At around 8:16am on September 4, hot-blast stove No. 3 at blast furnace No. 1 of Formosa Ha Tinh Steel Corporation (FHS) suffered an equipment failure that sparked a small fire nearby, extinguished within 14 minutes. FHS reported no casualties, though local authorities noted one worker was injured.Nhân DânBáo Xây dựng Blast furnace No. 1 was halted for inspection, furnace No. 2 kept running normally, and FHS committed to a fix within 48 hours.Giáo dục Thời đại
Looking at the numbers, the market had a full trading session to price this news, and the reaction was close to nothing, not because the news was overlooked but because the real scale of the disruption was much smaller than the headline suggested.
A full session to price it in, and almost no reaction
The incident happened at 8:16am, 44 minutes before the market opened on September 4. HPG closed at VND 21,700, up 0.46% (DB: HPG). NKG fell 0.46% to VND 10,750 (DB: NKG), HSG fell 0.47% to VND 10,600 (DB: HSG), and GDA held flat at VND 12,300 (DB: GDA). The same session, the VN-Index rose 1.39%, or 25.36 points, to close at 1,853.08 (DB: VN-INDEX).
In other words, steel stocks got no premium for a supply-disruption headline, and actually lagged the broader market. If investors had treated this as a serious event, the price reaction should have been clearer in both directions, favoring whichever names benefit or lose from a competitor's supply gap.
One detail is easy to misread on timing: the sharp decline in steel stocks actually happened on September 3, a day before the incident. HPG lost 2.26% (DB: HPG), HSG lost 2.29% (DB: HSG), SMC lost 2.91% (DB: SMC). Pairing the September 3 selloff with the September 4 fire into one causal story misattributes the cause, simply because the timeline doesn't allow it.
How much production was actually lost
FHS is Vietnam's largest integrated steel complex, with two blast furnaces and a combined design capacity of 7 million tonnes of crude steel a year, of which 3-3.5 million tonnes is hot-rolled coil (HRC).Người Quan Sát But design capacity and actual output are two different things. In the first half of 2026, FHS sold more than 3.1 million tonnes of steel for USD 1.7 billion in revenue, roughly 70% of full-year design capacity.Người Quan Sát
An FPTS report dated July 23, 2026 put nationwide crude steel output at 15.18 million tonnes for the half-year, with Hoa Phat leading at 46% market share and Formosa second at 20.5%.FPTS In HRC specifically, these two companies are Vietnam's only producers, and the ranking has flipped from a few years ago: Hoa Phat held more than 60% of HRC share by the end of 2025, after Formosa had led in the early 2022 period.Người Quan Sát
From there, the lost output can be estimated. If blast furnace No. 1 accounts for roughly half of FHS's capacity, about 3.5 million tonnes a year, then 48 hours offline equals roughly 19,000 tonnes of crude steel at design rates. That figure carries assumptions and is likely an overstatement: per Vietstock, each blast furnace runs four hot-blast stoves, three active and one on standby, so furnace No. 1 could still run at around 80% capacity if the repair drags on.Vietstock Set against the industry as a whole, 19,000 tonnes is less than a quarter of the country's daily crude steel output, based on the 15.18 million tonne half-year figure.
The nearest precedent: Hoa Phat Dung Quat, 2025
In May 2025, Hoa Phat's Dung Quat complex suffered a molten iron leak and had to halt blast furnace No. 1, which produces 1.5 million tonnes of construction steel a year. SSI estimated at the time that output would fall by around 60,000 tonnes over a 1-2 week repair, and judged the overall impact immaterial. HPG shares fell 2.4% in the September 12, 2025 session, then recovered.MarketTimes
Set side by side: an incident with roughly three times the lost output and five to seven times the repair time still cost the affected stock less than 3% in a single session. That is the reasonable expectation baseline for the September 7 trading session.
Three signals worth watching this week
The 48-hour window FHS set expires Saturday morning, September 6. Because this is a deadline the company set itself, whether furnace No. 1 restarts on schedule is a direct test of the credibility of its own guidance. If that date passes with no confirmation of restart, the working framework should shift from "single-session incident" to "supply disruption with an open-ended timeline."
FHS's steel price list is another signal worth tracking. On September 4, the company paused plans to issue a new price quote.Người Quan Sát Reopening that price list typically comes before any official statement, making it the fastest read on whether the production line is back to normal.
The point most likely to get flattened into one narrative is that the impact cuts differently across tickers. Hoa Phat sells HRC, so a competitor's supply shortfall is a favorable pricing condition. Nam Kim, Hoa Sen, and Ton Dong A, by contrast, buy HRC as an input for coated steel, so a bump in that price becomes an added cost. Same headline, two groups sitting on opposite sides of the price ledger.
Price backdrop and the risk in this framework
Global HRC prices are trending up: USD 1,230.06 a tonne on September 4, roughly 3.4% higher than a week earlier (DB: HRC_STEEL). But domestic prices run on each mill's own quote sheet and input costs, and don't track the global price chain closely. One clear sign of that: domestic D10 rebar has held flat at VND 14,210/kg since late July (DB: VN_STEEL_D10).
The biggest risk to this framework sits in the 48-hour figure itself. Per preliminary market information cited by Người Quan Sát, the incident is being assessed as relatively serious, with repairs potentially taking 2-3 months, though the same source notes this requires official confirmation from the plant.Người Quan Sát The two timelines differ by an order of magnitude, and only one of them comes from an official company statement.
It's also worth acknowledging that the flat reaction in steel stocks on September 4 may not entirely reflect the market correctly sizing the incident. At least two other explanations coexist: money flow that session concentrated in large-cap names driving the index, leaving steel names on the sidelines; and steel stocks had already released some selling pressure in the September 3 decline. The available data isn't enough to cleanly separate the contribution of each factor.
Conclusion
Based on data through the evening of September 4, the reasonable default framework treats this as a small-scale event: blast furnace No. 2 is still running, the lost output under the 48-hour scenario is less than a quarter of the country's daily crude steel production, and the Hoa Phat Dung Quat precedent shows a much larger incident produced less than 3% of price volatility in a single session.
That framework only changes on one specific signal: the morning of September 6 passing without FHS confirming furnace No. 1 has restarted. At that point the story shifts from a single technical incident to an open-ended supply gap, and Hoa Phat and the coated-steel names would genuinely sit on opposite sides of the same pricing story. FHS's price list over the coming week is the most useful signal to watch before any official confirmation arrives.