Robert Sterling – AsiaPresswire – Indonesia banned the export of unprocessed nickel ore in January 2020, pulling forward a deadline that had originally been set for 2022. Five years on, it is the most consequential industrial policy experiment in Southeast Asia, and the clearest test anywhere of whether a commodity-producing country can force its way up the value chain by legislating against its own exports.
The answer is genuinely mixed, and the mix is more instructive than either the celebratory or the dismissive version of the story. Indonesia did build a processing industry that did not exist. It also created chronic global oversupply, locked itself into a technology and capital structure it does not control, and discovered that banning an export is much easier than capturing the margin that sits at the end of the chain.
This piece is about what the policy actually built, what it cost, where the money actually goes, and what a company deciding whether to commit capital there needs to test before it writes a cheque.
The policy, stated precisely
The measure most people mean by “downstreaming” is a prohibition on exporting nickel ore, introduced through a mineral and coal mining regulation and brought forward to take effect in January 2020. The logic was straightforward: if ore cannot leave, processors must come to the ore.
Three things about it are worth being precise about, because the shorthand versions mislead.
It is not a tax. It is a prohibition. That distinction matters for modelling, because a tax can be priced and hedged while a prohibition reorders the physical geography of an industry.
It is not limited to nickel. Indonesia has applied the same logic to bauxite and has extended the approach, in various forms, to other commodities including copper and tin. Companies treating this as a nickel-specific rule are reading the policy too narrowly.
It is not static. Thresholds, permitted product forms, royalty structures, and domestic-market obligations have been adjusted repeatedly since 2020. Any model built on the rules as they stood in a given year needs re-checking against the rules as they stand now.
The instrument has also been paired with inducements: tax holidays, investment allowances, and the designation of industrial estates where processing is concentrated. The two largest are the Morowali industrial park on Sulawesi and the Weda Bay complex on Halmahera in North Maluku. Most of the country’s processing capacity sits inside those two sites, which is itself a concentration risk worth naming.
What the country actually became
On the narrow question of whether the policy worked as industrial policy — did processing capacity appear? — the answer is unambiguously yes.
Indonesia moved from exporting raw ore to exporting ferronickel, nickel pig iron, matte, mixed hydroxide precipitate, and, further along, nickel sulfate and precursor material. By the middle of the decade it accounted for a share of global mined nickel supply approaching or exceeding half, depending on the year and the source consulted, up from a fraction of that before the ban. Export revenue from nickel products rose from low single-digit billions of US dollars in the ore era to tens of billions afterwards — a genuine, large, and fast shift in the composition of the country’s exports.
Employment at the two main parks runs to tens of thousands of direct jobs. That is a real number in a country where the alternative use of remote land in Sulawesi and Halmahera was close to nil.
What is less often stated precisely is what that capacity is for. The overwhelming majority of Indonesian processing capacity was built for stainless steel — ferronickel and nickel pig iron feeding the stainless chain — rather than for batteries. The battery-grade portion, produced mostly through high-pressure acid leaching into mixed hydroxide precipitate and then into sulfate, is real and growing but remains the smaller part of the complex. A country that banned ore exports to serve the electric vehicle revolution ended up, in its first phase, mostly serving stainless steel.
Where the chain actually stops
| Stage |
What it produces |
Where Indonesia sits |
What it captures |
| Mining |
Laterite and saprolite ore |
Dominant global position |
Lowest margin, highest physical footprint |
| Smelting (RKEF) |
Ferronickel, nickel pig iron |
Very large, largely built |
Commodity margin, tied to stainless demand |
| Refining (HPAL) |
Mixed hydroxide precipitate |
Growing, capital-intensive |
Higher margin, exposed to cobalt and acid costs |
| Sulfate and precursor |
Battery-grade nickel chemicals |
Present but limited |
Approaching cell-grade pricing |
| Cathode and cell |
Cathode active material, cells |
Early / under construction |
The margin everyone wants |
| Pack and vehicle |
Battery packs, EVs |
Assembly only |
The margin downstreaming was aimed at |
The table is the argument. Indonesia is unambiguously dominant at row one, very strong at row two, present at row three, limited at row four, and not yet established at rows five and six. The stated ambition of the policy was row six. The delivered result, five years in, is mostly rows one to three.
