Asian Ferro-Alloy Stability Is Built on Cost Pain, Not Demand

Ferro-alloy prices across Asia are holding steady in late September 2026, but not for the reason most buyers would prefer. Coke prices in northern China have climbed $62.50 per tonne in under two months. Indian producers have exhausted the cheap raw material inventories that once protected their margins, and are now repricing against far more expensive imports. Across Malaysia and Indonesia, drought conditions are making firm export commitments hard to guarantee.

The floor under the market is real. But it is built from producer cost pain, not from any surge in steel demand.

That distinction is what matters most for anyone tracking steelmaking inputs. A price floor propped up by elevated operating costs behaves differently from one supported by end-user pull. It is more brittle. It holds only until producers must choose between defending margins and keeping furnaces running, and it can give way quickly if raw material costs retreat or downstream buyers find cheaper supply.

The Asian ferro-alloy market outlook heading into winter therefore hinges on the architecture of that floor, region by region. What follows here maps the specific cost pressures across each producing region, examines how they interact, and identifies the conditions under which current price stability could break down. It is the data needed to judge whether today’s steady prices reflect durable supply tightening or temporary cost pain, and what that answer means for positioning into the 2026-27 winter.

How rising coke and energy costs are squeezing northern China’s ferro-alloy smelters

Start with coke, because that is where the pressure originates. Fastmarkets assessed its weekly coke price (65% CSR, fob China) at $340 per tonne as of 22 September 2026, up from a midpoint of $275-280 per tonne recorded on 4 August 2026.

The coke price rise of $62.50 per tonne in under two months is the single clearest signal of how fast the cost stack is building for Chinese smelters.

That move does not stay confined to the coke line. One market participant told Fastmarkets that higher coke expenses alone had lifted ferro-alloy production costs by over 200 yuan (roughly $30) per tonne of alloy over about two months, based on a coke usage rate of around 0.5 tonne per tonne of alloy output. That is the coke layer.

Now add electricity. Certain northern smelters reported tariffs rising to 0.45 yuan per kWh in August 2026, up from 0.40 yuan per kWh in July. Two compounding cost lines, moving in the same direction, at the same time.

The upstream picture reinforces the squeeze. Dalian coking coal futures peaked at 1,729 RMB per tonne on 31 August 2026, the highest in over two years, before easing roughly 12% to 1,523.50 RMB per tonne. A fatal Shanxi mine accident in May 2026 triggered the supply squeeze behind that spike, and BI analysts cited by Mining Weekly expect full normalisation only in H1 2027.

The Shanxi safety shutdowns that triggered the coking coal supply squeeze were not a routine disruption; the fatal mine accident in May 2026 initiated a cascade of production halts whose full restoration timeline extends into H1 2027, keeping the cost floor elevated well beyond what a brief interruption would sustain.

Against that cost backdrop, Chinese ferro-alloy spot prices are holding rather than falling:

• Silico-manganese (65% Mn min, in-warehouse China): 5,550-5,700 yuan per tonne as of 18 September 2026
• Ferro-silicon (75% Si min, in-warehouse China): 6,000-6,300 yuan per tonne as of 23 September 2026
• Ferro-chrome (spot, basis 50% Cr, ddp China): 8,000-8,200 yuan per tonne as of 22 September 2026

What this tells you is that Chinese smelter margins are under genuine pressure from several directions at once. That is precisely why asking prices are steady even without strong downstream demand. China’s producers set the benchmarks against which global supply is priced, so their floor bid effectively anchors the wider market.

Ulanqab’s electricity competition: AI data centres versus ferro-alloy smelters

There is a structurally different risk layered on top of coke and tariffs. Ulanqab, one of China’s largest ferro-alloy hubs, produced 13.24 million tonnes of ferro-alloys in 2025, and it now hosts growing artificial intelligence data-centre infrastructure competing for the same low-cost power.

A market source cited by Fastmarkets flagged this competition as a potential upward pressure on smelter electricity costs, unless additional affordable generation capacity comes online.

