Strip the hype away and you get this: ocean freight rates are up, again, and the numbers say the cause is physical, not speculative. The Shanghai Containerized Freight Index closed at 3,509.53 points on August 28, up 2.9 percent week on week and up for a fifth consecutive week. US West Coast rates hit $6,940 per FEU, up 2.6 percent; US East Coast rates hit $10,046 per FEU, up 3.6 percent — the first time that route has broken ten thousand dollars. The Persian Gulf line rose 7.1 percent to $6,139 per TEU. Those are the raw readings. Let me now do what the marketing departments will not do and read them properly.
Rates are a symptom. The disease is effective capacity. The number that closes the argument is the congestion figure from Linerlytica: global port congestion reached 4.3 million TEU, above the pandemic-era peak of 4 million TEU. That is the line I want everyone to stare at, because it is the point where the story stops being about prices and becomes about physics. When 4.3 million containers are sitting in ports waiting to move, the vessels carrying the next cargo have nowhere to put it down, and the price of every slot rises as a direct mechanical consequence. That is not sentiment; that is a queue.
Three physical squeezes, one rate line
Let me decompose the congestion into its components, because a materials engineer does not accept a single-cause explanation. The first squeeze is the Red Sea rerouting: vessels are taking the long way around Africa rather than transiting the canal, which consumes effective capacity even when utilisation looks normal — the same volume is being carried on a longer pipe. The second is Panama: the canal cut daily transits to 34 ships from September 3, and will cut them again to 32 from September 15. That is a hard, scheduled reduction in a second major artery. The third is weather: typhoon Dolphin hit Ningbo and closed the port on August 7-8, locking in 2.4 million TEU of North American capacity.
Add those together and the 4.3 million TEU congestion number is not a mystery — it is the sum of a reroute, a schedule cut and a storm, all landing in the same quarter. What matters for anyone who actually moves goods is that all three are largely outside any single carrier’s control. That is the difference between a price cycle you can arbitrage and a constraint you can only manage around. The numbers say this is the second kind.
I want to be precise about the SCFI readings themselves, because the details carry information. A fifth consecutive weekly gain tells you the trend is not a one-off spike. The US East Coast crossing ten thousand per FEU tells you which trade lane is absorbing the most constraint — the route that depends on both Panama and the longer Red Sea alternatives. And the Persian Gulf line being the fastest riser at 7.1 percent ties this story directly to the security situation in the Gulf, which I have covered separately. The rates are not random; they are ranked by the physical difficulty of each route.
The GRI wave is a confirmation, not a cause
Now let me deal with the September 1 general rate increases, because they will get the headlines and they are the least interesting part of the story. Several major carriers — including COSCO, CMA CGM, Hapag-Lloyd, Evergreen and Yang Ming — implemented new GRI from September 1, with US West Coast 40HQ rates typically raised by $1,800 to $2,250. The no-hype reading: a GRI is a price announcement, and price announcements in a constrained market are confirmations, not causes. The carriers are not creating the scarcity; they are collecting the rent that the scarcity has already created. If congestion were falling, the same GRIs would fail in the market. They are rising because the queue says they can.
Maersk’s numbers back this up from the carrier side. Q2 revenue came in at $15.8 billion, up 20 percent, and the company raised its full-year underlying EBITDA guidance to $10.5-12.5 billion. Let me put that in engineering terms: the same fleet, carrying roughly the same cargo, earning 20 percent more revenue because each move takes longer and each slot is worth more. That is the definition of a capacity-constrained market — revenue per unit rises with the cycle time. The carriers are not doing anything clever; they are simply getting paid for the longer path and the longer queue.
At this point I should also note what the numbers do not tell us, because that is part of the job. The congestion figure is a snapshot, and congestion can dissipate faster than it builds — a clear weather window, a resolved labour issue, a seasonal demand dip. The rate line is a weekly reading and can snap back when the queue clears. Nothing in this data says rates are at a permanent new plateau. What the data does say, firmly, is that the current level is supported by physical constraints rather than by noise, and that predicting a fast normalisation would require ignoring every one of the three squeezes still in force.
The real constraint is the queue
Let me close with the takeaway I actually trust, because it is the one the numbers force and the headlines will not carry. The shipping market is not in a speculative bubble; it is in a physical constraint state, evidenced by 4.3 million TEU of port congestion above the pandemic peak, by scheduled Panama transits falling to 32 ships, and by a typhoon having locked 2.4 million TEU of capacity. As long as those constraints hold, high rates are the equilibrium, and the direction of the queue — not the direction of the press release — is the number to watch.
