Last September, a forty-foot container from Shanghai to Los Angeles cost $2,678.
This September, the same container costs $7,352.
That is an increase of 175% in twelve months, on the busiest import lane into the United States. If your costing still carries last autumn's freight number, it is short by roughly $4,700 per container — and that gap lands on whichever products happen to be in the box.
The number is not really the story, though. The story is that almost nobody saw it coming, including the people paid to see it coming.
In August, the Experts Said the Rush Was Over
The National Retail Federation publishes a monthly forecast of how much cargo is about to arrive at American ports. It is called Global Port Tracker, it is produced with a consultancy called Hackett Associates, and it is the number most retailers plan against.
On 7 August it ran under the headline Import Cargo's Early Peak Season is Winding Down. Jonathan Gold, who runs supply chain and customs policy at the Federation, explained why: “We had an early peak season this year as retailers brought in merchandise ahead of tariff changes in late July.”
It was a sensible call. Everyone had pulled orders forward ahead of the tariff deadline. Of course the autumn would be quiet.
Five weeks later, the same forecast was revised. September is now expected to be the busiest month of the entire year — up 9.6% on last September. Gold again, this time with a different answer: “We thought the peak season would be mostly behind us by now, but that's not the case.”
The ports back him up. The Port of Los Angeles just closed out the busiest three consecutive months in its history, moving more than 2.9 million containers across June, July and August. “We've put together an exceptionally strong summer,” said its executive director, Gene Seroka.
So the best forecast in the industry moved by about seven percentage points in five weeks — at precisely the moment freight rates were hitting their highs for the year.
Hold that thought, because it matters more than any number below.
What a Container Actually Costs Right Now
Here is where the market sits, per forty-foot container, as measured by Drewry's World Container Index — the weekly benchmark most buyers quote.
- Shanghai to Los Angeles: $7,352, up 2% on the week
- Shanghai to New York: $9,726, up 1%
- Shanghai to Genoa: $4,216, down 3%
- Shanghai to Rotterdam: $3,997, down 2%
Notice that the two American lanes are climbing while the two European ones are falling. The overall index has barely moved in a fortnight, which makes the market look calm. It isn't calm — it is pulling apart, and if you only watch the headline number you will miss that entirely.
For scale: a year ago that headline number was $2,044. Today it is $4,476 — an increase of 119%.
Three People, Three Explanations, All On the Record
This is the part worth arguing about, because the experts disagree in public.
One: the carriers took the ships away. When a shipping line cancels a scheduled departure, the industry calls it a blank sailing — the ship simply doesn't run, and all its space leaves the market. Drewry's tracker counted 47 cancellations out of 729 planned departures over five weeks, and 68% of them were on the Asia-to-America route. Less space, same cargo, higher price.
The carriers have also been raising prices across the board on set dates — a general rate increase. The one that landed on 1 September was, remarkably, the seventeenth of the year on that lane. CMA CGM, COSCO, Yang Ming and ZIM each added $2,000 per container. Evergreen, Hapag-Lloyd and HMM each added $3,000.
Two: the Panama Canal is running out of water. In an advisory on 20 August, the Panama Canal Authority cited poor rainfall and tightened the rules. Ships must now sit shallower in the water, which means they carry less, and fewer of them get through each day — the largest vessels were cut to nine crossings a day from 3 September, and the smaller locks dropped again on 15 September. Every container heading for the East Coast pays for that.
Three: people are just buying more. Maersk's chief executive, Vincent Clerc, rejects the whole premise that this is a temporary rush: “This is not a pull-forward. It is real underlying demand that has led us to increase our expectation of growth.”
These cannot all be the main cause, and which one you believe changes what you should do next. If it is the carriers, rates fall the moment they put the ships back. If it is the canal, rates fall when it rains. If Clerc is right, this holds into 2027 and you should be booking now.
No neutral source is going to settle that for you.
The East Coast Costs $2,374 More, and People Shipped West Anyway
New York is running $2,374 above Los Angeles per container. That is the canal, the longer voyage and East Coast demand, all showing up in one number.
What importers did about it is the interesting bit. Even with the West Coast looking cheaper on paper, they chose Los Angeles for its rail links and handling speed — because getting the box from the dock to your warehouse is a completely separate bill from getting it across the ocean.
Which raises a fair question. When the ocean leg swings by $2,374, does your routing decision actually get looked at again? Or does the container go where it has always gone?
For most importers it is the second one — not out of laziness, but because comparing the two routes honestly means comparing total delivered cost rather than freight, and that is a day's work nobody has while a purchase order sits waiting.
What This Does to the Quote on Your Desk
Four practical consequences.
A freight number more than a few weeks old is decoration. Seventeen price rises in nine months means this lane reprices roughly every two and a half weeks. A costing built in the spring isn't cautious or optimistic — it is describing a market that no longer exists.
The rate-increase date is a deadline. Prices moved on 1 September and again on 15 September. Cargo that made it into the terminal before the date paid the old rate. Whether yours did is usually decided by how fast the factory finished, not by your forwarder.
Who pays is about to become a conversation. If you buy at the factory door, this market is yours. If your supplier delivers to you, it is theirs — but they will come back for it, and probably in the next price round rather than this one. If it has been a while since you checked which side of that line your terms put you on, now is a reasonable moment.
A cancelled sailing is a stockout with a delay attached. Six per cent of departures pulled is not a rounding error when your reorder cycle assumes a fixed transit time. The goods aren't lost. They're late. Late is what empties a shelf.
Lock It In, or Ride It Out?
This is the real decision, and there is no clean answer.
Lock it in. The lane is up 175%, Maersk is telling investors the demand is structural, and nobody can negotiate with rainfall. Fixing your rate takes the single most volatile line out of your cost and lets you quote with confidence for a year.
Ride it out. You would be fixing at the top. Seventeen rate rises in a year is what a carrier does when it can, not when it must — and Europe is already softening on the same index in the same week. Space that was withdrawn can come back in a fortnight. Sign twelve months at peak pricing and you may spend most of 2027 paying above a spot market you could have used.
Both cases are built from the same figures. The tiebreaker isn't a forecast — the forecast has already been wrong once this quarter, publicly, by seven points. It is whether your business can absorb a shock or whether it needs certainty more than it needs the best price.
Those are different companies. They should probably make different decisions.
The Real Problem Isn't the Rate
Every argument above assumes you can tell, quickly, what a $2,000 increase does to the margin on one specific item.
In most import businesses, that is not a quick question. The freight sits in one place, the purchase order in another, and the cost per unit gets worked out by hand once the invoice arrives — by which point the decision it was meant to inform has already been made.
That is what actually hurts. Not the rate. The gap between the rate changing and anyone knowing what it meant.
The importers who came through this year in decent shape mostly weren't the ones who predicted the market. They were the ones who could reprice fast when it moved.
Check These Rather Than Trusting This Article
Freight moves weekly. Everything above was accurate on 10 September 2026, and some of it will be stale within a fortnight.
• Drewry World Container Index — the weekly rate benchmark, updated Thursdays
• Drewry Cancelled Sailings Tracker — which departures are being pulled, by lane
• National Retail Federation Global Port Tracker — monthly forecast of United States import volumes
• Panama Canal Authority advisories — transit limits, issued as they change
• Port of Los Angeles container statistics — what actually arrived, which is a useful check on any forecast
So — when the 1 September increase landed, how long did it take you to know which products were still profitable at the new number? And where does your freight cost live right now: on the purchase order, in a spreadsheet, or in the forwarder's invoice after the fact?