Chattogram NCT 15-Year Concession Clears Cabinet as Port Workers Sit In

Here is what is actually on the table, stripped of adjectives. A draft concession running fifteen years, covering the New Mooring Container Terminal plus the overflow yard, approved in principle rather than signed. A would-be operator from Dubai that has been negotiating since June. A labour response on 5 October — a day-long sit-in at the gate, the road to the administrative building blocked — a torch procession set for 7 October. And three demands: cancel the deal, publish a comparison of what the port earns under state operation against what automation would deliver, and review the twenty-two-year Patenga concession signed with Red Sea Gateway Terminal in December 2023.

Look at the shape of those demands. Only one of them is about tomorrow's productivity. The other two are about whether the paperwork itself can be trusted — what the old model earned, and whether the last long promise still binds. A dispute over efficiency can be settled with numbers. A dispute over whether a signed agreement holds cannot be settled with anything.

Why this lands in October 2026 rather than five years ago has two answers, and only one of them is inside the country.

The inside answer is arithmetic. An export economy that sends almost everything through one estuary eventually notices that dwell time is baked into the price of everything it sells. Every hour a box waits at Chattogram is levied against garments priced in European and American high streets, against competitors whose ports have been doing this faster for thirty years.

The outside answer is the calendar. Bangladesh graduating out of the least-developed-country category has been the fixed point everyone in Dhaka plans around, and it has been pencilled in for later this year. Come out of that category and preferential tariff access begins its scheduled erosion. When tariff preference erodes, the unit cost of a shirt has to fall somewhere else, and the two largest places available are labour practice and port cost. That reframes the whole affair: a port concession stops being a transport question and becomes the competitiveness question for an entire export base. Nothing on that basis is going to pass quietly.

Now zoom out, because this stops looking like local news at about a thousand kilometres.

On the map of the eastern Indian Ocean, Chattogram is an origin, not a hub. Volume bound for Europe and North America mostly does not leave on a mainline vessel. It goes south first — Colombo, Singapore, Port Klang — and transfers there. That makes Bengal one of the few large export economies whose access to global networks is leased from somebody else's harbourmaster.

This is not a trick of geography. It follows from draft, from berth length and crane outreach, all of which have run short against what mainline operators deploy. It is also why a single terminal matters far more than its tonnage suggests: the New Mooring Container Terminal is one of very few places in the country capable of physically receiving a large box ship. Whoever runs NCT is not running a terminal. They are holding the country's widest window onto the mainline world.

Which is why fifteen years is the number to argue about. It is longer than most governments, longer than a typical apparel sourcing cycle, longer than the life expectancy of most shipping alliances, and roughly the pay-back period of the civil works a deep-draft berth needs.

Who carries the consequences, and how far the spreading goes.

Closest hands first: the apparel factories strung along the road north towards Dhaka, and the Western buying offices whose critical path runs along that road and over that quay. Anything touching quay throughput touches ship dates, and ship dates in apparel are contractual rather than aspirational.

A ring out, the carriers. Feeder operators and mainline services whose Bangladesh rotations depend on Chattogram berth windows, and who have spent two decades planning around short dwell windows rather than reliable ones.

Further out, the rest of the country's maritime system. Mongla, and the deep-sea project down the coast that people have been watching for years, both get more interesting every month this one stays unresolved. Every stalled month strengthens the case for a second outlet, and every stalled month also delays it, because the same officials are busy.

And one layer rarely discussed: importers. Ports run in both directions. Ninety per cent of a nation's seaborne trade through one estuary means that industrial inputs and consumer goods coming in depend on the same few quays as the garments going out. A stoppage here is not an export story. It is one national pipeline with one valve.

The thing to plan against is not the dates in the news. It is the gap between them. In-principle approval arrived on 1 October. Workers were at the gate on 5 October. Four days. That is the reaction interval this file currently runs at, and there is no evidence yet of a different interval for the milestone.

Because in-principle approval is not signature, and signature is not handover, at least three further events are likely ahead: conclusion of negotiation and signing, notification of the charging regime, and physical transfer of operations. Three announcements. Three reactions at roughly four days. If any of your cargo moves through Chattogram between now and the middle of next year, build the calendar on that rhythm rather than hoping the pattern breaks.

And the season is working against everybody. The late-year windows for Western summer programmes are precisely the weeks in which this calendar is least forgiving, so every reaction lands on top of the heaviest bookings of the year.

Now the two things nobody has put in a headline, and the reason I do not read this as a simple efficiency-versus-labour story.

