Bangkok floods cut Ruili rail-to-port links as Golden Week congestion grips Asian gateways

Two things are happening at once here and they get worse together. Bangkok's flooding is a specific, physical problem on one corridor - the ICD Lat Krabang to Laem Chabang link that carries a large share of Thailand's containerised export cargo by road and rail. Water up to 60cm in Lat Krabang, Prawet and Bang Kapi is enough to stop trucks, and it is enough to make the rail link unreliable. Separately, the pre-Golden Week surge has pushed waiting times up at every major Asian gateway, which means the vessels that would normally carry Thai cargo are themselves queuing. A delay in Bangkok does not stay in Bangkok. It becomes a missed connection at a transhipment port, and then a missed delivery window at the other end.

Let me put the numbers in a way that is useful rather than just alarming. Shanghai at 3.66 days average waiting, with some terminals at four to seven days, is the number to watch if your cargo routes through Shanghai or tranships there. Ningbo at 2.51 days, with MSICT around four. Yantian at 1.83 days but with gate-in restrictions still in force, which matters more than the average because a restriction means there are days when you physically cannot deliver a container to the terminal regardless of how early you try. Singapore at 1.6 days. Busan at 2.0 days with BNCT yard density at 85%, which is the kind of figure that precedes a stop on accepting empties or a change in free time. Port Klang at 85% to 90% yard utilisation with a deep-water berth out of service until 1 October is the one I would watch most closely, because a berth outage reduces effective capacity twice - once for the vessel that cannot berth, and again for everything stacked behind it.

What all of this means for an exporter is that your schedule reliability, not your rate, is the thing under threat right now. Rates have been firm and the market knows it. What is harder to insure against is a booking that gets rolled. And here is the specific mechanism people miss: a rolled booking does not simply depart a week later. It departs a week later into a system that is more congested than the one it missed, so the delay compounds. In the pre-holiday window, one roll often becomes two.

Let me work through a case. Assume you are an exporter shipping 30 forty-foot containers a month from Thailand or South China to Europe or the US, and assume an average cargo value of $120,000 per container, so $3.6 million of goods moving a month. Assume regulatory and customer commitments require delivery within a fixed window, and that missing it triggers either air freight at a penalty or a discount. If a shipment that should sail in week 40 slips to week 41 and then week 42 because of congestion and a Thai feeder problem, you have moved roughly $1.2 million of inventory value by 14 days. At a 6% annual carrying cost, 14 days on $1.2 million is about $2,760. That is not the real cost. The real cost is the customer conversation, and if you end up air freighting even 10% of that volume to hold a delivery date, at, say, $6.00 per kilo and 400 kilos per container, that is $24,000 for three containers - about 10 times the carrying cost of the delay. So the money is not in the delay. The money is in the emergency recovery, and avoiding the emergency is worth doing this week rather than next.

Now the Thailand-specific part, because it has a distinctive shape. Laem Chabang is the deep-sea gateway and ICD Lat Krabang is the inland container depot that feeds it. When the road link is flooded, cargo that was consolidated and customs-cleared at Lat Krabang cannot get to the vessel. When the rail link is affected, you lose the alternative. So the practical question is not whether your cargo is ready. It is whether there is a live physical path between where the box is and where the ship is, and that changes day by day. Ask your forwarder for a daily position on the Lat Krabang to Laem Chabang corridor: which roads are passable, what the truck turnaround is, and whether the rail service is running. Do not accept a general statement that things are disrupted. Ask for the specific route and the specific status.

If you are an importer buying from Thailand, your question is different and it is about allocation. If your supplier cannot move out of Lat Krabang, their goods sit and your inventory position in your own DC degrades day by day. Work out your days of cover on the SKUs that come from Thai suppliers and compare it against the realistic recovery time. Assume the corridor takes 5 to 10 days to normalise after the water recedes, because there will be a queue of trucks and a backlog of containers competing for the same road space. If your cover is under three weeks on a Thai-sourced SKU and the recovery estimate is one to two weeks, you are in a genuine risk position and you should be talking to your supplier about priority allocation today, not next Monday.

