Fair Contracts, Fair Returns: How Other Countries Structured Their Biggest Mine Deals
Tanzania, Indonesia, Western Australia and Botswana compared with Oyu Tolgoi
By Dulguun Maidar
Report|October 31, 2026|10 min read
Key takeaways
Mongolia owns 34% of Oyu Tolgoi, but it financed that stake with loans, so 17 years on it has received no dividends. Other countries' experience shows the difference lies in three things: how the state's stake was granted, whether the benefit split is measured and reconciled in numbers, and whether the contract is public.
Oyu Tolgoi: under the 2009 agreement the state's 34% was financed by its partner's loans, which are repaid before any dividend. US$2.4 billion of debt was written off in 2022 and the loan rate cut by 2.5 percentage points in 2026, but no dividend has arrived.
Tanzania in 2019 took a 16% free carried stake in three Barrick gold mines and agreed to split economic benefits 50/50, with a yearly true-up to keep the split.
Indonesia in 2018 paid US$3.85 billion for 51% of Freeport's Indonesian unit, replaced the old contract with a special licence, and tied the extension to building a smelter at home.
Western Australia turns every major mine agreement into an Act of Parliament and publishes the full text.
For Mongolia: for new strategic deposits, take the state's stake as a free carried interest rather than a loan, set a numeric benefit-split target reconciled and published every year, have Parliament ratify and publish major contracts, and tie extensions to in-country processing.
Oyu Tolgoi: a stake without dividends
The Oyu Tolgoi agreement gave Mongolia ownership but postponed its cash. The state funded its stake with money borrowed from its partner, not its own capital, so the mine's profits first go to repaying that loan.
Exhibit 1
Item
Status
Investment Agreement
Signed October 2009; initial term 30 years, after which the state's stake can rise to 50%; the state's 34% can never be diluted (Oyu Tolgoi)
Funding the state's stake
The partner funds the state's share of costs and Mongolia incurs a debt; it is repaid with interest from the mine's cash flow, never directly by the state; dividends are paid only after the loans are repaid (Oyu Tolgoi)
US$2.4 billion of Mongolia's debt written off: US$1.4 billion of funding plus US$1 billion of accrued interest (NS Energy)
Taxes and payments
About US$6.1 billion in taxes, royalties and other payments since 2010 (GeoConversation)
Dividends
Originally expected in 2017; the earliest realistic estimate is now 2037 (GeoConversation)
2026 terms
Interest on the remaining loan cut by 2.5 percentage points (GeoConversation); the loan rate is now reviewed every 3 years instead of every 7 (Arctus)
Benefit split
According to the National Security Council, Mongolia received 25–31% of benefits over 17 years, mostly through taxes, against the 53% the agreement intended (Arctus)
Contract transparency
In 2019 Mongolia published 162 contracts, but most central-government contracts remained secret (NRGI)
The core problem. Ownership is not the same as benefit. When the state's stake is funded by loans, every cost overrun and delay grows the debt and pushes dividends further away. At Oyu Tolgoi, project costs rose from US$5.3 billion to US$6.75 billion and commercial production slipped by nearly two years (GeoConversation).
Tanzania: a free stake and a 50/50 split
When Tanzania settled a long dispute in 2019, it took the state's stake free of any loan and fixed an equal split of total economic benefits in numbers.
Exhibit 2
Chapter
What Tanzania did
4. Settlement agreement
20 October 2019: Barrick agreed to pay US$300 million to settle all tax and other disputes (Barrick)
3. State stake
The government holds a 16% free carried stake in each of the Bulyanhulu, North Mara and Buzwagi mines (Barrick)
7. Benefit split
Economic benefits are shared 50/50; the state's half counts taxes, royalties, clearing fees and its share of cash distributions; an annual true-up keeps the 50/50 split (Barrick)
5. Company structure
A new company, Twiga Minerals, manages the three mines and gives the government full visibility of and participation in operating decisions (Barrick)
12. Disputes
An Africa-focused international dispute resolution framework; the concentrate export ban was lifted (Barrick)
8. Local content
96% of the workforce is Tanzanian and 49% comes from nearby communities; more than 90% of procurement goes to Tanzanian suppliers (Barrick)
Results
US$4.79 billion injected into Tanzania's economy since 2019, including US$558 million in the first half of 2025 (Barrick)
What Mongolia can copy
Take the state's stake as a free carried interest, not a loan, so no debt builds up and dividends are not pushed back.
Agree the split of total benefits (taxes + royalties + dividends) in numbers, not just an equity percentage.
Reconcile the split every year and correct any gap the next year. Oyu Tolgoi's 53% target could be measured this way.
What doesn't transfer directly
Tanzania's deal followed an export ban and a mine closure; that path carries a high cost in investor confidence.
It came when Barrick took over the mines from Acacia, giving both sides a fresh start.
Indonesia: 51% and a processing condition
Rather than simply extend its old contract, Indonesia bought a majority stake, moved the mine to a licence regime, and tied the extension to processing at home.
Exhibit 3
Chapter
What Indonesia did
3. State stake
21 December 2018: state-owned Inalum paid Freeport-McMoRan and Rio Tinto US$3.85 billion, raising its stake in PT Freeport Indonesia from 9.36% to 51.23% (ESDM)
7. Regional share
10% of the 51.23% goes to Papua; Inalum lent the regional company US$819 million to buy it, repaid from dividends (ESDM)
1, 4. Contract → licence
The Contract of Work dating from 1967 was replaced by an IUPK special licence, extended by 2 × 10 years to 2041 (ESDM)
8. In-country processing
An obligation to build a smelter within five years (ESDM); the Manyar smelter in Gresik was substantially complete in June 2024 and began commissioning (Freeport)
3. Further stake
The government plans to raise its stake to 61% if the licence is extended beyond 2041 (Indonesia Business Post)
Results
US$4.3 billion paid to the state in 2025; expected to fall to US$2.6 billion in 2026 after the Grasberg landslide, then reach US$4.7 billion in 2027 (including US$1.9 billion in dividends) (Investortrust)
What Mongolia can copy
Tie extensions and new licences to time-bound obligations for in-country processing (smelting, refining).
Give the host region (aimag, soum) its own stake, financed from dividends.
What doesn't transfer directly
Indonesia paid US$3.85 billion in cash for its 51%; without that money, Mongolia buying a stake on credit would repeat Oyu Tolgoi's mistake.
Smelters are expensive; Mongolia would need to plan for power, water and transport first.
Western Australia: contracts ratified by Parliament and public
Western Australia gives every major mining and processing agreement effect through its own Act of Parliament, so the full contract text is open to anyone in the state's legislation database.
Exhibit 4
Chapter
What Western Australia does
4. Contracts as law
The government administers 64 State Agreements, each given effect by an Agreement Act of Parliament, across 16 sectors including iron ore, gold, copper, nickel, coal and gas (WA Government)
10. Transparency
Full agreement texts are published in the legislation database; for example the 1992 Hope Downs iron ore agreement (WA Legislation)
History
The first agreements date from the 1960s: Hamersley Range (1963), Mount Newman (1964), Robe River (1964) (WA Government)
What Mongolia can copy
Have Parliament ratify strategic deposit contracts as law and publish them in full. This gives investors stability and citizens oversight.
Fully implement the EITI requirement to publish all contracts granted, entered into or amended from 1 January 2021 (NRGI).
What doesn't transfer directly
Changing a contract ratified as law requires another law; given how often Mongolia's contracts have been disputed, the terms need careful preparation first.
Botswana: marketing rights
Beyond equity, Botswana made the right to sell its own output a central negotiating point; its 2025 agreement raises the state company's share of sales from 25% to 50% over 10 years. See the Botswana profile for details.
Comparison table
Across three questions, Mongolia's biggest gaps are a loan-funded state stake and no mechanism to reconcile the benefit split.
Exhibit 5
Question
Tanzania
Indonesia
Western Australia
Botswana
Mongolia (Oyu Tolgoi)
How the state's stake was granted
16% free carried
51.23%, US$3.85bn in cash
No direct stake
Debswana 50%
34%, financed by loans
Numeric benefit-split target
50/50, reconciled yearly
No numeric target stated
—
Up to 50% of sales
53% target, 25–31% actual
How the contract is secured
Agreement with government
IUPK licence
Act of Parliament, public
Agreement with government
Investment Agreement
Local conditions
96% local staff, 90%+ local procurement
Smelter
—
Diversification fund
—
Sources are in each country section.
A path for Mongolia
Exhibit 6
Phase
Action
Model
Owner
1. Make it visible (0–12 months)
Calculate Oyu Tolgoi's benefit split against the 53% target every year and publish it
Tanzania
Ministry of Finance, Erdenes Oyu Tolgoi
Publish all mineral contracts since 2021 in full, unredacted
Western Australia, EITI
Ministry of Mining
2. A standard for new deposits (1–3 years)
For new strategic deposits: a free carried state stake instead of a loan, a numeric total-benefit split, and a yearly true-up
Tanzania
Parliament, Government
Ratify strategic deposit contracts as Acts of Parliament
Western Australia
Parliament
3. Extensions and value added (3–5 years)
Tie licence extensions to numeric obligations for in-country processing, suppliers and workforce
Indonesia, Tanzania
Ministry of Mining
Give host aimags and soums their own stake or revenue share
Indonesia
Parliament
This path is a policy proposal, not a legal opinion; each phase needs detailed work with lawyers, finance specialists and contract experts.