This is not a criticism of execution so much as a description of what happened. Each step up the table requires a different input — not more ore, but technology, process know-how, qualification with cell makers, and access to a customer who will accept your material into a battery supply chain. Those are not things a prohibition on ore exports can compel. They have to be bought, licensed, or built over years.
The stainless anchor, and why it still sets the floor
Battery demand gets the attention, but stainless steel still absorbs the majority of primary nickel, and that fact is what makes the Indonesian build-out financially legible at all. RKEF lines producing ferronickel and nickel pig iron have a customer base that exists today, in volume, and does not require qualification. That is why the first phase of downstreaming went there rather than into battery chemicals: the capital was available, the technology was mature, the off-taker was known, and the product could be sold the day the line started.
This has a stabilising effect that is easy to miss. When battery-grade nickel prices weaken or when a cell maker delays a programme, the stainless chain keeps absorbing metal. It acts as a floor under offtake, and therefore under utilisation, in a way that a purely battery-facing industry would not have. It also means that the economics of most Indonesian capacity are driven first by stainless demand in China and India and only second by electric vehicle forecasts — which is the opposite of the framing most coverage uses.
The consequence for anyone modelling this market is that “nickel demand” is not one number. It is two largely separate demand pools with different price sensitivities, different customers, and different contract structures, connected mainly by the fact that producers can switch some capacity between them. Treating them as a single market is why so many forecasts of the 2021–2023 period were wrong in the same direction.
What it cost
The costs are real and they fall in four places.
Capital. Smelting and refining are among the most capital-intensive industrial activities in the commodity world. A rotary kiln–electric furnace line is expensive; a high-pressure acid leach plant is more expensive still, and its economics depend on ore grade, acid supply, and the cobalt credit in the byproduct stream. The billions of dollars invested came overwhelmingly from foreign capital, and the single largest source of that capital is Chinese.
Power. Indonesian smelting is heavily coal-fired, with captive power plants built alongside the lines. That has two consequences. The carbon intensity of the resulting nickel is high, which increasingly matters for buyers who have to report it. And it ties the economics of the metal to the cost and availability of domestic coal, which the government also prices and regulates.
Environment. Laterite mining clears land and moves very large volumes of material. The tailings question is the hardest one, and in parts of the country the practice of disposing of tailings at sea has drawn sustained criticism. Acid leaching creates its own waste stream. Companies sourcing Indonesian nickel and reporting against European battery or supply-chain due diligence rules should assume these issues will be raised, and should have sourcing documentation that withstands it.
Market structure. This is the cost that was least anticipated. Indonesia added processing capacity faster than global nickel demand grew. The result, visible through 2023 and into the following years, was sustained oversupply and a nickel price that fell well below the level at which the newest and highest-cost capacity is comfortable. Parts of the industry outside Indonesia — Australian laterite operations and several projects elsewhere — were squeezed or shut. Then the pressure came home: margins inside Indonesia compressed as well.
The part the brochures skip: who captures the margin
A prohibition on ore exports guarantees that processing happens in the country. It does not guarantee that the profit from processing accrues to the country or to its companies.
The ownership structure of much of the built capacity is foreign. Offtake is frequently tied to the same groups that provided the capital. Financing, engineering procurement, and construction were largely supplied from outside. In that arrangement, Indonesia receives employment, taxes and royalties, the depreciation of a large physical asset base, and the strategic position of being indispensable to a critical supply chain. Those are not nothing — employment and fiscal receipts are the point of industrial policy — but they are not the same as capturing the margin at the cell and vehicle end of the table.
There is a second-order effect here that deserves attention. Because the dominant capital and offtake relationships run through Chinese groups, Indonesian nickel has become entangled in the classification rules of other countries’ industrial policy. The United States clean-vehicle credit regime applies foreign-entity-of-concern restrictions that trace ownership and control through the supply chain. European rules apply different but overlapping tests. The practical consequence is that material which is perfectly saleable into one market may be disqualified from another, and the disqualification can be triggered by an ownership stake several levels up rather than by anything about the metal itself.