What should you actually do, in what order?

Start with visibility on your own cargo. By 1 October, every container you have moving out of South China or Southeast Asia should have a known current status: where the box physically is, whether the vessel is at berth or waiting, and what the next milestone is. If your forwarder cannot give you that for every box, escalate. You cannot make a decision on cargo you cannot see.

Next, address the Thai corridor specifically. For anything routing via Lat Krabang and Laem Chabang, ask for a written alternative: can the cargo be trucked directly to Laem Chabang from the factory, bypassing the ICD, or moved to an alternative port? The cost of doing that is real but it is finite, and it is usually small compared to the cost of a missed window. Set a decision trigger: if the corridor is still impaired on 3 October, switch to the alternative routing for anything with a hard delivery date before the end of October.

After that, protect your space. With waiting times like these, the risk is not that the rate goes up, it is that your slot is quietly reallocated to a bigger shipper. Get written confirmation of your vessel and your booking reference for every October shipment that matters, and request the roll-over policy in writing. Know in advance whether your contract gives you priority or puts you at the back of the queue, because finding out at the gate is too late.

Then deal with the empties and the free time, which is where the hidden money is. Yard density at 85% at BNCT and 85% to 90% at Port Klang means terminals will start restricting empty gate-in, and they will start enforcing free time strictly. Detention and demurrage charges in this environment are the easiest way to turn a manageable delay into a bad month. Assume the free period on your destination side will not be extended, and if you have containers sitting, either get them moving or get an agreement in writing before the clock runs out. If you have any say over your equipment terms, and I know most importers do not, ask anyway - the worst answer is no.

Now the honest part about alternatives, because there is no free move here. Switching from Thai origin to a different country is not a two-week exercise; it involves qualification, compliance and often customer approval. Switching ports inside the region, for example routing through Singapore or Port Klang rather than a congested Chinese gateway, adds transit time and cost and, given their own congestion levels, may not actually be faster. Air freighting solves the delivery date and destroys the margin - at roughly $6.00 per kilo against a containerised cost that works out to well under $1.00 per kilo, it is a 6x to 10x cost for a 4-week time saving. The only sane version of air freight is a partial one, for the units that hold a contract together, not for the shipment.

The mistakes I see most often in weeks like this, and I would avoid all of them. Not confirming the vessel name, just the sailing week. Assuming free time is negotiable at the arrival end when it usually is not. Consolidating more cargo at an ICD that is impaired, which increases the exposure rather than reducing it. And the big one: treating congestion as a rate problem when it is a reliability problem, and then discovering that the rate you fixed was never the risk.

My honest expectation is that the Asian gateways get worse before they get better, because Golden Week itself does not clear congestion, it just shifts it - the volume that did not move during the holiday arrives afterwards, into terminals that have already been running above comfortable density. And Thai corridor recovery depends on how fast the water goes down and how fast the backlog clears. I would plan for the corridor to be unreliable through mid-October, and I would plan for the network-wide waiting times to stay elevated into the second half of October. Build your October commitments on that assumption, and anything better than it is a pleasant surprise rather than a missed promise.

Let me add the cash-flow angle, because in a situation like this the freight invoice is not the number that hurts. It is the timing of the cash. Assume your terms are 30 days from bill of lading. If a shipment slips two weeks, the bill of lading date slips two weeks, which means your payment from the customer slips two weeks, which means a $1.2 million block of receivables moves out by 14 days. On a working capital facility at 6%, that is about $2,760 of interest cost, small on its own. But if it happens across a third of your October book, you are talking about several million of receivables displaced at the exact moment your own payments to suppliers are falling due. That is a liquidity problem, not a margin problem, and liquidity problems are the ones that force bad decisions - like accepting a discount to get paid early, which costs you more than the delay ever did. Worth a conversation with finance this week rather than a surprise at month end.

