doubling in the base case (tripling, or quadrupling on Trump's US goal) needing a doubling-to-tripling of conversion and enrichment the market "just isn't prepared" for. The policy stack behind it: the Russian import ban at full force in January 2028 once the waiver loopholes close, a Defense Production Act consortium survey putting US capacity at 6M lb by end-2027 → 35M lb by 2033 ("a bit aspirational"; even 25–30M replaces imports), a lobbied-for Strategic Uranium Reserve carrying FAST-41 preferential permitting, and UEC's planned 10,000-tonne domestic refining/conversion facility to end reliance on "just one 70-year-old facility in Illinois" — the fuel cycle's one missing gap, with the Naval Propulsion Program as the price-insensitive buyer underneath. Corporate: UROY's announced $1.1B acquisition of Sweetwater Royalties from Orion Resource Partners and Ontario Teachers' — the original 1860 Union Pacific land grant, 5.3M acres Cheyenne→Salt Lake plus the minerals beneath (largest public-company landowner in the US, ~90% of world trona/soda ash, five mines at ~$74M EBITDA / $30–50M FCF, split 50/50 with Interior across a 10.6M-acre checkerboard, soda-ash revenue up 2.5× in five years, with uranium, oil & gas, helium, wind, data-centre and battery-storage optionality unpaid-for) — explicitly a bridge, not a pivot: it funds acquisitions until URC's 27 royalties on 24 projects cash-flow $20–50M/yr in the 2030s. The next day (2026-AUG-19) the uranium leg is generalized into a five-chart exposure map with no tickers attached: AI/data-center consumption heading to ~600 TWh by 2035 (gas anchors the near term, nuclear + renewables take the long-term structural additions) with round-the-clock reliability — not raw volume — the binding constraint, so hyperscalers sign long-term, high-premium PPAs direct with nuclear operators and the edge is in off-grid / behind-the-meter baseload rather than utility-multiple operators; over 2021–mid-2026 physical uranium (+166%) led senior miners (+101.91%) and juniors (+65.11%) past the S&P 500 (+83.10%) and broad commodities (+37.16%), the metal leading the equities as a deficit tell → pair physical uranium with top-tier low-cost seniors; the U.S. pipeline splits into near-term restarts/uprates/life-extensions (engineering + utility firms that already hold a regulatory pathway) versus longer-dated SMR pure-plays screened on DOE cost-share plus industrial offtake; public favorability at a near-record 77% (vs 23% opposed) is treated as a quantitative regulatory-risk gauge — nuclear reframed as economic/national-security/grid-reliability policy, so support survives election cycles; and with only 7% of U.S. nuclear fuel domestic (>90% imported) against tightening Russian enriched-uranium restrictions, Western fuel-cycle infrastructure — domestic conversion and enrichment plus allied-nation (US/CA/AU) resource development — is her "highest-conviction" leg. (The post's closing Founders+ copper-developer pick is gated and not captured.)
• Aluminum is the playbook's cleanest expression — and the first one where the pick is captured. The metal she put "at the top of our Darkhorse commodities to watch in 2026" became a national-security problem and, in the last week of August, the flashpoint of the U.S.–Canada tariff fight (2026-AUG-26): the U.S. makes ~680,000 t of primary aluminum to China's 43M, imports 56% from Canada (UAE second at 8%), and when the deal to cut the tariff from 50% to ~25% collapsed on the Friday the 50% rate took effect Saturday, the U.S. threatened auto tariffs by Monday, and Canada announced duties of up to 50% on nearly 700 products from September 8. A two-day Pentagon war-game last summer had already named the weak point — high-purity aluminum, the grade that skins a fighter jet, with no real domestic source and ~90% of imports from the UAE — against a 2027 budget asking to lift missile procurement 188%. Her key inversion: because "the U.S. smelts almost none of its own," the tariff has no domestic industry to protect, so it functions not as a shield but as a price floor — it "holds the U.S. price of the metal well above the global price," and the few surviving smelters "earn more on what [they sell] than before the duty took hold in 2025," with DOE cash (up to $500M for the first new U.S. smelter since 1980, at Inola OK, >500kt/yr including 20kt of high-purity metal ring-fenced for defense) and the DPA reinforcing the wedge — "the same tools deployed across copper, rare earths, and other critical minerals earlier this year" — while rebuilding is "a decade-long project." The next day the August Pulse Premium issue names the operator (2026-AUG-27): Century Aluminum (CENX), buy up to $52 — two of only four U.S. primary smelters, selling into LME ~$3,225/t plus a US Midwest Transaction Premium near $2,000/t (~$5,200 delivered, 60% over the world price; a record ~$2,180 in February, triple its level two years earlier) while owing no import duty, and since "energy, a smelter's largest cost, does not rise with the aluminum price, most of that premium becomes margin, far more than a fabricator or a diversified miner keeps." That sentence is the whole method: value-chain position decides who keeps a policy-created premium — which is why July's Semi-Annual Update closed the fabricator Constellium (CSTM) at a more-than-double return in eight months and rotated upstream to the smelter. The metal backdrop: LME stocks halved this year to ~250,000 t, the lowest since 1990, Gulf outages cut 2Mt annualized (Mercuria: "a black swan"; ≥2Mt 2026 deficit), China pinned at its 45Mt cap (~60% of world output), against grid/AI-transmission and EV demand (31.7Mt vs 19.9Mt in 2020). Century specifics: ~760kt across Sebree, Mt. Holly (full nameplate end-June) and hydro/geothermal Grundartangi (Natur-Al, <4t CO2/t, ~a quarter of the industry average, an edge as EU carbon border adjustments tighten), all three at capacity for the first time in a decade; record Q2 EBITDA $326.9M on sales of $752.1M, net cash by late July, 55% of Jamalco for captive alumina, Hawesville sold to a TeraWulf affiliate for $200M, and a 40% stake in the $4B / 750kt Inola JV with EGA (60%) part-funded by a July-2026 proclamation letting Century import 300kt/yr at 25% from 2027. Risks she names in order: the tariff itself (an Aug-19 report of a 50%→25% Canada cut took the stock down 6% in a session — but "at 25%, the tariff still generates a substantial domestic premium, well above pre-tariff historical levels"), Glencore at 30% of shares and 44% of Q2 sales, execution, and the Oklahoma AG's suit against Inola. The entry discipline is the recurring one: buy the policy premium after a sentiment-driven drawdown — ~$21 → ~$70 → ~$44, −36%, while EBITDA hit records — because "the stock is trading as if that premium disappears entirely. We don't think it will."
• Tier matters for what the archive can capture: Pulse Premium monthly issues are readable and their "Action to Take" picks and buy-up-to prices are archived (ALM Jun-25, CENX Aug-27); Founders+ picks remain fully gated, so for those the archive captures the thesis and named evidence only.

The product — Prinsights

From her own descriptions in the material archived here (2026-JUN-04 interview, 2026-JUN-10 premium article).

How it serves retail investors:

• Policy literacy as an edge. An ex-Goldman MD reads export-control rules, defense statutes and deadline calendars — sources most retail investors never open — and maps them to tickers.
• A free evidence layer. Even without paying, the public articles name the thesis and the supporting M&A/policy facts; the subscription adds the specific pick and a buy-up-to price.
• Substack pricing. Portfolio-service economics at newsletter cost, versus the institutional research this kind of policy work usually lives in.

Posts

One dated page per article — each has its stock table and key points. Newest first. (Premium articles: the public summary is captured; the gated pick is not.)

To process — backlog

Prinsights posts not yet processed — newest first. None queued yet.

For personal study — not investment advice. Source material © Nomi Prins / Prinsights.