Senior Gold Miners
Top Companies Analysis and Review

Every financial outlet publishes annual production rankings. Newmont first, Agnico Eagle second, Barrick third. These rankings carry almost no investment utility. What separates miners over the long run is unit economics, where their mines sit on the life-cycle curve, and whether the government hosting their assets decides one morning to rewrite the rules.

Agnico Eagle Mines overtook both Newmont and Barrick in 2025 to become the number one component in GDX, despite having a smaller market cap than Newmont at the time. The ETF managers gave a smaller company a larger weighting. The weighting logic shifted from market-cap-driven to operations-quality-driven, and that shift tells you more about where senior gold mining is heading than any production table does.

Zijin Mining surged 35% in output year-on-year and jumped to fourth globally. Northern Star locked in future 3-million-ounce annual potential through the A$5 billion De Grey acquisition. AngloGold Ashanti's market cap crossed $55 billion by year-end. The North American duopoly is finished.

On February 3, 2026, Newmont issued a formal notice of default to Barrick over the Nevada Gold Mines joint venture, alleging systematic mismanagement. This triggered the formal default procedure under the 2019 JV agreement.

NGM was created in 2019 after Barrick dropped a hostile bid for Newmont and the two sides instead combined their adjacent Nevada assets. Barrick holds 61.5% and operates day-to-day. Newmont holds 38.5%. The complex includes Carlin, Cortez, 10 underground mines, 12 open pits, two autoclave facilities, multiple processing plants. RBC's Josh Wolfson has estimated NGM accounts for roughly 60% of Barrick's entire market value.

Newmont's allegation: Barrick diverted specialized personnel, heavy equipment, and technical resources from the joint venture to accelerate Fourmile, a project Barrick owns 100% and deliberately excluded from the JV in 2019. Fourmile sits right next to the JV-owned Cortez complex. Newmont called it a “managed decline” of the joint venture assets.

NGM's Q4 2025 production fell 23% year-on-year. During the same quarter, Fourmile next door was absorbing dramatically increased investment. If the production decline at the joint venture was even partly caused by resources being redirected to Fourmile, the implications are enormous. At $5,200 gold, every lost ounce from the JV represents thousands of dollars in opportunity cost, and Newmont owns 38.5% of that cost.

Newmont has also asserted ROFR and transfer restriction rights over Barrick's planned NewCo IPO, which bundles the NGM interest, Pueblo Viejo, and Fourmile into a new listed vehicle. If Newmont withholds consent, NewCo could be frozen entirely.

Mark Hill for Barrick: disagrees with the claims, can't say more due to JV confidentiality provisions. Natascha Viljoen for Newmont on the earnings call: same position, same provisions. Both companies clearly treating every public word as potential litigation material.

The JV operating structure has been standard in Nevada for decades because infrastructure costs are enormous and sharing makes economic sense. At $1,800 gold, the operator diverting a few engineers to a side project costs the non-operating partner a few million dollars. Nobody sues over that. At $5,000 gold, the same diversion costs hundreds of millions. The conflict of interest did not change. Its price tag did. Every existing JV in the gold mining sector should be re-examined with this in mind.

Barrick's Fourmile dilemma has no clean exit. Bringing Fourmile into the JV means handing over 61.5% of a 17 g/t future Tier One asset instead of keeping 100%. At $5,000 gold that gap is billions. Keeping it outside means permanently facing Newmont's legal allegations. The 2019 exclusion of Fourmile was a perfectly reasonable decision at the time. Early-stage find, nice grades, insufficient certainty. Six years and two resource doublings later, that same decision is the fuse on the biggest corporate governance conflict in gold mining in 2026.

Mark Bristow left the CEO role in September 2025 after nearly seven years. He built the modern Barrick through the Randgold merger. His successor Mark Hill's primary task appears to be executing the North American asset spinoff.

