Supply Chain Action Points
Read this first — the conclusion, and the moves to make:
- By 15 October, sort every air-eligible SKU by declared value per piece and remove everything under 30 euro from the air list; that 30 euro line is my own assumption, derived from holding the charge under 10% of value, not a figure from the article.
- Book board space against the second half of October rather than November: lock peak capacity for 20 to 31 October, then trim from 1 November once actual bookings land.
- Change the packing spec by the end of September, merging loose pieces that can ship as one item, targeting a 20% cut in total piece count and lifting average declared value per piece above 30 euro.
- Add a piece-count check at the origin warehouse, finished within 24 hours before dispatch, and hold any shipment where the count does not match the carton spec.
- Shift everything with a 5 to 12 day service requirement from air to LCL sea or rail, completing a 30% volume migration test by mid-October and keeping only sub-5-day cargo on aircraft.
The EU ended its 150 euro duty-free threshold for low-value imports on 1 July, replacing it with a 3 euro per-item charge (plus a 2 euro fee from November). China's low-value and e-commerce exports fell 7% year on year in May, Hong Kong export tonnage dropped 12% in early July, and Asia-Europe air volumes were down 14% YoY as the parcel-driven trade that buoyed air freight unwinds.
The Analysis
The route starts on a sorting floor in Kowloon, runs twenty minutes to Chek Lap Kok, gets built into a ULD on the terminal, and lands at Liege or Frankfurt or Amsterdam before the morning truck wave. Then somebody breaks the pallet down, sorts a few thousand pieces, and hands them to a parcel carrier. Every step in that chain is timed, and every step was built on one assumption: the goods inside are cheap enough that nobody wants to count them one by one.
On 1 July the EU stopped making that assumption. The 150 euro threshold went away, 3 euro per item took its place, and a further 2 euro lands from November. The volume response is already in the numbers. China's low-value and e-commerce exports were down 7% year on year in May. Hong Kong export tonnage dropped 12% in early July. Asia-Europe air volumes are down 14% year on year.
Read those three figures in the order they happened and the story changes. May comes before July. Exports were already falling two months before the rule applied, which means part of the drop is not the rule biting at all, it is shippers running the calculation early and deciding not to fly. Then July arrives and Hong Kong tonnage falls 12% inside ten days. Demand does not move 12% in ten days. A switch does.
That distinction tells you what kind of event this is. Not a slow erosion of parcel volumes, but a step change, taken deliberately, by people who had already done the math.
Friction lands on the terminal floor, because that is where the physical work happens. Air cargo is organised around density. You have a ULD, a fixed volume, a weight ceiling, and a loader who wants the build finished before cutoff. A consolidator fills that ULD with a few thousand small pieces, and the labour on the floor is measured in pieces per hour while the commercial side has always been measured in kilos. The new charge moves the cost onto pieces, and pieces are exactly the dimension the terminal floor is worst at. One more piece is another scan, another sort, another position in the bag, and the labour does not shrink when the value does.
The two ends of the chain react differently. In Hong Kong and Shenzhen the consolidation warehouses are built for small, light and fast, with packing lines laid out for single-piece orders. In Europe the receiving side has to take a pallet that used to move as an undifferentiated bag of parcels and break it into lines that each carry a charge. That is a slower unload. A slower unload at a cross-dock in Liege in the last week of October is not a paperwork problem, it is a dock door sitting occupied while the next truck waits.
Sellers of 8 to 20 euro items and the consolidators who built their business on them feel this inside one billing cycle. An item at 12 euro does not survive a 3 euro charge, and it certainly does not survive 5 euro. Goods that were already above 150 euro barely notice, because they were paying duty before and their clearance was never simplified. Pharma, electronics, spare parts, anything with high value per cubic metre, keeps flying and may even get better board availability, because the belly space the parcels used to occupy is now free.
Direction matters as much as size. This is a charge on goods entering the EU, so a European exporter selling into Asia is untouched, and an Asian exporter selling into the US is untouched by the rule while benefiting from the released capacity, because the same freighters and the same bellies serve overlapping lanes. If you sit on those lanes, this is the rare month when the constraint loosens instead of tightening, and the useful move is to lock board space you could not get in June.
Timing is where most planning goes wrong. The first switch has already been thrown, and July's tonnage numbers are the proof. The second switch is dated 1 November and it will not wait for 1 November. Look at what happened in May: shippers moved before the deadline, not after it. Adding 2 euro to an existing 3 euro charge is a 67% increase in the per-item cost, and anyone holding inventory will front-run that by two to three weeks. So the air peak is not in November. It is compressed into the second half of October, and then it falls off a cliff in the first week of November.
Most coverage will report the 14% decline and stop there. The part that costs money is the shape of the curve, not its height. A peak that arrives three weeks early, on a chain that plans by calendar month, is a chain that misses the peak entirely. If the capacity plan says November is the busy month, you will be holding board space in the wrong fortnight and paying for it.
Run the arithmetic on one shipment. Take 1,000 pieces with an average declared value of 12 euro, so 12,000 euro of goods; that is my assumption, not a number from the article. At 3 euro per item the charge is 3,000 euro, a quarter of the cargo value. From November, at 5 euro per item, it is 5,000 euro, or 41.7%. Keeping the charge under 10% of value requires an item worth 30 euro today and 50 euro from November. That is the entire filter, and it is brutal. Anything under 30 euro should not be on an aircraft.
There is a second bill nobody has priced. Per item is the most expensive phrase in this story, because the gap between a parcel and a piece is often a factor of two or three in the count. A shipment of 1,000 parcels containing 2,400 individual units is not the same consignment under this rule. Being counted at 1,000 or at 2,400 is the difference between 3,000 euro and 7,200 euro on one pallet. That is not a rounding question, it is a repack.
Doing nothing has two bad endings. The goods keep flying, the charge gets added to landed cost, the seller absorbs it until margin is gone, then raises the price and loses the sale, and freight becomes a tax on being small. Pulling everything out of the air is worse in a different way: three to five weeks added to a product that used to arrive in five days is, in e-commerce, indistinguishable from discontinuing it. Neither extreme is a plan.
The trap I expect is treating this as a demand problem and waiting for volumes to come back. They will not come back in the same shape. Booking November capacity on the assumption that the peak sits where the calendar says it sits is the second trap. Packing for weight when the charge is per piece is the third, and it is the one that costs the most per hour of floor work.
What you can do today is one thing: pull the SKU list, sort it by declared value per piece, and cut the line at 30 euro. Everything else grows out of that list.
↑ Back to the key points