For a company building a supply chain, this means the question is never simply “is there nickel available in Indonesia.” It is “which corporate entity produced it, who controls that entity, and where does the resulting material qualify.”
Qualification: the step nobody budgets for
Selling metal is one thing. Selling metal into a battery supply chain is a different and much slower business, and the gap between the two is where a great deal of announced capacity quietly stalls.
Battery makers do not buy material on specification sheets. They qualify it, plant by plant, through a process that runs from months to well over a year. Samples are produced, cell-level testing is run against cycle life and safety behaviour, the plant’s process control is audited, and the material is then approved against a specific production line. Change the plant, change the process, or in some cases change the ore blend feeding it, and the qualification has to be revisited.
This has three practical implications. First, a refinery that has not been qualified has no battery customer regardless of how good its product is on paper; it can only sell at commodity pricing, which is precisely the margin it was built to escape. Second, the bottleneck is not always chemistry — it is often simply the queue at the cell makers, who can only run a limited number of qualification programmes at once. Third, the value of an existing qualification is high and underappreciated in deal valuations, because it cannot be bought quickly at any price.
There is a related trap on the buying side. A purchaser that signs an offtake for “battery-grade nickel” from a plant under construction is not buying battery-grade nickel; it is buying a schedule, a qualification process, and a set of risks that sit between the two. Contracts that do not distinguish those states — and that do not allocate the cost of a failed or delayed qualification — are the ones that end up in dispute.
What other countries tried, and what happened
Indonesia is not the first country to try this, and the comparison is instructive because the results differ in ways that track the specifics rather than the ambition.
Chile has moved on lithium through a different instrument — a state-participation model for its salt flat operations, with the state taking control or majority stakes in new projects rather than a blanket export prohibition. The approach preserves more room for existing operators and leaves the question of where processing happens partly open. It also moves more slowly.
Guinea holds the world’s largest bauxite reserves and has pushed for domestic alumina refining for years, with results that have been slower and more uneven than Indonesia’s, largely because the capital and infrastructure requirements are less forgiving and the bauxite market is less concentrated.
Zimbabwe and Namibia both moved to restrict raw lithium exports, with Zimbabwe banning unprocessed lithium ore exports. The stated logic is identical to Indonesia’s. The scale of the resulting domestic industry, so far, is not — which is a reminder that a prohibition only works as industrial policy when there is capital ready to build against it and a customer willing to buy the output.
The Philippines is the closest analogue and the most interesting case. It is a large nickel ore producer and has debated an export ban for years without implementing one, which has left it exporting ore into the Indonesian complex it might otherwise have competed with. The comparison suggests that the speed and decisiveness of Indonesia’s move mattered at least as much as the move itself.
Across all of these, the pattern is consistent. Prohibitions work when three conditions hold: the country’s share of supply is large enough that the ban changes global trade rather than just diverting it, capital is available and willing to build on the country’s terms, and there is a market for the processed product that does not require the producer to win a qualification race first. Indonesia had all three for stainless-grade material and only the first two for battery-grade.
Questions companies actually ask
Can we get battery-grade material, or only stainless-grade? Both exist, but battery-grade capacity is the smaller share and is concentrated in specific plants. Qualification with a cell maker is a separate and slower process from buying metal.
Is the ore export ban at risk of being reversed? Treat it as permanent policy direction, not a temporary measure. It has survived changes of administration and legal challenge. Model it as a structural feature of the market.
What happens if the price stays low? Capacity built for a higher price does not disappear; it runs at lower margin, and consolidation follows. For a buyer, low prices are favourable. For an investor, they change the project’s whole return profile, and the projects most exposed are the newest and highest-cost ones.
How do we handle the carbon intensity? You report it. The footprint of coal-fired smelting is materially higher than alternative supply, and buyers subject to reporting obligations should be sourcing data at the plant level rather than the country level.
Does Indonesian material qualify for incentives in our target market? Check ownership and control, not just origin. This is the single most common and most expensive sourcing error in this market right now.