Whether that dynamic is structural or merely cyclical remains unresolved in current published commentary. It belongs on the watch list as an emerging risk rather than a settled conclusion, but for a hub of Ulanqab’s scale, even a modest sustained shift in power economics would ripple across the alloy complex.

India’s depleted stockpile problem: why cheap inventory is gone and what that means for export prices

For a long stretch, Indian ferro-alloy producers held a quiet advantage: cheaper coke and coking coal bought earlier and drawn down slowly. That buffer is now largely gone.

India relies on imported coke and coking coal, so with the low-cost inventory consumed, replenishment is happening at significantly higher international prices. The protected cost position has become a fully exposed one.

The margin damage was already visible in the data. BigMint’s benchmark high-carbon ferro-chrome (HC 60%, Si 4%) averaged INR 100,137 per tonne ex-works Jajpur in H1 CY 2025, down from INR 112,558 per tonne in H1 CY 2024, an 11% year-on-year fall. Export prices to key East Asian markets fell further still.

The volume story is starker than the price story.

Total Indian ferro-chrome exports fell 23% year-on-year in H1 CY 2025, the clearest single measure of producers withdrawing from the market.

Those lower prices cut into margins and forced selective participation, meaning some producers simply pulled back from sales at prevailing prices. Current assessments show where that leaves the export benchmark: high-carbon ferro-chrome (57-65% Cr) sat at $1.01-1.08 per lb cif dup Japan and $0.99-1.06 per lb cif dup South Korea as of 17 September 2026.

This dynamic was not new even a year earlier. Argus, in December 2024, framed Indian ferro-chrome around Rs 104,000-106,000 per tonne as likely to face further pressure into Q1 2025, a signal that margin compression has been a persistent theme rather than a fresh shock.

What this tells you is that Indian producers are being squeezed from both ends: higher costs and lower revenue per tonne. That limits how aggressively they can compete on price, while making them choosy about which shipments they commit to. India is one of the primary alternative supply sources for ferro-chrome into Japan and South Korea, so when its producers selectively withdraw, availability tightens at exactly the consuming hubs where stainless steel feedstock buyers are watching most closely.

Indian ferro-alloy export constraints have been building from multiple directions simultaneously; the margin compression from depleted cheap coke inventories compounds trade-side restrictions that have already redirected Indian manganese alloy volumes away from key consuming regions, leaving East Asian buyers with a narrower set of reliable alternative origins.

Southeast Asia’s hydropower squeeze: how water shortages are disrupting Malaysian and Indonesian ferro-alloy output

Malaysia earned its place in the ferro-alloy trade on one advantage above all: competitively priced hydroelectric power. Cheap, reliable water-generated electricity is what made its ferro-silicon and silico-manganese cost-effective on the export market.

That advantage is now under direct threat. An ongoing water shortfall is making it difficult for Malaysian producers to commit to firm export volumes, including their regular shipments to Japan.

Prices reflect a market pricing in that uncertainty rather than reacting to a fresh shock:

• Ferro-silicon (75% Si min, cif Japan): $1,290-1,310 per tonne as of 23 September 2026, narrowing from $1,290-1,330 per tonne the prior week
• Nickel pig iron (10-14% Ni, fob Indonesia): flat at $135-147 per nickel unit as of 23 September 2026

The Indonesian picture is worth handling carefully. Water scarcity has been flagged as a factor affecting local smelting activity, but the flat nickel pig iron price shows the impact has not yet reached alloy prices in any measurable way.

A caveat worth stating plainly: no traceable government or energy-authority data on reservoir levels or generation shortfalls is currently available in published sources. Any read on the hydropower impact remains qualitative, not quantified.

The Indonesian hydropower shortfall in South Sulawesi, documented by Indonesia’s ESDM Ministry and state utility PT PLN, saw aggregate generation collapse from 850 MW to 250 MW under severe drought conditions, a quantified measure of the supply stress that published ferro-alloy commentary has so far described only in qualitative terms.

What this means for procurement is more consequential than any single week’s price move. Malaysian producers unable to guarantee firm export volumes means Japanese and regional buyers now carry reliability risk on a supply origin they have long treated as dependable.