The concession itself will almost certainly not be what blocks this port in the next six weeks. Handing a terminal to anybody does not stop a gate. What stops a gate is labour action, and what stops it for months is labour action running on top of an unresolved review of contracts already signed — which is exactly what 5 October produced. A sit-in that blocks the road to the administrative building is survivable in isolation: gates reopen, trucks move, everybody catches up. A demand to reopen a twenty-two-year concession is survivable in isolation too, as a file somebody takes away to consider. Together they do something neither does alone. They put the whole contractual history of this port on the table at the same moment its next fifteen years are being negotiated, so nothing small can be settled by itself. There is no published end date on that overlay, and I have never seen an operating plan survive an open-ended overlay intact.

The fifteen years themselves carry their own suspicion, because they read like fifteen years of improvement and almost never are. Every concession I have watched in three decades has the same shape: the part that costs money arrives at handover, the part that saves time arrives years afterwards, and nobody publishes a calendar you can hold them to. Charges can be reset the day operations transfer. Cranes, yard equipment, gate automation, deeper draft, yard paving and the system integration behind all of it are procurement, civil works, training and commissioning, all carried out while the port keeps running, all of it imported into a working terminal. Three years is quick. Five is normal. Anyone signing a multi-year supply programme out of Bangladesh should build the cost model on the honest assumption that the new numbers land in year one and the promised performance lands somewhere around year four — and write that assumption into the contract rather than into somebody's memo.

Numbers now, with the assumptions written out.

Assume a European apparel buyer moving forty forty-foot containers a week out of Chattogram. Assume purchase order to store shelf takes sixty days. Assume three weeks of buffer stock on the faster-moving lines.

Assume one working week loses three days to disruption. Boxes that should have loaded do not load, and fall onto the following week's slot. One disrupted week therefore pushes roughly one weekly allocation — forty boxes — to the right. Three days lost in each of three consecutive weeks inside one quarter slides something like a hundred and twenty containers. Three weeks of buffer consumed. Against a sixty-day pipeline you are still ahead of the shelf. Add a fourth disrupted week in the same quarter and you are not.

Now put a price on the day itself, because dwell stays abstract until somebody invoices it. Take one forty-foot container of finished garments bound for a European distribution centre, invoiced at 80,000 dollars, roughly what a full box of mid-market woven apparel carries. Assume carrying cost at 8 per cent a year — financing, insurance, and the obsolescence you eat because fashion moves without asking — and one extra day tied up costs about 17.50 dollars. Assume the contract behind it allows 1 per cent per week against late delivery: a missed week is 800 dollars on that box, a little over 110 dollars a day once the clock starts. Assume demurrage and detention together at 60 dollars a day once free time runs out, cheap by regional standards. A day at this quay therefore costs somewhere between 90 and 190 dollars a box, depending on how late you already are.

What you set against that number decides everything. Assume 8,000 garments in the box, around 2,000 kilograms packed, sitting on a programme line that cannot slip. Flying it out at 12 dollars a kilogram costs roughly 24,000 dollars, thirty per cent of the invoice, to win back maybe four days. Re-routing through another gateway with a feeder add-on buys those same four days for around 400 to 700 dollars a box, plus eleven to fourteen days of transit you almost certainly do not have. Put the two side by side and the honest conclusion is uncomfortable: 24,000 dollars of air freight buys you somewhere between four months and nine months of further dwell. Every sourcing director I have watched reaches for the air waybill around day five. That is not a decision. It is a reflex, and it is how a four-day reaction turns into a thirty per cent write-off.

Arithmetic of a different kind belongs on the map. Two long contracts will now run side by side through one gate. Patenga was signed in December 2023 for twenty-two years, giving it a horizon near 2045. A fifteen-year concession approved in October 2026 runs to roughly 2041. Two counterparties, two expiry dates four years apart, two escalation paths, sharing one hinterland, one road corridor and one labour pool for something close to fifteen overlapping years — and one of those contracts now has a demand to reopen it attached. Nobody designing a network likes that picture, because the port no longer has a single counterparty whose interests line up with the port as a whole.

Then the crudest one. Assume generously that every remaining gateway Bangladesh owns can between them take ten per cent of what Chattogram moves in a normal month — generous, since they are smaller and further from the garment belt. Ninety per cent of the flow still has nowhere to go but wait. Dependence above 90% means the missing fallback is itself the exposure, and it is the sort of exposure that does not show up in anyone's risk register until the year it is needed.

Assume what that fallback actually costs, because diversion looks free on a slide. Mongla sits roughly 250 kilometres from the garment belt on a river approach that caps vessel size; add the road leg and a box runs about 300 to 450 dollars more than the direct Chattogram routing, and it carries seasonal draft limits and a labour calendar of its own. Payra further down the coast, and Matarbari behind it, are still closer to plans than to outlets: no regular feeder network, no hinterland road worth the name, not yet the draft they were drawn for. What genuinely moves cargo today is routing north-west through Kolkata or Haldia — three to five extra days, a land border with its own paperwork queue, roughly 500 to 900 dollars a box once road, handling and a second documentation set are counted. Colombo is not a fallback at all; it is where most of this volume already goes. None of these is instant. All of them are days slower than waiting, which is the arithmetic sitting behind that ten per cent.

