SCFI extends winning streak to six weeks, hits 3,590.05 as transpacific outruns Europe

The Analysis

3,590.05. That is the SCFI print dated 4 September 2026, up 80.52 points, plus 2.3% on the week, and the sixth consecutive weekly gain. Every headline I saw this week led with the streak. The streak is the least useful number in the release.

Underneath it, the same release gives three lanes pointing in three different directions: Far East to US West at 7,242 dollars per FEU, up 4.4%; Far East to US East at 10,324 dollars per FEU, up 2.8%; Far East to North Europe at 2,643 dollars per TEU, down 2.7%. Two units, three lanes, one composite. If you are still budgeting off the composite, you are budgeting off a number that no ship ever charged anybody.

So let me take it apart the way I take apart every rate notice: what is the base, what is the unit, and what does one box actually cost.

The units have to come first, because this is where a cost sheet goes wrong before it ever reaches the money. US West and US East are quoted per FEU, the forty-foot box. North Europe is quoted per TEU, the twenty-foot equivalent. Three figures sit side by side in one release and they are not the same kind of figure. Set 10,324 against 2,643 and you get a ratio of 3.9. Halve the forty-foot rate to put it on a twenty-foot basis and you get 5,162 against 2,643, a ratio of 1.95. The ratio that gets forwarded around the office is roughly double the one that survives normalization. Halving is a crude conversion and I will say so plainly: a forty-foot box has never cost exactly two twenty-foot boxes, which is precisely why the industry prices it at a discount. Use the halved number for scale, never for settlement.

The composite has the same problem at a different altitude. Work out the dollar move on each lane. US West added 305 dollars per FEU, because 7,242 divided by 1.044 is 6,937. US East added 281, because 10,324 divided by 1.028 is 10,043. North Europe gave back 73, because 2,643 divided by 0.973 is 2,716. Those three sum to 513 dollars. The composite moved 80.52 points. The two numbers are unrelated by construction. SCFI is a weighted index on a fixed base, not a dollar average of three lanes. 80.52 points is not 80.52 dollars, and the sentence that the market added eighty dollars a box this week is not a simplification of the release, it is a different claim altogether.

Capacity explains the split, and capacity follows the per-TEU number, not the composite. On a twenty-foot basis the same release prints 3,621 for US West, 5,162 for US East and 2,643 for North Europe. The gap between US West and Europe is 978 dollars per TEU. Between US East and Europe it is 2,519. No scheduler needs a meeting to decide where the next hull goes. Europe is not weak because European demand collapsed in the week to 4 September. It is weak because the ships that could serve it are being paid better somewhere else, and the lane that loses the hull prints a lower rate and calls it a soft market. The two-speed market in the headline is a scheduling outcome before it is a demand outcome, which matters to you because scheduling reverses faster than demand does. A carrier can pull a string of sailings off a lane inside a month. A demand recovery takes two quarters to show up in a booking book.

Put the numbers on weight, because weight is the version your warehouse understands. Assume a forty-foot box carries 20,000 kilos of chargeable cargo, which is my assumption and not a figure from the source. US West works out at 0.362 dollars per kilo, US East at 0.516. Assume a twenty-foot box carries 17,500 kilos and North Europe works out at 0.151. The East Coast is three and a half times the European lane per kilo. Now put the same numbers on cargo value instead, because weight is only half the story and the other half decides who actually hurts.

Assume a 60,000 dollar FOB value per forty-foot box: freight is 12.1% of value to the West Coast and 17.2% to the East. Assume 30,000 dollars per twenty-foot box to Europe: freight is 8.8% of value. A furniture importer and a semiconductor equipment importer are reading the same index and living in two different economies. One of them has freight eating a sixth of the invoice. The other treats it as a rounding line.

The transmission window is set by your booking date, not by the sailing date and not by the Friday print. SCFI goes out for the week just booked; the container that leaves Shanghai in three weeks was priced whenever you lifted the rate, which for FAK cargo is the moment the booking confirms and for contract cargo is the moment the contract says. Three clocks run at three speeds. Spot books reprice the same week. Contract books reprice at the next service contract review or at the MQC shortfall, typically thirty to ninety days out. Surcharge changes arrive with a notice period the carrier sets, and that notice is the only part of the sequence you can plan around with any confidence. If you are on a contract and you have not read the surcharge clause this quarter, you do not know your exposure. You know your base rate, which is a different document.

Here is the number that got no coverage this week. The East Coast premium over the West Coast is 3,082 dollars per FEU. Last week it was 3,106. It narrowed by 24 dollars, and it did that while the West Coast rose 4.4% and the East rose 2.8% in the same seven days. Run the differential forward at those two paces and the premium closes in about 23 weeks: 7,242 times 1.044 to the n equals 10,324 times 1.028 to the n gives n of 22.9. That is arithmetic on the source numbers, not a forecast from the index, and it says something specific.

The market is telling you the West Coast is getting expensive faster than the East, but not fast enough to justify re-routing a fourth-quarter program. Anyone who moves an Asia-to-Chicago program from a West Coast port to an all-water East Coast routing on the strength of one 4.4% print is paying a switching cost, a longer transit and a new set of inland connections to chase a 24-dollar-a-week convergence. Re-route when the premium is inside two hundred dollars a box. Not when a percentage looks dramatic.

Do the monthly arithmetic on a mid-sized book, because percentages do not survive contact with a volume plan. Assume forty FEU a month to the West Coast and twenty to the East, which is my assumption and not a figure from the source. Last week's move alone costs 12,200 dollars on the West Coast book, 305 times 40, and 5,620 on the East Coast book, 281 times 20. That is 17,820 dollars of additional freight in one week, on volume you had already committed before the print came out. Carried at the same weekly pace for a quarter it is 231,660 dollars. I would not carry it that far. Six straight weeks is a fact. Seven is a hope.

The asymmetry is the part that gets lost in the meeting. At a 7,242 base, a 4.4% week is worth 319 dollars a box and a 2.7% week is worth 196. In dollars, the upside observed so far has been bigger than the downside, because the base is high and because the up-moves printed in this run have been larger than the down-moves. That is the entire argument for locking a share of the book rather than waiting for the turn. Wait and be right, you save 196 a box. Wait and be wrong, you pay 319. You need to be right better than three weeks in five just to break even on the wait, and I have not met a shipper who is right three weeks in five on timing. The people who are right on timing in this market are usually the ones who sold you the space.

The expensive mistake in a week like this is treating the index as your invoice. Your invoice carries the base rate plus whatever the surcharge schedule says plus whatever the booking was actually confirmed at. The index is a basket of claimed rates for one week, not a price you can buy at. Reconcile four consecutive prints against four consecutive invoices before you put either number in a budget; if the variance runs above three percent, the problem is in your cost model, not in the market. My line: if US West prints above 7,600 per FEU while North Europe stays under 2,700 per TEU, lock sixty percent of fourth-quarter US-bound volume inside that week. Above that line you are not buying freight, you are buying lottery tickets with your own margin. Six weeks of gains is a trend. The seventh week is when the people who did not lock start calling it a crisis.

↑ Back to the key points

— By Vivian Zhao