Japan's steel exports fell 6% in Jan–Sep 2026, but the headline hides a rotation: HRC collapse, US surge, semis retreat

Japan's January–September 2026 iron and steel product exports came in at 21.50 million mt, down 6.0% year on year, with September alone at 2.32 million mt — 6.0% below August and 7.1% below September 2025, according to MOF customs data compiled by JISF and reported by SteelOrbis. The headline number is unremarkable in isolation (2025 was already down 4.2%); what matters is where the tonnage went and which products disappeared.

The decline is decelerating, but September broke the pattern

SteelOrbis's own monthly series shows the year-on-year decline peaking early and then narrowing steadily:

Source: SteelOrbis monthly headlines. My interpretation: the trend confirms a very weak first half followed by stabilisation around an annualised ~28.5–29 million mt run-rate, well below Japan's historic 40 million mt-plus peaks.

The Jan–Sep figure ticked 0.2 points worse than Jan–Aug because September itself fell 7.1% year on year — the weakest single month since spring. The improvement through mid-year therefore looks less like demand recovery and more like the base effect of a disastrous Q1 fading out.

HRC is the whole story — and semis are the second act

Of the ~1.37 million mt implied year-on-year decline in the nine-month total (calculation from published percentages), HRC alone lost roughly 863,000 mt — about 60% of the shortfall. Semis (−15.0%) and CRC (−7.9%) account for most of the rest. Meanwhile galvanized sheet grew 3.2% and bars jumped 39% on a small base.

Note: the listed lines total only ~67% of the reported 21.50 million mt, so the table is a selected extract, not a full breakdown. Figures as reported; sums and shares are my calculations.

The pattern is analytically important. The weakness sits in upstream, high-volume carbon flat products and semis — exactly the lines that carry integrated mills' blast-furnace utilisation — while downstream coating and niche long products grow. September's 38.2% month-on-month collapse in semis shipments points to weaker intra-Asian slab and billet flows to ASEAN and Korean re-rollers, not to Japanese finished-steel demand.

Destination rotation: US up 23%, China and Taiwan down double digits

*Shares calculated by the author; source gives volumes and percentages only.

The US is the only major market with strong growth, and it is now roughly two-thirds the size of Japan's China book. That is a meaningful reversal for a trade relationship long constrained by Section 232. Japanese mills are evidently finding workable routes into the US — likely a mix of agreement-based or quota-type arrangements, downstream/automotive supply and US mill demand — though the source gives no mechanism, so this driver remains inference, not fact.

At the same time, double-digit falls to China (−13.8%) and Taiwan (−14.1%) reflect both Chinese demand weakness and Chinese export competition squeezing Japanese material out of Asia. Korea (+7.7%), Japan's most reliably growing near-shore market, is the partial offset.

What this means commercially

• Product mix, not just volume, is under pressure. A 10% fall in HRC plus 15% in semis hurts integrated mill economics far more than a 3% rise in coated sheet helps them. Japanese producers are effectively being pushed down the value chain by Asian price competition.
• Export dependence is shrinking in importance, but not painlessly. At ~28.5–29 million mt annualised, exports are running well below the 2020–2024 norm, concentrating mills' exposure on a soft domestic market.
• The US is becoming a strategic outlet. A 23% rise on a 1.0 million mt base is small in absolute terms but signals a durable redirection of Japanese supply away from depressed Asian markets.
• Semis deserve watching. The −38% month-on-month September reading, if sustained, would signal a broader contraction in Asian re-rolling activity that affects billet and slab pricing region-wide.

Caveats

These are MOF customs product weights — not mill shipment statistics — and the JISF/MOF originals could not be retrieved for this analysis, so figures are single-sourced to SteelOrbis. The destination list may be a selection rather than the full top-market set. The interpretation of US growth drivers is analytical, not reported. Readers should verify against the JISF English statistics release (next scheduled 30 November) before building forecasts on the product-level detail.