So: pulse or change of shape? Today, telling you straight, this is a pulse. It carries a pulse's signature — a dated committee decision, a reaction four days later, a list of demands somebody could answer. If it resolves, what it leaves behind is a tariff schedule and nothing more, and next year people write about Chattogram in exactly the language they used last year.

Three things together would turn it into something structural. The agreement gets signed and handed over without stoppages lasting more than a few days at a time. The investment comes attached to a published calendar — quantities and dates, not adjectives. And a measurable share of the boxes that today transfer at Colombo, Singapore or Port Klang begin loading direct at deeper draft in Chittagong. Watch the third figure. It is the only one that tells you whether anything on the map has moved; the first two only tell you paperwork has moved.

Three other things would turn it into decline instead. Consecutive weeks of stoppage through the heaviest months of the shipping year. A contradiction at the centre of the port's own contracting — the review of a twenty-two-year agreement succeeding while a fifteen-year agreement is being signed, after which no future decision-maker in Dhaka believes any port contract anywhere in the country is final. And the coastal alternative slipping further behind, so no second outlet appears over the horizon while this one is argued over.

Run it the other way for a minute, because 7 October can be the top of this story as easily as the start of one. Suppose the torch procession passes quietly, somebody actually publishes the comparison of what the port earns under state operation against what automation would deliver, the twenty-two-year Patenga review gets a written timeline, and signature lands before the year is out. On those facts everything above still reads as a pulse, and it stops being a pulse at exactly one step, which is not the signing. Signed paper changes who collects the tariff. What changes what this port can physically receive is a published investment calendar with quantities and dates on it: cranes landed, automation commissioned, draft deepened, each carrying a month. That is where I would move my own reading, and nowhere earlier. If instead the deal closes on adjectives, then even a perfectly peaceful winter leaves you exactly where you are today.

Underneath all of it sits the condition that decides most of this and answers to nobody in Dhaka: whether the apparel programmes stay put. If European and American buyers shift even a tenth of their Bangladesh allocation in this sourcing round, the case for every crane in this deal weakens and the timeline stretches correspondingly. Ten years ago Chattogram was already the port everyone watched on this map with a particular kind of dread. Fifteen years from now somebody will be writing about whether October 2026 was the season the pattern changed, and they will be able to tell you from one figure: how many boxes stopped driving south to catch a ship.

What I would do this month, standing where the cargo owner stands.

Stop treating Chattogram as though it were a single input that always arrives. If more than ninety per cent of your Bangladesh volume moves over one quay and you hold no second plan, the response is not to read the news more carefully. It is to have one alternative gateway or one alternative origin pre-qualified and priced before January, even at a higher unit cost. Paying something for an option beats discovering the option never existed.

Then put the reaction interval into the booking calendar. Every announced milestone in this process — signing, tariff notification, handover — carries a precaution: no critical booking loaded inside the following seven days. Four days plus your own margin, and nothing more complicated than that.

Rewrite the assumption inside your cost model too. Any programme running three years or longer should assume higher port charges from handover and lower dwell from around year four, with somebody named as carrying the gap between them.

And watch one indicator instead of watching the headlines: the share of your Bangladesh boxes that transfer, and where they transfer. Falling means the concession is doing what it says. Rising means the port has gone indirect again and your pipeline has lengthened, whatever anyone says about it.

One piece is left over, and it is the one that separates being informed from being ready. Get the diversion worked out on paper by your own operator before you need it — alternative routing, road leg, how much spill the other gateways can actually take, cost per box and how long activation takes. Port gates do not close politely, and they do not close on a schedule you get to choose.

• Pre-qualify and price at least one alternative gateway or one alternative origin for 10% of Bangladesh volume before 31 January 2027, so the option exists before it is needed.
• Apply a seven-day no-critical-booking rule after every announced milestone in the concession — signing, tariff notification, handover — built on the four-day reaction interval observed between 1 and 5 October.
• Rewrite every Bangladesh programme of three years or more to assume new port charges from handover and dwell improvements from roughly year four, and name the party carrying the gap.
• Track monthly the share of Bangladesh boxes that transship and the port of transfer; treat a rising share as the trigger to activate alternative routing within two weeks.
• Get a written diversion plan from your own operator by 30 November 2026 covering routing, road leg, spill capacity at the other gateways, cost per box and activation lead time.