The Mali saga with Loulo-Gounkoto deserves more space than it usually gets. The timeline: October 2024, new mining code demanding higher state participation and royalty rates. November, disagreements over export authorizations and tax compliance. December, Malian authorities blocked gold exports. January 2025, full operational shutdown. March, government-appointed provisional administrator took over. July, limited operations resumed at roughly 25% of normal capacity. People on Barrick's side said privately that even after full resolution, restoring normal output would take at least four months because equipment, technical systems, and spare parts supply chains all need reconnection time.

Loulo-Gounkoto produced about 578,000 ounces in 2024, roughly a third of Mali's entire industrial gold output. The shutdown dropped Mali's 2025 industrial gold production by 22.9%. Total national output came in at 48.2 tonnes against a government target of 54.7. A senior mines ministry official privately attributed the shortfall to “Barrick problems.”

The detail from the provisional administration that matters most: the government administrator could not operate Barrick's proprietary technical systems or the mine-specific operational methodologies. That is why capacity recovered to only 25%. The administrator had the mining rights, the physical infrastructure, the labor force. What he did not have was the decades of geological knowledge about how to handle specific ore types in specific geological conditions at that specific mine. Equipment maintenance protocols. Mineral processing parameters. The institutional memory that lives inside a technical team's heads and in their internal documentation systems. A government can confiscate all the physical assets of a mine overnight. The knowledge to run those assets takes years or decades to accumulate. Mali demonstrated this gap very clearly, and it has implications for every jurisdiction where resource nationalism is on the rise, because it changes the calculus of what a government actually captures when it seizes a mine. The answer from Mali: about a quarter of the production capacity, at least in the short run.

Barrick dropped “Gold” from its name. Reko Diq copper-gold is targeting end-of-2028 production. Lumwana Super Pit expansion advancing. Both years away from contributing.

The overall picture for Barrick: production declining for six straight years, AISC trending toward $1,950, reserves at 85 million ounces (down from prior year, grade 0.98 g/t), the CEO replaced, the largest single asset under a formal default notice from the JV partner, the planned North American spinoff potentially blocked by that same partner's ROFR claims. And the capital allocation response: early 2026, 50% of free cash flow to dividends, $1.5 billion in buybacks completed. The stock returned 195% in late 2025 because $5,000 gold lifts all boats, but gold does not stay at $5,000 forever, and when it normalizes, the operational trajectory underneath is what is left.

Full-year production approximately 5.7 million ounces. Strongest cash flow in company history. AISC around $1,358 per ounce. $6 billion buyback program. Significant dividend increase. Production fell 14% year-on-year. 2026 guidance: further decline. Peñasquito's gold output share dropping. Yanacocha's Quecher main pit winding down. Ahafo North in Ghana reached commercial production in 2025 but does not have the tonnage to offset the broader slide.

Newmont owns more mines than anyone else and the portfolio as a whole is uneven. A lot of assets, not enough great ones. At $3,000 gold the elevated AISC becomes a free cash flow cliff.

What matters most about Newmont in 2026 is not its standalone operations story but the Nevada dispute, which has already been covered at length above. The default notice, the ROFR assertion over NewCo, the demand for operational overhaul at NGM before any spinoff proceeds. If Newmont prevails in forcing Barrick to either fold Fourmile into the JV or make major operational concessions, the value implications for Newmont's 38.5% stake are substantial. If the dispute escalates to litigation or arbitration, both companies' management teams will be consumed by it for years.

AEM's AISC advantage over its peers has been noted already: $1,275 midpoint guidance versus $1,566 for Newmont and $1,637 for Barrick. The gap against Barrick approaches $400 per ounce. At multi-million-ounce scale that runs to hundreds of millions in annual free cash flow differential. The market understands the cost advantage to some degree and has priced it partially into AEM's premium. What has received far less attention, and what ultimately matters more for long-term positioning, is the reserve replacement picture.