There is also a documentation risk that shows up in weeks like this and it is completely avoidable. When trucks are stuck and terminals are refusing gate-in, people start making informal arrangements, and informal arrangements produce clean cargo problems. If a container gets dropped somewhere other than where the bill of lading says, or if a delivery is made without the correct release, you can spend weeks unwinding it. Make sure the release process is respected even when the pressure to move boxes is high, and make sure you know who is authorised to accept a delivery on your behalf. I have seen a shipment delivered to the wrong consignee during a port disruption, and it took far longer to resolve than the disruption itself.

One more thing about making the call on air freight, because it comes up in every one of these weeks and it is almost always made too late and for the wrong volume. The right way to decide is to look at it product by product, not shipment by shipment. Ask which SKUs have a hard contractual date, which have a customer that will actually penalise you and how much, and which have a retail or promotional window that makes a late delivery worthless rather than merely late. For the first group, air freight may be rational even at several times the cost. For the third group, air freight at 6 dollars a kilo is pointless, because late is the same as never. For the middle group, the answer depends on the penalty number, and most penalty clauses are smaller than people assume before they read them. Pull your contracts out and read the actual penalty wording. In my experience half the shipments that get air freighted do not need to be, and the reason is that nobody did the arithmetic until the cargo was already at the airport.

On the Thailand side specifically, there is a question worth asking your supplier that has nothing to do with transport and everything to do with how fast they recover. Ask them where their finished goods physically are right now, and whether they hold stock at the factory, at the ICD, or at the port. A supplier who holds finished goods at the factory and has its own trucking can route around a flooded ICD in a way that a supplier who has already consolidated into a container at Lat Krabang cannot. The difference in recovery time between those two profiles is not days, it is weeks. If you are choosing between two Thai suppliers for future volume, that operational detail is now a selection criterion, and it should be scored alongside price.

Something I would also keep in view is that congestion like this is where the market quietly separates good forwarders from cheap ones. Your forwarder's value in a normal month is a rate. In a week like this it is information, an alternative plan and a phone that gets answered. If your forwarder is quoting you rates but cannot tell you which Lat Krabang roads are open or what the actual truck turnaround is, you are paying for a commodity when you need a service. And this is the week to find out, while the problem is live, so that you know who to keep on your panel for next year. Note it down while you are annoyed. It is the cheapest vendor assessment you will ever do.

To close, let me state what I think the actual risk is, stripped of the noise. The numbers are bad but they are not catastrophic, and most cargo will move, late. What changes lives in weeks like this is the small number of shipments where a two-week delay interacts with a contractual date, a customer relationship or a seasonal window, and those are the ones that should get your attention now. Thirty containers delayed is an operational event. Three containers that break a contract is a commercial one. Spend your week on the three, not on the thirty.

And keep one thing in proportion: waiting times of three to four days at Shanghai are elevated, not unprecedented, and the terminals are still working. This is a week to be organised, not a week to panic. Organised means you know where every box is, you have a written alternative for the one corridor that is genuinely broken, and you have identified the handful of shipments where a delay turns into a loss. That is a day of work. Everything beyond it is just watching.

Leo

• By 1 October, confirm the physical status and next milestone for every container moving out of South China or Southeast Asia, including vessel name and berth or waiting status.
• For Lat Krabang to Laem Chabang cargo, obtain a written alternative routing (direct trucking to Laem Chabang or an alternative port), and switch everything with a hard pre-November delivery date if the corridor is still impaired on 3 October.
• Secure written confirmation of vessel name and booking reference for every October shipment, together with the roll-over policy, rather than relying on a sailing week.
• Check days of cover on all Thai-sourced SKUs: where cover is under three weeks against an estimated one to two week recovery, contact the supplier about priority allocation today.
• Beat the yard restrictions: clear empties and import boxes before free time expires, and assume no extension at Busan, Port Klang or Yantian, where densities are already at 85% to 90%.
• Plan for the Thai corridor to be unreliable through mid-October and network waiting times to stay elevated into late October, and treat any air freight as a partial, contract-saving move only.