Three ways this resolves
The next five years are genuinely uncertain, and the honest framing is scenario rather than forecast.
Consolidation. Prices stay low enough that marginal capacity stops running, and the industry concentrates into the lowest-cost operators. Indonesia’s position strengthens on volume while its fiscal take per tonne weakens. This is, on present evidence, the path of least resistance.
Escalation. Indonesia moves decisively up the table — cathode, cell, and eventually vehicle assembly, supported by incentives and by requirements attached to domestic market access. This requires technology partnerships and customer qualification that no export rule can mandate. Several projects are aimed at it. Whether they reach commercial scale on schedule is open.
Fragmentation. Buyers in the United States and Europe build qualified supply outside Chinese-controlled chains, while Chinese-linked capacity serves the rest of the market. Indonesian volume keeps growing but splits into two pools with different prices and different customers. Elements of this are already visible.
These are not mutually exclusive. Consolidation and fragmentation can happen at the same time, to different parts of the industry.
Glossary
- Laterite — The ore type Indonesia mines, requiring different processing from the sulphide ores common elsewhere.
- RKEF — Rotary kiln–electric furnace, the route to ferronickel and nickel pig iron for stainless steel.
- HPAL — High-pressure acid leaching, the route to mixed hydroxide precipitate for battery-grade material.
- MHP — Mixed hydroxide precipitate, the intermediate product that HPAL produces and refiners convert to sulfate.
- FEOC — Foreign entity of concern, the ownership-and-control test that determines clean-vehicle credit eligibility in the United States.
The risks worth pricing
Technology lock-in. Much of the built capacity was designed around the ore, the capital, and the customer relationships available at the time it was built. If battery chemistry shifts away from high-nickel formulations, that capacity is less valuable than its construction cost.
Environmental liability. Land disturbance and tailings practice are live issues now and will remain so. Expect tightening, and expect the compliance cost to land on producers and then on buyers’ documentation requirements.
Concentration. Two industrial parks carry most of the capacity. A single-site disruption — power, labour, logistics, regulation — is a supply-chain event, not a local one.
Policy drift. Royalty rates, export levies, domestic processing obligations, and local content requirements have all moved. A model built today should be stress-tested against at least two alternative policy settings.
The commodity cycle itself. Nickel is a cyclical commodity with a volatile price history, including a short squeeze in 2022 that remains the clearest illustration of how unstable this market can be. Any project economics built on a spot price rather than a through-cycle assumption should be redone.
A checklist before committing capital
- Establish which row of the chain you are actually investing in, and whether the margin at that row justifies the capital
- Trace ownership and control of your counterparty to the level required by your target market’s rules — not to the level that is convenient
- Source emissions data at plant level, and confirm it survives your reporting obligations
- Model at least two nickel prices and two policy settings; if the project only works at one, it is not a project
- Confirm power supply, logistics, and port access are contracted rather than assumed
- Check whether your offtake customer has already qualified the specific plant, or whether that process has not started
- Ask what happens to your asset if battery chemistry moves away from high-nickel
The honest bottom line
Indonesia’s nickel policy is neither the triumph its supporters describe nor the cautionary tale its critics prefer. It did the thing it set out to do: it moved processing into the country, created a large industrial base where there was essentially none, and changed the structure of a global commodity market within five years. Very few industrial policies in the developing world have done that much, that fast.
What it did not do is the harder part. Banning an export moves the location of an activity. It does not move the location of the margin, the technology, or the customer relationship — those are earned at the top of the chain, and they are earned by qualification, process capability, and access, none of which respond to a prohibition.
For a company deciding what to do about it, the practical conclusion is unromantic. Indonesian nickel is abundant, cheap, and increasingly central to the world’s battery supply. It is also concentrated, carbon-intensive, policy-exposed, and tangled in other countries’ ownership rules. Those facts are not contradictory. They are the deal, and the companies that do well here are the ones that priced all of them rather than the ones that found the cheapest tonne.
source
https://newsroom.seaprwire.com/press-releases/technologies/indonesias-nickel-downstreaming-what-a-decade-of-industrial-policy-actually-produced/
