Supply Chain Action Points
Read this first — the conclusion, and the moves to make:
- By 31 October, get the berth and draft terms in writing for your lane: maximum arrival draft, any tide or trim condition, and whether the call carries a load-plan restriction; do not quote a customer on the phrase ultra large capable without this.
- Add hop count as an explicit column to your lane map within two weeks, and put every lane with two or more handlings on the list for a direct-call test in the first-quarter 2027 tender.
- For the next sixty days, quote from the ninety-fifth percentile rather than the mean, and log every delivery that lands outside the quoted window; that log is your evidence for repricing the promise.
- Run a two-quarter parallel from January 2027: twenty to thirty per cent of Mexico-origin volume on a direct Manzanillo call, the rest on the current routing, reviewed at the end of the first quarter against measured spread and not against the mean.
- Check quarterly whether the only-terminal status still holds, starting 31 December, because that status is the whole of the differentiation and it ends silently.
Contecon Manzanillo (CMSA), ICTSI's Mexican unit, won approval to handle vessels with drafts up to 15.5 metres, making it the only terminal at the Port of Manzanillo able to serve ships up to 400 metres long and around 24,000 TEU. The clearance lets the Pacific coast gateway capture larger Transpacific calls and ease pressure on busier West Coast ports.
The Analysis
Ask anyone in this business to describe a port and they will describe cranes. Ask them what actually decides which ships can call, and the answer is under the water, invisible from the quay and paid for decades before anyone thinks to mention it.
On 6 September, Contecon Manzanillo, the Mexican unit of ICTSI, was cleared to handle vessels drawing up to 15.5 metres, which makes it the only terminal at Manzanillo able to serve ships up to 400 metres long and around 24,000 TEU.
A draft approval sounds like a technicality. It is closer to a change of class, and class is the variable that decides which strings a node can even be written into.
Three numbers carry the story: 15.5 metres, 400 metres, 24,000 TEU. Around them sit one status and one claim. The status is that no other terminal at Manzanillo can do this. The claim is that the gateway can now capture larger Transpacific calls and take pressure off busier West Coast ports.
Draft is the least glamorous capital in the system and the most decisive, which is why it is worth being precise about what it actually is. A crane can be ordered and delivered in a couple of years. Depth cannot. Fifteen and a half metres represents a dredged channel, a turning basin, a berth line built to hold a ship of that length, and a maintenance budget that never ends, because channels silt. That is why a draft approval changes a node's class rather than its capacity. Capacity is how much you can move. Class is which ships can be bothered to come.
Now the reallocation, which the word pressure obscures. A very large call does not add capacity to the Transpacific. It moves a call. If a ship of that size berths at Manzanillo, the cargo it carries was going somewhere before: discharged at a busier West Coast port and railed or trucked south, transhipped at a hub, or split across two smaller calls. All three of those routings lose volume when the direct call appears. So the pressure that eases is not congestion in any simple sense. It is the number of calls competing for berth windows at ports that were already full, and the volume that leaves them does not disappear from the ocean, it changes address.
There is a ceiling inside the approval, and it deserves attention because it is where an importer gets surprised. Up to 15.5 metres is, by construction, a bound. If a ship of that class would otherwise sit deeper, the difference does not cancel the call, it moves into the load plan: the vessel sails a little lighter, or the carrier trims with bunkers and ballast, or the cargo is split across calls. The constraint does not vanish. It migrates from the port's channel to the carrier's stowage computer and then to the booking desk, which now works inside a weight and draft envelope. Cargo that is heavy for its volume feels that first, so steel, minerals, bagged agricultural goods and machinery meet the ceiling before consumer goods do, and they meet it in the form of a booking restriction rather than a port problem.
What will not appear in a port announcement is the reason this approval is worth more than it looks. A container's journey is a chain of handlings, and every handling is a random variable with a mean and a spread. Add a transhipment and you add a dwell distribution at a hub. When you combine independent distributions the spreads do not add, they combine in quadrature, which means the tail thickens faster than the middle slows. Removing a hop therefore does not mainly remove its average. It removes its variance. What a direct call from a Pacific gateway actually sells is not a shorter average transit, it is a shorter worst case, and the worst case is what your safety stock is sized against. That is where the money is, and it is booked as working capital, which is why nobody attributes it to a port.
Who this lands on. Exporters in the Bajio automotive corridor and the Guadalajara electronics cluster gain most, because their cargo is high in value density and sensitive to arrival date, so a tighter tail converts almost directly into inventory they no longer have to hold. Bulk and project shippers gain least: when your lot is large and your schedule is planned a quarter ahead, a variance reduction is worth little because your buffer is a stockpile, not a promise. Small exporters gain nothing directly, since a very large call is allocated in blocks, and their benefit is second order, arriving only if the big call drags feeder frequency up behind it.
The timing is the part people get wrong. The approval is a dated event; the effect is not. A terminal gets cleared, and then a carrier's network planning round decides whether to write a string into it, and those rounds happen once or twice a year. So the first real call is a planning decision, not a ribbon. For an exporter the effect lands at the tender, which means the question to ask now is not whether Manzanillo can take the ship but whether any string you can buy is being rewritten to use it.
Take an arithmetic, with the figures mine and none of them from the announcement. Assume a shipper moving thirty FEU a month from the Bajio to Asia, and assume the present routing includes one transhipment with a mean dwell of six days and a spread of three days, against a total door-to-vessel distribution with a spread of 4.6 days. Remove the hop and the remaining spread is the square root of 4.6 squared minus three squared, which is about 3.5 days. Written as a ninety-five per cent buffer, that is 1.645 times each spread, so the buffer falls from roughly 7.6 days to roughly 5.7. Call it two days. Two days is not dramatic and I will not pretend it is. What matters is where it lands: it lands on the promise, not the price. On a lane where the customer's line runs on your delivery date, two days of tighter promise is the difference between being in a tender and watching one.
Run the counterfactual. Without the approval, Manzanillo stays a regional call, the Bajio keeps its hop, and nobody notices, because the cost of a hop is paid in safety stock, and safety stock is booked as working capital and never attributed to a port. Nothing breaks. That is precisely the problem with structural costs: they are only visible when someone removes them, which is why ports compete on depth rather than on price.
The trap is quoting a customer on the phrase ultra large capable. A draft approval is a permission, not a schedule, and a schedule is what you sell. The second trap is treating the exclusivity as permanent. The value of this clearance rests entirely on being the only terminal at Manzanillo that can do it, and that status ends the day a neighbour dredges. Your negotiating position is a lease on a fact, and facts of that kind expire.
Depth is the slowest variable in this industry and the one that redraws the map. Ten years from now nobody will remember the clearance date, but the strings will have been rewritten around it or will not have been, and that is the only thing that will still be visible. Watch for the second caller and for the second dredge. Until both appear, use the gateway for what it is, which is one fewer handling on a lane you already run, and keep the promise you quote two days inside the one you could quote.
↑ Back to the key points