Supply Chain Action Points
Read this first — the conclusion, and the moves to make:
- Within 7 days, rebuild the September budget lane by lane using the lane change rather than the composite: East Coast at plus 3.6% a week on a 10,000 base is USD 347 per FEU per week, West Coast at plus 2.6% on 6,940 is USD 176. Replace any 2.9% assumption in the model.
- Compute freight as a share of goods value before deciding anything: divide the lane rate by your actual payload per FEU. On a 20-tonne payload the West Coast is USD 0.347 per kilogram and the East Coast USD 0.50. Any product where freight exceeds 12% of FOB value goes to the top of the routing review.
- Lock 60% to 70% of September East Coast volume by 5 September and leave the balance on spot. At a 3.6% weekly pace the arithmetic puts the lane near USD 11,520 by the end of September, so the cost of waiting one week on 40 FEU is about USD 13,900.
- Re-run the West Coast plus inland comparison against a hard ceiling of USD 3,060 per FEU, which is today's East Coast minus West Coast spread. If the inland move and the extra transit days come in under that, switch the volume; if not, stop revisiting the question for the month.
- For contracts priced on a monthly index average, add a term referencing the weekly print by 15 September. A monthly average in a market moving 3.6% a week systematically underprices your cargo by roughly half a week of movement, which on 40 FEU is about USD 7,000 a month.
The Shanghai Containerized Freight Index closed at 3,509.54 points on 28 August, up 2.9% weekly and its highest in 25 months. Shanghai-US West Coast rates reached USD 6,940 per FEU (+2.6%) while East Coast spot breached USD 10,000 per FEU for the first time since 2022, up 3.6%. Shippers face sustained pressure on transpacific space through September as peak-season demand and Panama Canal limits persist.
The Analysis
Three thousand five hundred and nine point five four. Up 2.9%. A twenty-five-month high. Three numbers, and if you put any of them into a September budget without doing one more division, you will be wrong by a measurable amount.
The lane numbers are the ones that matter. Shanghai to US West Coast settled at USD 6,940 per FEU, up 2.6% on the week. US East Coast spot crossed USD 10,000 per FEU for the first time since 2022, up 3.6%. Those are the only two figures in the release that describe a route somebody actually ships on.
Everything else is a basket. A basket is fine for a headline and useless for a booking.
Run the subtraction before anything else, because the percentages are hiding the size of the move. A 3.6% weekly gain on a rate that now sits at USD 10,000 means the East Coast rate was about USD 9,652 the week before, so the week added roughly USD 347 per FEU. A 2.6% gain on USD 6,940 means the West Coast was about USD 6,764, so the week added about USD 176 per FEU. One percentage point of difference between the two headlines turns into almost double the money. That is the first thing a rate table does to you when you read it as percentages instead of dollars.
The composite moves less than the story suggests. At 3,509.54 and plus 2.9%, the index added about 99 points on the week, and 2.9% is a weighted average sitting between a 2.6% West Coast and a 3.6% East Coast. If your cargo goes to the East Coast, the headline understates your week by roughly 20%: budgeting at 2.9% on a 9,652 base gives you an increase of about USD 280 per FEU, and the actual number was about USD 347. Sixty-seven dollars a box, per week, is the price of using the wrong number. It is not a rounding error when it repeats.
The twenty-five-month reference point needs the same treatment. Twenty-five months before the end of August 2026 is July 2024, so a twenty-five-month high means this is the strongest print since the summer of 2024. It does not mean the index is at a record, and it certainly does not mean the East Coast is at a record: the release is explicit that this is the first time above USD 10,000 since 2022, which makes 10,000 a re-entry into old territory rather than a new peak. Those are different risk postures. A market at a record has no reference above it. A market re-entering a level it last saw four years ago has a very clear reference above it, and everybody in the room remembers what happened next.
The causal chain is short and it is not mysterious. Peak season demand is running into a Panama Canal that is still restricting transits. When the canal gives you fewer slots, the boxes do not disappear; they either wait, or they go the other way, which for Asia to the US East Coast means the long all-water route or the West Coast plus inland rail. Both alternatives are more expensive per box than a canal transit was, and the East Coast premium is the price of that substitution showing up in the spot market. The West Coast number is not rising as fast because it is absorbing the displaced volume at the same time as it absorbs peak season, and a 2.6% weekly move on a 6,940 base is what absorbing looks like when the alternative is worse.
Who feels this, in order. A shipper moving low-value, high-volume goods across the Pacific is destroyed by it, because the freight is a large share of the landed cost and there is no way to pass it on inside a seasonal window. A shipper moving high-value goods barely notices. Take a 20-tonne payload in a 40-foot box, which is my assumption for the arithmetic and not a figure from the source. At USD 6,940 per FEU the West Coast rate is USD 0.347 per kilogram. At USD 10,000 the East Coast is USD 0.50 per kilogram. For goods with a value of USD 3 per kilogram, freight is 11.6% of goods value on the West Coast and 16.7% on the East Coast. For goods at USD 30 per kilogram, the same rates are 1.2% and 1.7%. Same index, same week, one shipper loses five points of margin and the other loses a rounding error. That is what the phrase value density means when you stop using it as a slogan.
The timing is compressed. The release says pressure runs through September, and the print is dated 28 August. That gives you roughly four weekly readings before the month is over, and the market is telling you that each of them is likely to be higher than the last. Every week you leave a September booking unbooked is a week priced at the next print rather than this one.
The part the headlines have not done for you is the base question, and it is worth stating plainly. A percentage is a statement about a base, and the two bases here differ by 44%. East Coast at 10,000 against West Coast at 6,940 is a spread of USD 3,060 per FEU. That spread is the number that should drive a routing decision, not the percentage change. It says: any shipper who can reach the East Coast market through the West Coast plus an inland move should do so if and only if that inland move, plus the extra transit days, costs less than USD 3,060 per FEU. Above that, pay the all-water rate. Below it, rail it. The index will not make that decision for you, and neither will a 3.6% headline.
Put the whole thing on a monthly budget. Assume 40 FEU a month to the East Coast, which is my assumption. At 10,000 that is USD 400,000 of ocean freight. At the prior week's 9,652 it was USD 386,100. One week of index movement, on one month of one lane, is USD 13,900. Extend the same weekly rate of change across the four remaining weeks of September and the arithmetic puts the East Coast at about USD 11,520 by the end of the month, which is 15% above where it is now. That is arithmetic, not a forecast, and it is the reason to lock volume rather than to wait for a dip that the Panama Canal has no incentive to create.
The counterfactual is the useful part. If the canal restrictions had eased in July, the displaced volume would have gone back through the canal, the all-water alternative would have carried less, and the East Coast premium would have been narrower than USD 3,060. Every dollar of that spread is a subsidy to whoever owns the alternative capacity, and right now the alternative is rail. Shippers who booked rail capacity in June are being paid by this market. Shippers who waited are paying it.
The trap is arithmetic laziness in a specific form: quoting a monthly average when the contract settles on a weekly print, or budgeting a lane at the composite rate because the composite is the only number in the press release. Both errors are small in a flat market and expensive in one moving 3.6% a week. The number that tells you when to lock is not the index. It is the East Coast print against your own all-in alternative, and that breakeven is USD 3,060 today. Cross it and you are no longer shopping; you are just paying.
↑ Back to the key points