DHL and FedEx tighten China–US express capacity from Sept 20, contract shippers first

The Analysis

I read the forwarder briefings this morning and the first thing that hit me was a familiar knot in the stomach. DHL and FedEx are pulling China–US express capacity behind a contract wall starting 20 September, and if you are a spot shipper moving parcels across that lane, you just lost your two easiest doors.

From where I sit, this is not some distant quarterly reshuffle. I have watched smaller importers get boxed out of peak-season air space year after year, and the timing this time looks worse than usual. My gut says anyone still buying spot on these two carriers needs to move this week, not next.

Say what you will about the big integrators, but when both DHL and FedEx quietly decide the same lane is their tightest at the same moment, that is a signal you ignore at your own cost. DHL has flat-out called China–US its most constrained corridor, and the numbers behind that are ugly: the 2026 US-bound space was booked out a full month before the mid-October 2025 peak even arrived. Let that sink in. We are talking about capacity that normally tightens in October being gone by early September. Whoever planned to ride spot rates into the fourth quarter just got cut from the roster.

I have been on the wrong side of this before. A few years back a client of mine ran a tidy little spot-dependent operation shipping consumer electronics out of Shenzhen, no contract, just calling the carrier desk every Monday for space. One October the desk stopped picking up. Rates tripled overnight and half the boxes missed the Black Friday window. The client ate the cancellations and nearly lost two retail accounts. That is the exact movie playing again, except the curtain is rising in September, a full month earlier than the disaster I watched.

Here is the mechanic of what changed. From 20 September, both carriers are running China–US on a contract-first basis. If you do not hold a signed block-space or annual deal, you are effectively at the back of a line that has no room for you. The forwarders I talk to say spot shippers are being redirected to UPS, or to a 21-day China–Europe rail-plus-truck combo that sounds clever on a slide but is a different animal in practice. UPS is not magic either; they will feel the same capacity squeeze and they will price it. And that rail-plus-truck route to Europe is genuinely a 21-day play, not a substitute for a five-day air file when your customer is screaming for stock on the other side of the Pacific.

FedEx is doing something similar on the Europe side. From late September they are limiting Europe lanes to contract clients. So the escape hatch of 'fine, I will route via Europe' is closing too, unless you already have paper with them. The pattern is clear: both integrators are protecting their contracted volumes and letting everyone else fend for themselves. If your plan was to zigzag around the tightness by hopping lanes, the lanes are all tightening at once.

Now look at the demand picture, because this is why they can do it. Asia outbound air volume climbed 18.7 percent in the first half of 2026, but capacity only grew 9.3 percent. That gap, roughly a nine-point spread, is the whole story. When volume runs almost twice as fast as capacity, the carriers hold the whip. And fuel surcharges are sitting at 43 to 48 percent on top of base rates. So even if you do find space, you are paying a surcharge that is nearly half the freight again. I have seen surcharges in the 30s before, but low-to-high 40s is the kind of number that flips a margin-thin shipment into a load you should never have booked.

Step back and think like an importer or exporter who has to keep goods moving. The risk is not just cost, it is lead time and reliability. If your China–US air file slips from five days to 'whenever we can fit you,' your inventory buffer blows out, your cash sits in transit longer, and your customer service line lights up. For a brand running lean safety stock, a two-week slip is not an inconvenience, it is a stockout. And stockouts in peak season are how you lose shelf placement for a year. The cost line on the invoice is the part everyone watches; the stockout is the part that ends relationships.

Let me put real numbers on it so this is not hand-waving. Assume you are moving 50 parcels a day on China–US express, average 8 kilos each, and your current all-air spot cost is about 6 US dollars per kilo including the fuel surcharge at the 45 percent mark. That is 50 times 8 times 6, roughly 2,400 dollars a day, call it 72,000 dollars a month. The line I would follow is to lock 80 percent of that volume to ocean or rail. Ocean on that lane might run 1.2 dollars a kilo all-in. So assume you shift 40 of those 50 daily parcels to ocean.

That is 40 times 8 times 1.2, about 384 dollars a day versus 1,920 before, saving roughly 1,536 dollars a day, around 46,000 dollars a month. The catch: ocean adds about 28 to 32 days of transit. So you are trading 46,000 dollars a month for a one-month lead-time stretch on 80 percent of volume. You cover that with forward inventory and you keep the remaining 20 percent, 10 parcels a day, on a contracted air block so your fast-moving SKUs still fly. Do the math on your own lanes, but the shape holds: move the bulk to slow modes, keep a contracted sliver for speed.

The hard part is that the 80 percent target assumes you actually have ocean or rail capacity locked, and that you are not discovering in October that the boat is full too. Asia outbound is hot across modes, so ocean is not a free pass. You still need to book early and you still need a backup carrier, because a single ocean contract with one line is its own single point of failure. I have watched companies congratulate themselves for 'solving' air with one ocean deal, then watch that one line skip a sailing and take the whole plan down with it.

That is why I keep two backup carriers minimum. One integrator contract is not resilience, it is a polite dependence. If DHL squeezes you, your UPS block better be real, and you better have a forwarder with block space on a third option. The briefings point at UPS as the spot redirect, but do not mistake 'redirected' for 'guaranteed.' Call them this week and get a number in writing, because a verbal 'we can probably fit you' is worth nothing when the deck is full.

