Turkey's Steel Growth Is Real, but September Reveals Where the Cracks Are

Turkey's nine-month steel statistics, released by TCUD at the end of October and carried in English by SteelOrbis on 2 November, look reassuring at first pass: crude steel output up 6.1% to 29.77 Mt, finished steel production up 2.7% to 32.72 Mt, and the export/import cover ratio improving from 77.9% to 84.5%. The headline framing — "production increases, exports and imports decrease" — is accurate but obscures the more interesting story, which is a divergence between a strong first half, a deteriorating second half, and a trade structure that is shifting in ways TCUD itself is worried about.

The September signal is weaker than the year-to-date signal

September crude steel output of 3.21 Mt was up only 0.8% year on year, a sharp deceleration from the +6.1% nine-month run rate. That figure is independently confirmed by worldsteel's September release (3.2 Mt, +0.8%), which is the only number in the TCUD package cross-checked against a second primary statistical source. It is worth putting the arithmetic plainly: nine-month output of 29.77 Mt averages 3.31 Mt/month, so September ran about 3% below the year's average pace. The +6.1% headline is being carried by the first half.

The demand side is where the divergence is sharpest. TCUD's nine-month consumption figure of 29.5 Mt (+2.2%) reads as modest growth, but Turkish-language coverage of the same release (TRT, CNBC-e, 12punto) reported September finished steel consumption down 6.9% year on year to 2.9 Mt. If that is right, domestic demand effectively reversed in the third quarter while producers kept rolling. That combination — output holding up, home demand contracting — explains why TCUD is pressing on export mix and import defence simultaneously.

Slab up 16.6%, billet flat: the capacity story behind the headline

The product split of crude output is more informative than the total. Billet production was flat at 17.81 Mt while slab rose 16.6% to 11.96 Mt. The two reconcile exactly to the 29.77 Mt total, which is a useful consistency check. The message: Turkey's flat-product, integrated/concasting route is expanding, while the billet-fed EAF long-product segment — much of it serving construction and rebar exports — has plateaued.

That has implications beyond one year's data. A slab-heavy growth profile implies more domestic hot-rolling capacity chasing a flat-steel market that remains structurally short. But it also implies higher slab demand from the very mills that will need feedstock, which is precisely why Turkey's semis imports remain so large despite rising domestic slab output.

Semi-finished trade is the real anomaly

Semis are doing something unusual in both directions. In January–September Turkey imported 5.71 Mt of semi-finished steel (−7.9%) — 42.8% of total steel imports — while simultaneously exporting 749,284 t of semis, up 64.8%. Importing slab to feed flat mills while exporting billet or slab in volume is not a sign of a balanced industry; it is a sign of a sector whose rolling and casting capacity no longer matches its product mix and cost structure.

The export/import arithmetic is otherwise clean: 4.76 Mt flat + 6.08 Mt long + 0.75 Mt semis = 11.59 Mt, matching the reported export total. On the import side, 6.62 + 1.00 + 5.71 = 13.33 Mt against a reported 13.35 Mt — a 0.02 Mt residual, plausibly rounding or an unlisted category, and the only non-reconciling item in the release.

Unit values tell a better story than volumes

The most striking data in the release are the per-tonne prices, which the reported totals do not make explicit (all unit values below are my calculations from the disclosed volumes and values):

Two things stand out. First, September's export mix moved up-market — more value per tonne on 6.4% less volume. Second, imports collapsed by 22.3% in volume but their unit value rose about 12%: the cheap, low-value material that once flooded Turkey's long-product market has been squeezed out, leaving higher-grade flat and semis.

On the nine-month figures the pattern is less flattering: exports average ~$682/t against imports at ~$701/t. Turkey pays more per tonne for what it buys than it earns per tonne on what it sells. This is the empirical basis for TCUD's repeated call to shift the export mix toward higher value-added products, and it is a structural problem that one year of volume growth does not fix.

The EU problem is visible in the numbers, not in the release

What the SteelOrbis summary does not contain is the most consequential fact in the release, reported by CNBC-e and other Turkish outlets quoting TCUD Secretary General Veysel Yayan: Turkey's steel exports to the EU fell about 23% to roughly 3.7 Mt in January–September, a loss of around 1.1 Mt.

That is consistent with the new EU regime in force from 1 July 2026: an annual tariff-free quota of 18.3 Mt across steel product categories, with a 50% out-of-quota duty once quotas are exhausted. Turkey's country-specific hot-rolled coil allocation reportedly fell from roughly 1.6 Mt to 642,294 t per year, with total Turkish allocation across steel categories at about 2.86 Mt. Add CBAM verification requirements and the pressure on Turkish flat exports is now binding, not theoretical.

The corollary matters for readers outside Turkey: the 1.1 Mt Turkey lost in the EU did not disappear from global supply. It went looking for other markets — hence the 2.6% growth in long-product exports and the 64.8% surge in semis exports. Turkish mills are re-routing tonnage toward MENA, the Americas and Southeast Asia, and competing harder in those markets on price.

What to watch

Three things will determine whether 2026 ends as a growth year or a plateau year for Turkey.

1. Flat-steel import penetration. Imports supplied 44.1% of total steel consumption and 58.5% of flat steel consumption in January–September. Turkey is still structurally short of flat product, and until the slab and hot-rolling investments now underway convert into finished output, that dependency persists. The question is whether the import decline reflects genuine substitution or merely weaker demand.
2. Semis two-way flow. If slab imports and semis exports continue to rise together, Turkey is effectively using its rolling capacity as a tolling operation — importing intermediates and exporting partially processed steel at premium prices to customers outside the EU's quota system.
3. EU access post-CBAM. With Turkish HRC quota cut to roughly 642,000 t/year, the next round of trade data will show whether Turkish mills can recover EU volume through low-carbon certification or whether the 1.1 Mt loss becomes permanent.

TCUD's own prescriptions — value-added export mix, low-carbon investment, defence against dumped and subsidised imports — are correct but are essentially a restatement of the industry's structural problem. The nine-month data show Turkey producing more steel than a year ago, earning slightly more per tonne on exports, and yet still importing more than 13 Mt of steel, more than half of it flat product that its own mills cannot yet make competitively at scale. That is a more useful starting point than the headline growth rate.