Italy's combined flat and long steel imports fell 3.08% year on year to 6.62 million tonnes in January–August 2026, with the value of those flows down only 1.64% to €4.82 billion, according to a SteelOrbis analysis of Eurostat data published on 20 November. On its own, that is a routine statistical release. What makes it worth reading is the arithmetic underneath: volumes fell about twice as fast as value, and the supplier mix rotated sharply toward intra-EU origins.
A modest contraction with firm unit values
| Segment | Jan–Aug 2025 | Jan–Aug 2026 | Volume change | Value change | Avg. value/t |
|---|---|---|---|---|---|
| Flat | 5.794 mn t | 5.692 mn t | −1.76% | −0.48% | €703 → €712 (+1.3%) |
| Long | 1.031 mn t | 0.923 mn t | −10.50% | −7.33% | €802 → €830 (+3.54%) |
| Total | ~6.83 mn t | 6.62 mn t | −3.08% | −1.64% | ~€718 → ~€729* |
*Total unit values are my calculation (€4.90bn/6.83mn t and €4.82bn/6.62mn t); the source reports only the segment-level averages. The implied 2025 totals reconcile exactly with the reported volumes and values, so the dataset is internally consistent.
Two things follow. First, the decline in the import bill is almost entirely a volume effect — roughly 1.5% of unit-value uplift offset a third of the tonnage lost. Italy is not being flooded with cheap steel; if anything, the average tonne crossing its border got more expensive. Second, the divergence between the two segments is large enough to be a signal rather than noise: longs contracted six times faster than flats in volume terms, yet their unit value rose nearly three times as fast.
That combination is not what a demand-driven downturn looks like. A weak Italian construction market would normally pull in cheaper, lower-value long products from third-country mills to fill the gap — compressing, not lifting, average long prices. The observed pattern is more consistent with a supply-side reallocation: fewer, higher-cost tonnes arriving, with the marginal low-price supplier priced out. My reading is that this reflects the interaction of EU safeguard quota economics and the cost of servicing the Italian market from distant origins, rather than a collapse in Italian apparent consumption. The source does not attribute causes, and this is analysis, not reported fact.
Germany absorbed the entire flat-product contraction
The flat-supplier league table is where the report earns its keep.
| Flat supplier | Jan–Aug 2025 (t) | Jan–Aug 2026 (t) | Change |
|---|---|---|---|
| Germany | 587,048 | 869,530 | +48.1% |
| France | 419,679 | 584,965 | +39.4% |
| Turkey | 699,879 | 574,936 | −17.9% |
| South Korea | 610,702 | 445,006 | −27.1% |
| India | 261,734 | 377,846 | +44.4% |
Germany's increase alone (+282,482 t) is nearly three times the entire net fall in Italian flat imports (about 102,000 t). Put differently: strip out the German gain and Italian flat imports would have risen by roughly 180,000 t. France adds another +165,000 t. The top five suppliers accounted for about 50% of flat volumes in 2026 versus 44.5% in 2025 — a modest but real concentration increase, driven almost entirely by EU origins.
The third-country losses are equally stark: Turkey −125,000 t, South Korea −166,000 t. Those two alone gave up roughly 290,000 t, more than the combined German and French gains. The obvious inference — which the data support but do not prove — is substitution. When Turkish and Korean tonnage becomes relatively less attractive, whether through safeguard quota exhaustion, anti-dumping exposure or freight and lead-time economics, the nearest alternative for an Italian service centre or automotive Tier 1 is a German or French mill across the Alps.
India is the exception that complicates the "fortress Europe" narrative: +44.4% to 377,846 t makes it the only non-EU origin in the top five to grow, and it now sits within 70,000 t of Turkey. Indian flat supply into Italy is small in absolute terms but is clearly finding a niche — plausibly in commodity-grade hot-rolled coil and coated products where quota availability and price still clear the barrier.
Longs: China's retreat leads the contraction
| Long supplier | Jan–Aug 2025 (t) | Jan–Aug 2026 (t) | Change |
|---|---|---|---|
| Spain | 225,932 | 203,669 | −9.9% |
| Germany | 150,301 | 157,904 | +5.1% |
| Czech Republic | 80,016 | 82,171 | +2.7% |
| China | 103,031 | 76,630 | −25.6% |
| France | 61,527 | 67,665 | +10.0% |
China's absolute loss (−26,401 t) is the single largest contribution to the long-product contraction, marginally ahead of Spain's (−22,263 t). The long segment is structurally different from flats: it is small (0.92 mn t), overwhelmingly intra-EU at the top of the table, and the four EU suppliers shown all held or gained share. The top five account for roughly 64% of long volumes.
There is a technical point here that matters for anyone modelling Italian trade. The long-product universe in this dataset is narrow — HS 7213–7217 covers bars, rods, angles, sections and wire rod, but excludes seamless and welded tube (HS 7304–7306) and railway material, both of which are material long-product import categories for Italy. So the −10.5% should be read as a statement about merchant bar, wire rod and sections, not about long products in the widest sense. The flat basket (HS 7208–7212) is more complete but still omits coated downstream products under HS 7210 in part and, more importantly, excludes stainless (HS 7219–7220), where Italy is a significant net importer.
Why the September preliminary figures should not be over-read
The report carries preliminary September non-EU data: 114,513 t of flats (−12.0% month on month) and 14,279 t of longs (+8.8%). Two cautions apply.
First, these cover third-country suppliers only, because intra-EU partner data had not been disclosed at the time of publication. Since Germany and France together supplied about 1.45 mn t of the 5.69 mn t flat total in January–August, over a quarter of Italian flat imports are structurally invisible in the September number. The monthly run-rate implied by the eight-month total is roughly 712,000 t of flats; 114,513 t of non-EU flats is a different animal entirely and cannot be compared with it.
Second, month-on-month comparisons on non-EU-only flows are noisy at these volumes. A 12% fall in non-EU flat arrivals is around 15,000 t — within the range of a single late-arriving vessel or a quota-cycle effect at the start of a new safeguard period. The more useful reading is directional: third-country flat pressure on Italy eased into September while third-country long arrivals ticked up, which is consistent with the year-to-date mix shift rather than a reversal of it.
What to watch
The January–July release from the same outlet showed imports down 7.8%; the eight-month figure is −3.08%. That is a meaningful deceleration in the rate of decline, driven by a stronger August. If that momentum persists, the full-year 2026 total could land close to flat against 2025 — a very different outcome from the double-digit contractions reported in parts of 2024–25.
The questions that will decide it are structural rather than cyclical. Whether Germany and France can sustain 40–50% growth rates in flat supply to Italy once their own order books tighten; whether India's growth is a durable niche or a quota-arbitrage window; and whether the Chinese long-product retreat continues once EU safeguard measures on the relevant product categories are reviewed. For Italian re-rollers and service centres, the practical implication of the first eight months is that the import competition they face has become more European and less Asian — which changes the freight, lead-time and currency calculus of competing against it, even if the headline tonnage is lower.
Data note: All volume, value and unit-value figures are as reported by SteelOrbis from Eurostat data, covering HS 7208–7212 (flat) and HS 7213–7217 (long). Total-segment unit values, supplier-share calculations and the counterfactual on German flat supply are my own calculations from the reported figures. Cause-and-effect interpretations are analytical and are not attributed to the source.
Source & methodology +
MetalMate separates original-source evidence from summary and analysis. Verify the original source before procurement, legal or engineering decisions.
- Article status
- Published · Version 1
- Source date
- 20 November 2026
- MetalMate publication
- 20 November 2026
- Verification
- Automated evidence checks; not human reviewed
