The facts. Sea-Intelligence estimates that 1.7m teu is currently idled by port congestion — a volume that would rank as the world’s eighth-largest container carrier if it were a fleet. Late vessels are arriving five to five and a half days behind schedule. Maersk frames the underlying imbalance simply: Far East volumes have grown 25% since 2024, while terminal capacity has grown 10%. Carriers now describe the strain as structural, and expect freight rates to be set increasingly by port and inland capacity rather than by the ships they deploy. (Source: Lloyd’s List, 17 August 2026)
A fifteen-point gap that ships cannot close
Twenty-five against ten. That is the whole story in two numbers, and it explains why the industry’s usual reflex no longer works. For two decades, capacity problems in container shipping were answered with tonnage: order bigger ships, deploy more of them, absorb the demand at sea. Applied to this gap, that reflex makes things worse. More boxes arriving at a shore that cannot process them do not become throughput — they become queue.
The constraint has moved from the water to the quay, and the quay operates on a completely different clock.
Two ways to add shore capacity, on very different timescales
The first is to build. In a single week this month, Lyttelton committed NZ$821m to terminal expansion, the US Army Corps cleared a new deepwater container terminal in Louisiana, and Montreal broke ground on its new facility. That is not coincidence — it is an industry reading the same numbers. But a terminal announced in 2026 handles its first box several years later.
The second is to extract more throughput from terminals that already exist: more machines on the yard, longer gate hours, additional shifts, tighter stacking strategies, less downtime. This route is measured in months, not years.
Between now and the moment the new terminals open, the entire burden falls on the second route. That is a long window, and it is where equipment demand lives.
What congestion does to a machine fleet
Congestion is not simply more work — it is work distributed badly. When vessels arrive five days late, they do not arrive late in an orderly way: they bunch. A yard fleet sized for average demand fails at the peak, and the peak is precisely when the terminal is being judged on turnaround.
Three consequences follow, and they are the ones operators tend to discover in sequence. Utilisation per machine rises, which compresses overhaul intervals and accelerates parts consumption. Redundancy stops looking like waste — a spare unit that idles half the year pays for itself the first time it covers a breakdown during a berth window. And the cost of downtime changes character: a machine out of service on a quiet yard is an inconvenience, while the same machine out of service on a congested yard is a delayed vessel.
Lead times decide who buys what
Here is the part that rarely makes it into the freight-rate commentary. New equipment sits in the same industrial queue as everything else. A terminal that needs more capacity this quarter cannot order its way there — the machines would arrive after the problem they were bought to solve.
That leaves three instruments that operate on the timescale of the problem itself: buying used, renting, and overhauling what is already on the yard. Demand under these conditions does not simply grow; it redistributes toward whatever can be put to work quickly.
What to watch next
Whether the wave of terminal announcements continues, and where. The three this month were in New Zealand, the United States and Canada — European availability of released equipment depends on whether Mediterranean and North European operators follow.
Whether spot rates hold. Carriers calling the strain structural is a forecast, not a fact. If congestion eases through the autumn, the urgency drains out of the shore-side capacity question quickly.
And weather, which in a system with no slack stops being background noise. Average vessel waiting time at Shanghai stood at 87 hours after Typhoon Dolphin. In a port with spare capacity that is a bad week. In a port already at its limit, it is the event that propagates across an entire trade lane.