Canadian Malartic is Canada's largest open-pit gold mine. Open-pit mining ends in 2029. The underground Odyssey mine is already taking over. Odyssey holds 6.03 million ounces of proven and probable reserves at 3.14 g/t plus 12.7 million ounces inferred. Underground mining projected through 2042, cumulative output 8.5 million ounces. The mine runs on an LTE network with automated electric equipment, on-demand ventilation, worker detection and localization.

East Gouldie inferred resources grew 62% year-on-year in 2025, adding 2.8 million ounces. Selected intersections from the eastern extension drilling: UGED-071-029 at 1,010m depth, 19.8m at 3.5 g/t; UGED-075-057 at 929m, 11.9m at 4.9 g/t; UGED-095-004 at 990m, 9.3m at 6.8 g/t. The 6.8 g/t over 9.3 meters is a strong result by Canadian underground standards. These intersections contributed the bulk of East Gouldie's year-end resource additions.

Now, the Fill the Mill strategy. Odyssey will use about one-third of the Canadian Malartic plant's capacity. From 2028, approximately 40,000 tonnes per day of throughput sits idle. That plant is already built, the depreciation already amortizing. AEM's approach: continuously acquire and explore within the plant's catchment radius, finding satellite ore to feed into existing infrastructure. The Marban deposit from O3 Mining is projected to contribute about 15,000 tpd and roughly 130,000 ounces per year starting around 2033, running about 9 years. Processing satellite ore through an existing plant costs a fraction of what a standalone mine build would require. In the Abitibi Greenstone Belt, where mineralized trends are dense and geological conditions are thoroughly mapped, there is no shortage of candidates.

Nothing like this exists at Newmont or Barrick. Barrick's growth story is about Fourmile (currently triggering a legal war with its JV partner) and copper projects years from production. Newmont's growth story is a question mark attached to a declining production profile. AEM's growth story is a mill with 40,000 tpd of spare capacity sitting in the middle of one of the world's most productive gold belts, waiting to be fed. The economics improve with every satellite source added because the fixed cost base does not grow. This is an industrial strategy wearing the clothing of a mining company, and no industry overview gives it the space it warrants.

2026 exploration and corporate development budget: $565 to $635 million, about $384 million for exploration alone. Detour Lake underground ramp advancing, underground indicated resources at 3.47 million ounces. Hope Bay Patch 7 being drilled. All in Canada. About 85% of gold production and 87% of mineral reserves in Canadian jurisdictions. Forty-two consecutive years of dividends since 1983.

Guy Gosselin, AEM's exploration VP, said in the February 2026 update that the company has built the strongest project pipeline in its history with exploration upside arguably the best in the gold mining sector. AEM's management is known for understatement, which makes that assessment unusual from them. The exploration data from Odyssey, Detour Lake, and Hope Bay appears to back it up.

Production up 35% year-on-year. Global number four. Zijin does not work the way Western miners do. Instead of spending decades optimizing one world-class asset, it stacks volume through rapid-fire global acquisitions. Akyem from Newmont in 2025. A $4 billion bid for Allied Gold in early 2026. Western miners take a year or two per deal. Zijin runs several simultaneously.

State backing provides cheap financing and high tolerance for jurisdictional risk. Chinese mining engineering teams execute on timelines and costs that Western operators cannot match. ESG standards and governance transparency lag well behind Western peers and the stock carries a persistent discount. For international investors, owning Zijin is closer to an emerging-market growth bet. If Zijin sustains this acquisition pace it could be second or third in global production rankings within a few years, which would force a rethinking of how GDX-type indices are constructed.

Kinross: over 2 million gold-equivalent ounces for the second straight year, Paracatu and Fort Knox saw grade improvements, Great Bear in Ontario is the future catalyst but years from production, three U.S. development projects advancing. A steady operator. Allocation capital.

Northern Star: KCGM includes one of Australia's largest open-pit gold mines, the A$5 billion De Grey acquisition could theoretically push annual production to 3 million ounces, fiscal 2026 guidance already cut due to “isolated negative events” late in 2025. Australian labor inflation and integration risk mean the growth curve will be bumpier than the investor presentation suggests.

Neither company is shaping the strategic direction of the industry right now. Kinross is executing competently without ambition. Northern Star is ambitious but unproven at the scale it is targeting.

Franco-Nevada's 2026 revenue is over 80% from gold and precious metals royalties and streams. These companies lend money to miners during the construction phase when capital needs are most acute, and in return they get the right to buy a portion of future output at well below market price. When the mine overruns its budget, when grades disappoint, when the host government rewrites the mining code, the royalty company's exposure is limited because it does not operate anything. Gold comes out, royalty gets paid.

The Nevada dispute has been instructive. Royal Gold holds royalty interests in the Cortez district. As Fourmile development accelerates, Royal Gold's cash flow goes up regardless of which side wins the Newmont-Barrick fight. Wheaton Precious Metals has also been publicly citing the NGM dispute as validation of the streaming model. The spectacle of the world's two largest gold producers locked in litigation over who diverted whose engineers from their joint venture is strong advertising for a business model that avoids operating mines entirely.

Every historical gold rally came with AISC rising in step. Energy, labor, equipment costs inflate, and margin expansion falls short of the gold price move. The 2025-2026 cycle broke the pattern. Gold roughly doubled from $2,500 to $5,000. Industry-average AISC barely moved. Automated mining, digital mine management, electrified haul trucks, all seeing higher penetration in recent years, structurally suppressing marginal cost growth.

There is a complicating factor though. When gold surges, miners tend to defer low-grade ore because high-grade material at elevated prices already throws off enough cash. This suppresses per-ounce AISC in the short term but accelerates reserve depletion. How much of the “decoupling” is genuine technological progress and how much is cyclical grade management is hard to untangle. The distinction matters enormously for forecasting margins when gold normalizes. If it is mostly technology, margins have structurally shifted upward. If it is mostly grade selection, the AISC snapback when miners return to lower-grade ore could be sudden.

At gold in the $3,500-4,000 range, top miners' margins remain comfortable. Below $3,000, margins compress across the board. AEM at $1,275 guidance retains ample room. Barrick approaching the $1,950 guidance ceiling has almost none.

After Mali, the market put an explicit price tag on safe-jurisdiction premiums. Same per-ounce profit from a mine in Ontario versus the West African Sahel receives valuation multiples 30% to 50% apart. Q3 2025 AISC data showed clear regional divergence: Africa, Oceania, and South America saw the steepest cost increases, North America's pace notably slower.

Before 2024, jurisdiction risk occupied a boilerplate paragraph in the risk factors section of investment memos. Almost no sell-side analyst applied explicit jurisdiction discounts in valuation models. Mali turned the boilerplate into the centerpiece of the memo.

AEM has 85% of production and 87% of reserves in Canada. Barrick lost nearly a year of production in Mali and faces a JV dispute in Nevada. That jurisdictional contrast explains a significant portion of the valuation gap between the two companies and it will continue to widen as the market prices jurisdiction risk more explicitly.

Gold broke $5,000 in early 2026, briefly touched above $5,400. What determines whether this lasts: central bank gold purchases maintaining their four-year extraordinary pace, the Fed's rate path, Barrick's NewCo IPO navigating Newmont's blocking claims, AEM's Fill the Mill strategy continuing to find feed in the Abitibi belt.

Barrick excluded Fourmile from the JV in 2019. Early-stage discovery, decent grades, insufficient certainty at the time. Six years and two resource doublings later, it is a 17 g/t project with 15-million-ounce-plus potential, and also the fuse on the biggest corporate governance conflict in gold mining this year. Mining decisions take five or ten years to show whether they were right, and the answers tend to arrive in ways nobody planned for.