Supply Chain Action Points
Read this first — the conclusion, and the moves to make:
- By 28 July 2026, get written confirmation of which services move to Wando Welch and which to North Charleston, and whether export receivals and empty returns sit inside the pause; re-quote every August booking that still names no terminal, targeting 100 percent terminal-confirmed coverage.
- By 31 July 2026, pre-book gate appointments and drayage for the first ten days of August, and hold 15 percent of August volume as a switchable tranche to Savannah or Norfolk with rate and transit already quoted in writing.
- From 1 August 2026, measure gate turn time weekly against your July baseline; trigger is two consecutive weeks more than 25 percent above baseline, which moves the switchable tranche under a named decision-maker rather than a committee.
- By 15 August 2026, write a truck-or-rail fallback into the contract for the top 20 percent of SKUs by margin on any barge-dependent European inland leg, with 72-hour activation and a capped surcharge.
- By 30 September 2026, settle the structural question: if the pause runs past 1 October, treat the lost share as gone and rebuild three-year corridor cost with Savannah and Norfolk trucking inside the total rather than beside it.
Charleston will temporarily pause container handling at its Leatherman terminal from August 1, shifting services to Wando Welch and North Charleston. Separately, Northern European congestion persists, with barge delays of 72+ hours at Rotterdam and Antwerp and yard utilization near 90%, relief not expected before August.
The Analysis
Spread two maps out side by side, because this arrived as two stories and it is one. On one map, a harbour in the American Southeast takes one of its container terminals out of service on the first of August and pushes the work onto the two it names as receivers, Wando Welch and North Charleston. On the other, the river mouths of northern Europe are holding barges more than seventy-two hours at Rotterdam and Antwerp, yards near ninety percent, relief not expected before August.
Read apart, one is a local inconvenience and the other a seasonal headache. Read together, they show capacity being withdrawn from the layer of the network that was meant to absorb the overflow, in the same weeks the overflow begins.
The facts fit in one hand. Container handling at Leatherman pauses from 1 August, services redistributed to Wando Welch and North Charleston. Barge delays past seventy-two hours at Rotterdam and Antwerp. Yard utilisation near ninety percent. Relief not before August. Note what is absent: no volume for the terminal being paused, no end date, no word on whether export receivals and empty evacuation sit inside the pause, and no map of which service goes where.
Start at the European end, where the mechanism is most often read backwards. A yard at ninety percent is not almost full; it is a yard that has stopped buffering. The stack holds a box for the two or three days between discharge and the next leg, and when the next leg is a barge running seventy-two hours late, the box waits in the only place with space. Utilisation climbs because the inland leg failed, not because the ocean leg overdelivered. Treat the percentage as the disease and you buy the wrong cure: more yard is slow and dear, while barge capacity and barge scheduling are neither.
Then the screw turns. Overflow from a stalled barge goes to truck and rail, both with hard ceilings in drivers, slots and wagons, both dearer per box. The congestion does not stay in the water. It changes mode, and in changing mode it reprices every box that touches the port, including the ones that never meant to touch a barge.
Charleston is the opposite kind of event. Nothing failed there; something was withdrawn. A terminal is a gate, a berth, a stack and a rail connection, and pausing it does not delete throughput, it transfers it. What Leatherman handled arrives at two terminals built for their own traffic, with their own gate lanes and appointment systems. The volume does not vanish. It queues, and the queue stands on the road outside the gate, where drayage bills by the hour.
The figure that decides the outcome is the missing one: how much of the port's volume Leatherman carried. A ten percent transfer is absorbed in a fortnight by a terminal that is well run. Thirty percent is a different port for the rest of the year. Until somebody produces that number, a plan resting on the other two will cope is a guess wearing a timetable.
Who feels it first depends on which side of the water they sit. An importer whose distribution centre sits inside the Charleston drayage radius has the real problem, because the box can go to Savannah or Norfolk and the warehouse cannot; the truck leg from a substitute port is not a detour, it is another journey at another rate with another driver pool. A mid-sized importer without dedicated equipment takes the worst of the allocation at the two remaining terminals and pays spot drayage for it. A consolidator has the least room, since fewer working berths means fewer chances at a connection, and an LCL box that misses its cut waits a week. Exporters lose quietly at both ends: a paused terminal shortens receival windows, a clogged European yard delays empty returns, and neither appears in an import statistic.
Here is what the two headlines say only when read together. Carriers facing northern European congestion do what carriers always do, which is omit calls and hunt for alternative gateways. That valve works because a network keeps spare nodes. The spare nodes on the American side are the secondary East Coast ports, and one of them goes short of a terminal on the first of August. Discretionary volume that would normally test Savannah and Norfolk now lands while Charleston is short, so three ports draw on one drayage pool, one set of rail slots and one driver market along a corridor a few hundred miles long. A network does not carry two buffers. It carries one, and this month both ends reach into it.
The other unwritten part is what a pause does to booking behaviour. A terminal called temporarily closed has an unknown forward schedule, and unknown schedules do not produce brave decisions. Shippers route around uncertainty for longer than they route around a known constraint. Give an end date and traffic returns inside a quarter of it. Give none, and what leaves may not come back for a year, which means the port pays for a maintenance window with market share.
The clock runs like this. Discharge at the two remaining terminals strains in the first week of August, because July bookings cannot be undone. Gate appointments and turn times follow in two to three weeks. Rail ramp dwell follows in three to six, which puts the worst of it in the second half of August and early September, and drayage contracts reprice a quarter after that. In Europe the sequence is inverted and faster: seventy-two hours of barge delay and a yard at ninety leave no lag to absorb, only a queue to join.
Do the arithmetic, assumptions mine. Take an importer moving forty FEU a month through Charleston, my number and not the article's. Each box re-handled at another terminal picks up three extra days of yard dwell and one extra truck leg, also mine. Forty times three is a hundred and twenty box-days the port did not have last month, and a yard has no elasticity: the hundred and twenty-first box-day does not cost more, it moves to the next box, and the next box is somebody's cargo sitting on a chassis outside a gate. In Europe, assume a barge of two hundred TEU, conservative and my assumption again. Seventy-two hours across three such barges is six hundred TEU that either waits or goes by road, and by road at two per truck that is three hundred tractor movements hunting three hundred slots in a week that never had them.
The counterfactual argues for patience. Had Charleston kept Leatherman running through August at reduced productivity, it would have held its receival windows and its share and paid in cranage hours and safety margin. A pause buys certainty of execution with certainty of service. That is defensible for a terminal operator, expensive for a shipper, and the two are not the same party.
The trap is reading temporarily as briefly. The costlier one is treating the alternative gateway as free capacity. It is the same corridor and the same drivers, and this month it is also Europe's overflow valve.
Harbours are the oldest thing in this trade and the slowest to change. A berth is built for fifty years, a crane for twenty, a drayage contract for one, a booking for a fortnight, and every congestion story lives in the gap between those clocks. The network is not breaking. It is being asked to move two full loads through one set of valves, and the valves are on the land side, where nobody is watching the water. Watch the gate, not the berth.
↑ Back to the key points