The numbers say what the marketing won’t: this is not about carriers pricing aggressively; it is about a pipe that is too full. For shippers, the actionable reading is unchanged but now evidence-backed: book early, plan for cycle-time inflation, and assume the constraint is real until the congestion line falls below the pandemic peak. That is the spec-verdict, and it is the one the data authorises. When the queue clears, the rates will follow — and not a day before.
What the data cannot tell you yet
Let me be disciplined about the limits of this data set, because an engineer who overreads a gauge is no better than a marketer who overprices a product. The congestion figure is a point-in-time measurement of a fast-moving system; the rate index is a weekly snapshot of a market that can reverse within a month. Neither number tells you the duration of the constraint. A typhoon passes, a canal schedule can be revised upward, and labour situations resolve faster than they build. The data as it stands supports the current level of rates; it does not certify how long the level holds, and any forecast of that duration would be speculation dressed as engineering.
There is also a measurement caveat I want to put on the record, because it is the kind of thing my readers deserve to know. The 4.3 million TEU congestion figure is a modelled estimate from a private analytics firm, not an audited statistic, and the SCFI is a surveyed index whose composition reflects a specific basket of routes and carriers. Both are the best instruments available, and I use them as such — but “best available” is not the same as “complete,” and a prudent reader should hold a small margin of error around every number in this column. The direction is what the instruments support; the exact magnitude carries uncertainty.
What the data does allow me to say, without hedging, is that the rate level is being supported by identifiable physical constraints — the reroute, the canal cuts, the typhoon, the congestion — rather than by speculative positioning alone. That is a meaningful distinction, because a speculative bubble corrects on sentiment while a physical constraint only corrects when the constraint lifts. For anyone planning freight spend into the fourth quarter, the difference between those two worlds is the difference between “wait and see” and “book now.” The evidence points to the latter, and that is the practical conclusion the numbers authorise and the one I will act on myself.
Putting the numbers to work: a checklist for planners
Let me convert the data into a working checklist, because a rate analysis that cannot be acted on is only half a job. First, assume the constraint persists into the fourth quarter — that is the base case the evidence supports, and it is the case that leaves you least exposed if it turns out to be wrong. Second, book long-haul capacity earlier than usual and prefer routes that do not depend on the two constrained canals; the reroute premium is built into every quote right now, but the worst position to be in is the one where you need a slot at the last minute in a constrained market. Third, watch the weekly congestion line as the deciding indicator: when it falls back below the pandemic-era peak of four million TEU, that is the signal that the physical support is loosening and rates have room to correct — not before.
I want to also speak to the temptation this kind of story always produces, which is to treat a single reading as a trend. Five weeks of rising SCFI is a trend; it is also a short one, and the same index has produced sharp reversals inside a month in this very cycle. The discipline is to update on the data, not on the mood: if the congestion line starts falling, the analysis changes even if the news coverage has not caught up. Conversely, if the congestion line keeps climbing, a high rate is not a bubble — it is a price that is finally telling the truth about the pipe. The numbers are the arbiter, and they always will be.
That is the honest end of this data note. The rate levels are high and, on the available evidence, physically supported. The congestion figure is above the pandemic peak, the canal is cutting its schedule, and a typhoon has already taken a meaningful chunk of capacity out of the system. The queue is the constraint, the queue is the arbiter, and the queue is what will decide when this cycle turns. Strip the hype away, and that is what the numbers actually say.
Let me state the bottom line in the fewest words, the way a spec sheet closes. This market, at scale, is not a pricing story; it is a capacity story, and the capacity is physically constrained in three independent ways at once. That’s the real constraint — not carrier strategy, not seasonal demand, not sentiment, but the simple fact that the pipe is full. No hype is required to make the point, and hype would only obscure it. Rates are high because the system is congested, and rates will stay high while the congestion stays. That is the entire analysis in one breath, and it is the one I will act on.
For the reader who wants to track this themselves, the discipline is simple and I recommend it: mark the weekly SCFI and the congestion estimate in the same notebook, because the two move together and the gap between them is where the surprises live. A rate rise with flat congestion is a different signal from a rate rise on rising congestion, and the difference is exactly the difference between noise and a physical event. Watch the queue. The numbers say what the marketing won’t, and they are saying it clearly enough for anyone who takes the time to read them.