If that constraint persists into Q4 2026, Japanese buyers face a choice heading into higher winter demand: pay up for alternative origins, or accept supply uncertainty. Either way, the competitively priced Malaysian stream can no longer be assumed. That is a variable worth tracking closely.

The price floor’s weak points: three conditions that could erode cost-driven stability

The floor holding up Asian ferro-alloy prices is not a fixed given. It is a conditional structure with specific failure modes, and knowing them is the difference between a clear watch list and a vague sense of unease.

Three conditions could erode the current stability:

1. Raw material cost retreat. Dalian coking coal futures already pulled back roughly 12% from their 1,729 RMB per tonne August peak to about 1,523.50 RMB per tonne, and full Shanxi supply normalisation is expected in H1 2027. If coke costs ease, the cost-side floor moderates with them.
2. Producer throughput pressure. When margins compress and volumes fall, as they did in India with a 23% year-on-year export drop in H1 2025, producers face a choice between pricing discipline and cash flow, and cash flow tends to win.
3. Demand weakness. The entire floor rests on producer costs, not end-user pull. If downstream steel demand deteriorates further, the pressure to accept below-cost pricing intensifies.

These operate on different clocks. Raw material retreat could begin within months if Shanxi recovers faster than expected. Throughput pressure is already visible in Indian behaviour. Demand weakness sits in the background, amplifying the other two if it deepens.

Why Chinese coke capacity cuts complicate the retreat scenario

The raw material retreat risk comes with an important qualifier. Even if Shanxi coking coal supply normalises in H1 2027, that does not automatically translate into equivalent coke supply.

Total Chinese coke capacity fell to 558 million tonnes in H1 2025, down 17 million tonnes year-on-year. On top of that, 27.55 million tonnes of outdated 4.3-metre coke ovens are targeted for full shutdown by 2026, a permanent reduction rather than a temporary one.

China’s steel industry work plan for 2025-2026, released by the Ministry of Industry and Information Technology, prohibits new coke production capacity and mandates ultra-low emission upgrades, embedding the structural capacity constraint into official industrial policy rather than treating it as a cyclical outcome.

What this tells you is that the coke cost floor baseline has structurally shifted compared with prior cycles. Coking coal supply recovering does not mean coke prices fall as far as they might have, because the refining capacity to turn that coal into coke has contracted. It is a partial offset to the retreat scenario, and it deserves weight in any forward view.

The coke price retreat scenario carries a specific mechanism worth tracking: import substitution and recovering Shanxi output arriving simultaneously could compress coke costs faster than the current consensus expects, which would directly moderate the cost floor under Chinese ferro-alloy prices.

What the Asian ferro-alloy supply map looks like heading into winter 2026-27

Pull the three regional threads together and a single picture emerges. Chinese coke and energy costs, Indian inventory depletion, and Southeast Asian hydropower each provide price support independently. None, on its own, is enough to push prices higher without an improvement in demand.

The winter shapes up as one of supply-side defensiveness rather than genuine tightening. Producers are protecting margins, not cutting capacity, so volumes are still flowing, just at prices that reflect elevated costs.

The headline variable to watch through winter is whether Chinese coke prices stabilise or retreat as Shanxi supply recovers. That input connects most directly to the largest ferro-alloy producing region.

The takeaway for positioning is direct. Current price stability is a holding pattern sustained by producer cost pain, not a signal of improving fundamentals. The prudent stance is to monitor the cost-side variables, particularly Chinese coke and the Ulanqab power wildcard, rather than read steady prices as recovery.

For anyone making calls in steelmaking input markets over the next two quarters, that distinction between cost-floor stability and demand-driven recovery is the most important one to get right. Reading stability as strength, or dismissing it as noise, both carry real cost.

Investors exploring how trade policy is reshaping global ferro-alloy flows alongside the cost pressures covered here will find our full explainer on ferro-alloy safeguard measures, which details how the European Commission’s 2025 framework is redirecting supply away from traditional export routes.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.

Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors.