Scrap Market Review — August 2026
Market analysis from TDC Ventures. Not trading advice.
The month in one line
Turkish mills raised the price of what they sell and left the price of what they buy alone — and by the close of the month the two destinations that matter to European sellers were no longer the ones the market quotes.
Our track record this month
We publish a scored record of every forecast we make. The full, unedited version — ranges, actuals, verdicts, errors — is at Forecast Track Record.
August was our worst month, and the honest number is the point of publishing one.
- Resolved this month: four scored calls — two in range, two missed. A 50% hit rate, against 80% on our published-benchmark record to date.
- August mean absolute error: $7.30/t, against $3.09/t across the whole book. One call did most of that damage.
- Three new forecasts written, all still open and targeting early September.
- Separately, our daily directional calls ran 7 clear hits and 18 partials from 25 scored at full horizon — no outright misses, which is a different and gentler measure: direction over days, not levels over weeks.
The book as a whole now stands at 8 of 10 in range on a published-benchmark basis (80%), mean absolute error $3.09/t, median $1.78/t, direction accuracy 60%. Ten forecasts resolved, three open.
Calls that landed
Turkish consolidation, called a month ahead. Our July review said Turkish shredded would consolidate near $390–400 rather than break out or give back its freight premium. Through August the Kallanish assessment traded $392.00–$393.50 — inside that band every session. No breakout, no collapse. The Fastmarkets leg sat marginally below it at $387.50–$388.73, which is the honest asterisk on an otherwise clean call.
The Pakistan index leg, resolved early in the month. Our 30 July forecast put the Port Qasim shredded index at a central of $413 with a $406–420 range. It printed $414.21 — an error of +$1.21/t, the second-tightest call on our book.
The margin read, published as it happened. Through the last week of August we said the Turkish rally was mill margin, not raw-material cost. Rebar for export ran +$22.50 (+3.93%) to $590–600/t and billet +$10.00 (+1.87%) while every Turkish scrap leg printed 0.00%, widening the gap between finished steel and its feed to about $218.77/t. Nothing in the scrap tape has since contradicted it.
Where we missed — and what it taught us
The worst call on our record resolved this month, at $18.00/t.
Our 30 July Pakistan shred forecast carried two declared bases. The index leg hit. The transaction leg set a central of $388 with a range of $383–396, and observed transaction levels reached around $406/t CFR Port Qasim on 19 August — ten dollars above the top of our range.
We had named the risk ourselves. Our primary upside case was a mill-restocking wave, and we sized it at +$5–10/t. It delivered roughly double that. Getting the mechanism right and the magnitude wrong is still wrong.
The lesson we adopted, and have since written into our method: a transaction-basis central is only as fresh as the transaction series beneath it. Ours was anchored on observed levels from earlier in the summer while the published index had moved on. Where a benchmark and a transaction series disagree, the benchmark is the one with a refresh schedule. We now forecast dark markets as a spread to a live anchor rather than off a stale transaction level — adopted as standing policy on 18 August.
There is a second, smaller miss worth naming. Our July review said the Pakistan index would plateau, not extend. It held $413–417 for three and a half weeks — then printed $419.96 on 28 August, a fresh high for the series. Right for most of the month, wrong at the close.
The month’s market arc
August’s defining feature was a disconnection between finished steel and its feed. Turkish mills lifted export rebar nearly 4% while the scrap they melt moved a dollar in either direction depending on which assessment you read. That is margin expansion at the mill, not cost pressure from the yard — and it is the single most useful distinction a scrap seller could have drawn last month.
The Turkish scrap chain went almost nowhere, twice. It round-tripped through the middle of the month and finished at $376.23 on the US-origin index — within a dollar of where it spent most of August. European flats, meanwhile, re-rated hard: Spanish hot-rolled coil +3.90%, southern European plate +3.57% in a single week.
Pakistan broke late and for the right reason. After four static sessions the Port Qasim shredded index jumped +$6.33 (+1.53%) to $419.96 on 28 August — a fresh high for the dataset. The stated cause was supply, not demand: tighter collection and container availability lifting offers into steady buying. A supply-led rise does not need buyer conviction to hold, which makes it a different animal from a demand rally.
And the destination hierarchy inverted. On matched 28 August prints, shredded into Nhava Sheva ($397.28) was paying $8.55 more than shredded into Turkey ($388.73) — despite longer freight from every European origin. The premium of Port Qasim over Nhava Sheva compressed to $22.68, its narrowest in our series. If Turkey is no longer the best bid, every origin negotiation anchored on “the Turkish number” is anchored on the wrong one.
Iron ore did not lead any of it. The 61% Fe Qingdao assessment finished at $97.79, again failing to transmit into scrap for a fifth week. In non-ferrous, copper’s prompt premium collapsed from $535 to $56 in four sessions before re-widening to $254 — a structure moving violently in both directions on a market that barely moved.
Structural signals we’re tracking
Two export regimes are now squeezing scrap from opposite sides, and they are three years apart. On 27 August a US allocation order took effect covering black mass and tungsten waste and scrap: 100% of covered monthly sales must go to US persons, and covered material must remain physically in the United States, with a waiver path. It runs to 27 August 2027. Read it precisely: an allocation order with an exception process, not an export ban — no shipping document cures it, because it is a seller-side accounting obligation.
Pulling the other way, the EU Waste Shipment Regulation restricts exports to non-OECD destinations from 21 May 2027, with the approved-country list due 21 November 2026. US material is being pulled inward now; EU material faces a gate in under two years. Origin economics that look settled today have a dated expiry, and the November list is the first hard checkpoint.
The third signal is the one August produced: destination hierarchy is not fixed. A market that has priced off Turkey for years just spent a month with India bidding higher. Whether it persists is open — the assumption that it could not happen is not.
What we’re watching next month
Three specific calls for September. These are what next month’s review gets scored against.
- Turkish shredded has held a $387.50–393.25 envelope for four weeks. We expect it to stay inside it, and the trigger that would break it is a European or Baltic cargo reported above $380 CFR — nothing has printed above that line all episode.
- Pakistan at a fresh high on supply, not demand. Our open forecast is $408–418 for the 4 September print. The 28 August print at $419.96 is already above that range — we are publishing the call as written rather than moving it, and we expect to be scored on it.
- Does India stay above Turkey on matched-date shredded prints? It is +$8.55 today. Convergence within a dollar, or a reversal, would tell us August was an artefact rather than a shift.
How we forecast
We anchor to published benchmarks and observed transaction levels, cross-check every call against leading indicators — freight, iron ore, finished-steel margin — and declare the resolution basis in writing before the target print exists, so a call cannot be quietly re-scored against whichever number happens to flatter it. Every forecast is then scored against the first print on or after its target date, hit or miss, and published. The misses stay up.