Supply Chain Action Points
Read this first — the conclusion, and the moves to make:
- By 8 September, calculate the chargeable-to-gross weight ratio for the last eight weeks on your Europe lane, by SKU family; anything under 0.75 means a quarter of what you buy is air.
- By 15 September, re-cut cartons and pallet build on the two worst families to reach 0.80, then re-measure in week 40 and keep the change only if the ratio holds.
- From 1 October, move the tender cutoff for palletised cargo one full working day earlier, and confirm the build slot with the handler in writing rather than by phone.
- By 15 September, check the volumetric divisor written into each air contract and confirm which dimensions it is applied to, because a divisor applied to the outer carton instead of the pallet is a different invoice.
- Keep at least 40% of fourth-quarter volume on a rolling weekly position until week 40, and decide it on your own shipped kilos rather than on the weekly lane index.
Combined China and Hong Kong-Europe chargeable weight rose 1% week-on-week in week 34 (17-23 August), the first recovery since June, though mainland volumes stayed 8% below a year earlier. Mainland-Europe spot rates climbed for a third straight week to USD 4.14 per kg, still about 24% below the mid-June peak of USD 5.43 per kg as the EU de-minimis removal weighs on e-commerce.
The Analysis
There is a moment at a cargo terminal that explains this whole week, and it is the moment a pallet goes on the scale and then somebody walks around it with a tape measure. What you pay for is not what it weighs. It is the larger of two numbers, and which of the two wins is decided entirely by what is inside the carton.
The figures for week 34, 17 to 23 August: combined China and Hong Kong to Europe chargeable weight rose 1% week on week, the first recovery since June. Mainland volumes are still 8% below a year earlier. Mainland to Europe spot rates climbed for a third straight week to 4.14 dollars a kilo, still about 24% below the mid-June peak of 5.43. The removal of the EU de-minimis threshold is still working its way through e-commerce.
One per cent is not a recovery you can plan a quarter on. What I want to know is what changed inside the aluminium, because that decides whether the next six weeks are a turn or a reshuffle of the same tonnage.
On a lane that moves thousands of tonnes a week, one per cent is a handful of pallets either way, and a single large consignment can manufacture it. The direction is the readable part, and the direction now points up while the volume underneath it is still down: mainland to Europe is 8% below where it sat a year earlier. Rates rising for a third consecutive week against volume that is still negative year on year is not a demand story. Something changed in the shape of the cargo, and shape is a physical thing.
Think about what was on those aircraft in June. It was parcels. A parcel is light and it is bulky: a carton of phone cases, a carton of cosmetics, a carton of small electronics, each one mostly void fill by volume. Fill a freighter with that and the aircraft runs out of cubic metres long before it runs out of payload. Then the de-minimis threshold came off, the parcel flow thinned, and the space was taken by palletised general cargo, which is the opposite animal: dense, stacked, shrink-wrapped, and heavy enough that the aircraft runs out of payload before it runs out of room. Same tube of aluminium. Different contents. And because the billing unit is the chargeable kilo, the same aeroplane reports a different number depending on which of the two it is carrying.
Here is the arithmetic, and every number I supply beyond the article is my own assumption. Air freight bills the greater of gross weight and volumetric weight, and the standard volumetric divisor turns 6,000 cubic centimetres into one kilo. Take a pallet 120 by 80 by 160 centimetres. That is 1.536 cubic metres, which at that divisor is 256 kilos of chargeable weight. Say the pallet actually weighs 180 kilos, a normal density for boxed consumer goods. You are buying 256 and using 180. Seventy-six kilos, close to thirty per cent of the invoice, is air. At the 4.14 dollars a kilo the article reports, that is roughly 314 dollars a pallet paid for nothing. Twenty pallets a week is 6,290 dollars a week, and thirteen weeks of it is a little over 81,000 dollars. None of that is a rate problem. All of it is a loading problem, and loading is the one thing on this list you control without a single negotiation.
Who this hurts depends on what they ship. A parcel consolidator on the mainland is still 8% down on last year and cannot fix it with density, because their cargo is what it is and their only remaining lever is price. A shipper of machinery, tooling or dense electronics is fine and always was, because they weigh out before they cube out and the volumetric rule never bit them. The one getting squeezed in the middle is the mid-size seller who converted from parcels to palletised air freight after the threshold changed. They moved onto a channel billed in chargeable kilos, and they brought parcel-shaped cargo with them.
You feel this at the cutoff, not on the invoice. Loose parcels can be tendered late because they get consolidated and built up at the handler. A pallet needs screening, a build plan and a contour that fits a unit load device, and the people who do that go home at some point in the evening. So the deadline for a pallet moves earlier than the deadline for a parcel on the same flight, often by most of a working day. If your warehouse is still working to the tender time it used a year ago, you are not late for the flight. You are late for the build, and the flight leaves without you either way.
Here is the part nobody printed. Chargeable weight is not a physical quantity. It is the larger of two numbers, and which one wins is decided by what is inside the box rather than by how much cargo there is. A 1% weekly rise in chargeable weight can therefore be a rise in density rather than a rise in cargo, and the reporting cannot tell the two apart. The article reads the gain as the market turning. The dock reads it as a different pallet on the same aeroplane. The check costs nothing: ask your handler for the volume-to-weight split of what is actually flying your lane, and put it next to the same weeks last year. If kilos are up while cubic metres are flat, you are looking at a mix change, and the recovery is narrower than the headline suggests.
Get that wrong and the mistake is expensive in a specific way. Read 4.14 as demand coming back and you commit forward volume and accept a floor, and then October arrives with the parcel tail finished and a market that is smaller and denser than the one you priced. Read the 8% decline as the real signal instead, and you plan for less cargo at a better density, which is the only version of this market where you win on cost per unit instead of on the rate line. The trap is emotional. A number that has stopped falling feels like a number that is about to rise, and those are two different things wearing the same clothes.
The thing I would do before any of that is walk over to the scale with a tape measure. Take one pallet you shipped last week, measure it, weigh it, divide. That answer belongs to you, it costs nothing, and it does not depend on anybody's index.
↑ Back to the key points