Steel the Future
Steel · Markets
“One of the worst times in 20 years”: Why steel feels this bad in 2026
The shop-floor complaint is real. The official reports say the same thing — with one caveat: the pain is not evenly distributed.
“I only have 3 years in the steel industry, but it sucks real bad right now. The people that have been in the industry for 2 decades also say it’s one of the worst times they have ever seen.”
— A friend in the millsThat is not just venting. It matches what governments, trade groups, and mill earnings say — if you keep one distinction in mind. The global market is in a genuine overcapacity slump. The U.S. market is split. Some mills are making money behind a tariff wall. Others spent 2025 idling plants and cutting thousands of jobs.
The numbers behind the mood
In June 2026 the OECD published its Steel Outlook 2026 and did not mince words. It called the situation a deepening crisis. The core problem is simple: too many mills, not enough buyers.
Planned capacity additions through 2028 run as high as about 139 million tonnes. Demand barely moves. That is how you get low prices, thin or negative margins, and veterans saying they have seen this movie before. A readable summary of the OECD findings is here.
The World Steel Association’s April 2026 short-range outlook is a little less dire on demand, but it is still a slog: global steel demand up only 0.3% in 2026, then 2.2% in 2027. China is still contracting. India is the main bright spot. worldsteel press release.
China is still the weight on the market
China consumes roughly half the world’s steel. When its property sector cracked, mills did not cut capacity in line with lost domestic demand. They exported.
Chinese crude steel output in 2025 was well below the 2020 peak. Demand fell sharply. Exports did the opposite: Chinese mills shipped a record ~131 million tonnes in 2025. That is more steel than the entire EU produces in a year, sent onto markets that were already soft.
Bloomberg’s June 2026 read is useful: not a cliff collapse, but a long plateau after the property crash, with manufacturing and exports only partly offsetting the hole. Bloomberg story.
The OECD also flagged subsidies. It said the median subsidization rate for Chinese steel firms had roughly doubled since 2019 and was many times higher than for OECD producers. That is why mills in Europe, Korea, Japan, Russia, Thailand and elsewhere keep talking about crisis, closures, and utilization rates that would not keep a plant alive without protection or cheap energy.
EUROFER’s mid-2026 outlook is the European version of the same story: modest demand improvement, production still near record lows, consumption still millions of tonnes below pre-pandemic levels. EUROFER outlook.
The U.S. is a different market — and even there, it depends where you work
Section 232 tariffs, raised to 50% in 2025 and extended to more downstream products, changed the U.S. math. Imports dropped sharply. Domestic prices stayed high. Hot-rolled coil in the U.S. has been trading well above $1,000 a short ton for much of 2026, while export markets elsewhere sit far lower.
That is why two true statements can exist at once:
Some U.S. producers are doing fine
Nucor’s leadership has told Congress the American industry is performing well, import share is the lowest many executives have seen in decades, and the company has poured billions into new capacity. Data-center construction has helped certain product lines. Steel Market Update.
Some U.S. producers had a brutal 2025
Cleveland-Cliffs posted a $1.4 billion net loss, cut about 3,500 jobs, and idled or closed multiple operations after weak auto demand and messy trade fallout with Canada. Management now talks recovery in 2026. Manufacturing Dive.
U.S. Steel, now under Nippon Steel, also had a rough 2025 before results improved as furnaces restarted and the tariff-supported price level held.
If your friend works at an integrated mill, in auto-exposed flat roll, rail, plate, or tubular, three years in the industry can easily feel like a dead end. If they work at a well-positioned mini-mill selling into construction and data centers, the same year can look like a boom. That split is why shop-floor talk and CEO testimony sometimes sound like two different industries.
Fastmarkets’ recap of 2025 in the Americas is a good blow-by-blow of how tariffs, auto weakness, mill idlings, and price swings piled up. 2025 timeline.
Is it really one of the worst stretches in 20 years?
Compare it to the last two reference points.
2008–09 was a demand crash. Orders vanished almost overnight with the financial crisis.
2015–16 was the last big China overcapacity crisis. Global excess capacity peaked near 750 million tonnes. Prices collapsed. Plants closed. Employment in steel took a beating. That is the episode the OECD is now measuring against.
2025–26 looks more like 2015–16 than 2008–09: not a sudden freeze, but a long grind of too much steel chasing too little demand, made worse by subsidies and export surges. Some analysts argue this episode has already lasted longer than the mid-2010s slump because Chinese domestic demand never really rebounded and new capacity kept appearing in other countries.
Tariffs did not solve the overcapacity. They walled off one rich market. That is why U.S. HRC can sit near multi-year highs while mills in unprotected markets talk about utilization below 30% and plants shutting down.
What to watch next
- Does China actually cut capacity, or only cut output for a few weeks? Output cuts without permanent closures just reload the next export wave.
- Auto production in North America. Integrated mills are betting on a rebound. If it doesn’t arrive, the U.S. split stays ugly.
- How leaky the tariff wall gets. Downstream products and third-country routing have already been a fight.
- India and infrastructure. That is where demand is still growing in a meaningful way.
- Energy and interest rates. High power costs in Europe and high rates in places like Russia have been as damaging as cheap imports.
None of that is a reason to tell your friend they’re wrong. For a lot of people in this industry, 2025–26 has been one of the worst stretches in a generation. The data says the global market is structurally sick. The U.S. exception is real, but it is a policy exception, not proof that steel demand suddenly got healthy.
Further reading
- OECD Steel Outlook 2026
The main official report. PDF is linked from that page. - OECD summary: excess capacity heading for 745 Mt
Shorter write-up of the same outlook. - worldsteel Short Range Outlook, April 2026
Demand forecast for 2026–27 by region. - Bloomberg: China’s steel market after the property crash
Why Chinese demand flattened instead of collapsing overnight. - EUROFER 2026–27 steel market outlook
The European industry’s own read of a still-weak market. - Cleveland-Cliffs after a loss-making 2025
Layoffs, idled mills, and the 2026 recovery pitch. - U.S. mills on tariffs and investment
The protected-market side of the story. - Fastmarkets: 2025 steel trade timeline in the Americas
Month-by-month on tariffs, auto weakness, and mill cuts.
Sources cited above are from mid-2025 through September 2026 industry reports and earnings coverage. Conditions vary by mill, product, and country. This is an industry explainer, not investment advice.
Comments
Post a Comment