The Mongolian Mining Trap
Why Owning the License Is Not the Same as Owning the Asset
Mongolia periodically attracts capital. Capital consistently disappears.
This is not a story about a small landlocked country with an unfortunate geography. It is a story about a system — one that operates wherever Chinese state capital has established infrastructure dependency before equity terms are set. Mongolia is where that system is most legible. The DRC, Indonesia, and the rare earth processing chains underpinning the AI infrastructure buildout are where it is most consequential.
Robert Friedland found the copper at Oyu Tolgoi — and understood early that China held the cards on what that copper was worth. China found something the market took longer to price: that controlling the destination of the metal is more valuable than owning the ground it comes from. Those are not competing discoveries. They are sequential ones. And the gap between them is where most institutional capital has been losing money ever since.
Over two decades, capital from Canada, Russia, Australia, the United States, and China itself has entered Mongolia’s mining sector with licenses, financing, and conviction. Most of it has left through arbitration tribunals, dissolved offshore vehicles, or write-offs that institutional compliance systems never saw coming. The pattern is consistent enough that it cannot be explained by bad luck, bad governance, or bad contracts alone.
Something structural is happening. This piece names it.
The Mechanism
China does not need to own a critical minerals asset to control its economics.
It needs to own what the asset depends on.
Railway access. Offtake destination. Processing capacity. Border throughput. By the time a Mongolian mine is producing at scale, every link in its value chain runs through a Chinese decision point — not because China seized anything, but because those dependencies were established quietly, years earlier, while everyone else was focused on the headline transaction.
China Energy’s approach to Tavan Tolgoi, one of the world’s largest coking coal deposits, illustrates this precisely. China’s largest state-owned energy enterprise did not outbid competitors or force a political outcome. It waited — for more than a decade — while successive Mongolian governments failed to develop the project through international tenders and bilateral negotiations. It waited while the railway that would move the coal was being planned, financed, and built. Then, when the infrastructure dependency was fully established and no alternative off-take route existed, it moved.
Mongolia’s parliament ratified the cross-border railway agreement in April 2025. The terms reflect a decade of patience, not a negotiation.
This is not a story about Chinese aggression. It is a story about a specific strategic capability: dependency is engineered before ownership is formalized. The equity holder gets the license. The patient capital gets the economics.
Western institutional capital has no equivalent playbook. It underwrites assets. China engineers the conditions under which assets can be monetized — or cannot.
What the Public Record Shows
The arbitration record documents what happens when outside capital enters this corridor without understanding the mechanism.
Khan Resources (Canada) — Exploration licenses for a uranium deposit cancelled by regulatory fiat. The investor assumed the license was the asset. A tribunal awarded $100 million in 2015 finding expropriation. Mongolia settled for $70 million in 2016. The mine never operated. The license was not the asset. The license was the hostage.
Paushok v. Mongolia (Russia) — A 68% windfall profit tax on gold sales above $500 per ounce destroyed investment economics overnight. The Mongolian Central Bank took physical custody of the investor’s gold and deposited it abroad. A UNCITRAL tribunal found direct breach of fair and equitable treatment. The tax was repealed — after the capital had left. Retroactive legislation moves faster than any stabilization clause you did not negotiate before the political cycle turned.
Beijing Shougang v. Mongolia (China) — A Chinese state enterprise lost its iron ore licenses and initiated arbitration under the China-Mongolia bilateral investment treaty (PCA Case 2010-20). The tribunal found it lacked jurisdiction to rule on whether an expropriation occurred — because the BIT only permitted arbitration over the amount of compensation, not the fact. A Chinese state enterprise. A bilateral treaty. No remedy. Even the most legally sophisticated capital discovers treaty architecture failures only when the dispute is live.
Three nationalities. Three mechanisms. One outcome. But these are only the cases that reached tribunals.
What Doesn’t Reach Tribunals
The more important extraction dynamic resolves quietly — through offshore holding structures, liquidation proceedings in foreign jurisdictions, and transfer pricing arrangements that never appear in any public filing.
The architecture is standard across frontier market mining investment: an offshore holding company controls an intermediate entity which owns the Mongolian-licensed operating company. The structure has legitimate uses — tax efficiency, investor protection, simplified share transfer. It also creates a vulnerability that the arbitration record cannot capture.
When an offshore-held Mongolia mining vehicle encounters financial difficulty, value distribution follows a predictable direction. Intercompany loan balances, management fee arrangements, and offtake transfer pricing determine where value has accumulated within the structure before any formal insolvency begins. By the time a winding-up proceeding is initiated, the question of what remains for equity holders — including local Mongolian partners — is often already answered.
The operating company’s mining license sits beneath this as a legally ring-fenced but functionally stranded asset. The practical outcome: a license holder attached to a producing mine, selling output at distressed prices to the only available buyer, with no leverage to renegotiate and no legal pathway to recover value lost at the holding level.
In 2024, Mongolia’s parliament added a new layer. A sovereign wealth fund law now requires the government to take a 34% non-compensated stake in any deposit designated as “strategic” — with no limitation on future designations. Every foreign-held mining asset in Mongolia now carries a latent expropriation risk activatable by parliamentary vote with no advance notice.
The legislative response was designed to address extraction. It created a new mechanism for it.
When Sophisticated Capital Loses Too
In October 2024, Trafigura — one of the world’s largest commodity trading houses — confirmed $1.1 billion in losses from its Mongolian refined oil products business. According to Trafigura’s own annual report, the alleged misconduct involved manipulation of receivables over five years — invisible to the company’s accountants in Geneva and Singapore despite operating in a market representing less than 0.3% of its total trading volume.
The mechanism was not sophisticated. The information asymmetry was.
The corridor does not discriminate by the nationality or sophistication of the capital it extracts from. Canadian miners, Russian gold investors, Chinese state enterprises, and one of the world’s most capable commodity trading houses have all paid tuition in this corridor. The common factor is not who they were. It is what they didn’t know — and when they didn’t know it.
Why Mongolia Is The Most Readable Version Of A Global System
Mongolia’s relevance to this analysis is not geographic. A landlocked country sandwiched between Russia and China will always face asymmetric dependencies — that observation requires no intelligence to make.
What makes Mongolia analytically valuable is different: it is the only place where Chinese industrial policy execution is small enough to read in full — border crossing by border crossing, offtake contract by offtake contract, parliamentary vote by parliamentary vote. What takes decades to become visible in the DRC or Indonesia is legible here in a single project cycle.
Mongolia is not the destination. It is the place where you learn to read the system before it appears somewhere that actually moves your portfolio.
The same dependency architecture — logistics control established before equity terms are set, offtake destinations determined before production begins, processing capacity controlled downstream before upstream investment is made — operates across every critical minerals corridor Chinese state capital has entered at scale.
Copper from Mongolia crosses the border into Chinese smelters on terms set years before the ore arrives. Mongolian coking coal feeds Baotou’s steel operations, which underpin the northern rare earth processing system — meaning border throughput decisions in southern Mongolia ripple directly into rare earth output schedules weeks later. REE processing is controlled so completely at the downstream stage that holding the mining license is the least leveraged position in the entire chain.
If value can be extracted before the border crossing in Mongolia, it can be extracted at every subsequent processing stage. The corridor is not the anomaly. It is the template.
Investors who cannot read the system here will not read it in the DRC, in Indonesia, or in the rare earth processing chains that underpin the AI infrastructure buildout they are currently financing.
Closing
Mongolia is not struggling to make the case for its mineral endowment. It is exceptionally efficient at extracting the capital that case attracts.
The system is not broken. It is functioning exactly as designed.
The only question for any investor entering this corridor — or any corridor where Chinese state capital has established dependency before ownership — is whether they are inside the extraction layer or upstream of it.
That distinction is not visible in standard diligence. It requires reading the corridor at a different layer, earlier, before the dependency is complete.
The Mongolian Mining Trap is the first in a series examining how China’s resource architecture operates across critical minerals supply chains — and where the early warning signals are visible before price discovery confirms them.
About MSIQ
The author has spent over a decade working at the operational level of the Mongolia-China resource corridor — across mining finance, offtake negotiation, and commodities trading — with prior senior roles at international financial institutions covering Mongolian and Northern Chinese resource markets.
MSIQ principals maintain direct exposure to the markets they cover. We do not publish from the sidelines.
MSIQ is currently tracking active holding-structure stress indicators and border logistics anomalies in the Mongolia-China corridor — and how those signals translate into broader critical minerals chokepoint dynamics across processing, export licensing, and supply chain policy. To discuss current corridor signals: info@midstreamiq.com
Sources
Khan Resources Inc. v. Mongolia — ICSID Case No. UNCT/10/2. Award 2015. Settlement 2016.
Sergei Paushok, CJSC Golden East Company and CJSC Vostokneftegaz Company v. Mongolia — UNCITRAL. Award 2011.
Beijing Shougang Mining Investment Company Ltd. and others v. Mongolia — PCA Case No. 2010-20.
US Department of State — 2025 Investment Climate Statement: Mongolia.
Interfax — Mongolia Parliament Ratifies Gashuunsukhait-Gantsmod Cross-Border Railway Agreement. April 2025.
Mining.com — Mongolia-China Railway Extension to Increase Coal Transport Capacity by 30Mt. 2025.
Trafigura — FY2024 Annual Report. October 2024.