Now the traps, because this is where people get hurt. Account backup sounds obvious until your backup carrier asks for a credit check and a two-week setup that misses your ship date. Contract terms matter: read the minimum monthly commitment, the penalty for under-utilization, and whether the rate is net of or plus the fuel surcharge, because at 43 to 48 percent that distinction is real money on every kilo. A rate that looks cheap 'plus surcharge' can be more expensive than a rate that looks high 'net of surcharge.' Know which one you signed.

Customs is the quiet killer. A rail-plus-truck file that crosses more borders than a direct air waybill will draw more inspection touches, and every touch is a day. Build that into your lead-time promise to the customer, do not discover it in transit when the phone is already ringing. I have seen a 'clever' multi-modal route save 30 percent on freight and then lose it all to a single customs hold that nobody had budgeted for in the timeline.

I would also watch the free-demurrage window on any ocean shift. When you move volume from air to ocean you are landing bigger batches less often, which means longer warehouse dwell if your receiving is not ready. A full container sitting outside free time is a fee you did not budget. Coordinate with your distribution center before you reroute, not after the box is on the water and the clock is running.

For the Europe escape hatch specifically, the 21-day rail-plus-truck combo is real but it is a plan, not a panic button. If your goods are US-bound, Europe is a detour that adds cost and a second customs entry; only use it for genuinely Europe-destined freight. Do not let a forwarder sell you a clever route for US cargo just because the China–US lane is tight. That is how a five-day shipment becomes a 30-day one with two customs bills and a smell of desperation that buyers can feel.

The opening move I give a client sitting on spot China–US air right now is this week: take stock of your lane, separate the 20 percent that must fly from the 80 percent that can sail or rail. Call your carrier rep and ask for a contract quote in writing, including the fuel surcharge basis. Open a UPS account and a second forwarder relationship before you need them. Book ocean for the October and November windows now, because those slots fill exactly when air does. Set a trigger: if your spot rate crosses, say, 7 dollars a kilo or your space confirmation slips past 72 hours, you execute the shift automatically, no debate at the desk. And keep enough cash buffer for one extra month of in-transit inventory, because the slow mode ties up working capital longer than you think.

Personally, my worry is not this September. It is that contract-first becomes the permanent shape of these lanes and spot shippers simply stop having a chair when the music stops. If you have been treating express as a tap you turn on when needed, that tap is being welded shut. Build the contract, build the backups, build the inventory plan, and do it before the mid-October peak proves me right. The carriers are not going to leave the door open for you.

A word on what a real block-space deal should look like, because a lot of people sign the first thing put in front of them and regret it in November. You want a committed volume that matches your true floor, not your hopeful peak, so you do not pay penalties for under-utilization. You want the rate pegged to a published index or at least capped on fuel, so a surcharge jump from 43 to 48 does not gut you. You want rollback rights if the carrier fails to provide space, because a contract that books your money but not your boxes is just a loan to the carrier. Read the force-majeure and the 'capacity allocation' clauses; those are where they keep the escape hatches.

On the forwarder question, do not assume your current broker has the muscle. The good ones have block space they guarded for years and they ration it to clients who bring steady volume. If your forwarder has been a casual touch-and-go relationship, this is the quarter that shows. Ask them directly how much China–US contract space they actually hold, not how much they can 'probably get.' The answer tells you whether you need a second broker, and the second broker should be one whose name you have heard on the dock, not one you found on a search result this morning.

Cash flow is the part that surprises people. When you shift 80 percent of volume from five-day air to thirty-day ocean, the goods are still yours, still paid for at the factory, but they sit in a box for a month instead of a week. That is working capital parked on the water. For a business running on thin float, one extra month of inventory in transit can be the difference between making payroll and not. This is why I tell clients to model the cash, not just the freight. The 46,000 dollars a month you save on rate is real, but if it forces a 200,000 dollar line of credit you did not have, the math flips. Talk to your financier before you reroute, not after.

Customer communication is the soft skill that saves the hard numbers. If you quietly stretch lead times and your buyer finds out from a late delivery, you have a problem that no freight saving fixes. Tell your key accounts now that you are pre-building inventory for peak and that their replenishment window should open a week earlier. Most buyers would rather hear 'I am covering you' than 'your box is late.' The carriers are squeezing everyone; a customer who knows you planned ahead stays a customer.

A smaller shipper should not think this advice is only for the big boxes. Say you move 10 parcels a day, 8 kilos each, same 6 dollars a kilo spot. That is 480 dollars a day, about 14,400 a month. Shift 80 percent, 8 parcels, to ocean at 1.2, and you drop to about 77 dollars a day on that share, saving roughly 320 a day, close to 10,000 a month. The shape is identical; only the absolute number shrinks. The mistake small shippers make is assuming they are too small to get a contract. You are not. Block-space deals come in sizes, and a 10-parcel-a-day commitment is a real number a carrier will take if you show up before the space is gone. Waiting because you are small is how you end up smallest in the queue.

On picking which SKUs earn the fast 20 percent, do not let the sales desk decide by instinct. Pull your last six months of China–US files and rank by margin times velocity: the items that carry the fattest margin and the strictest delivery promise are the ones that must stay on contracted air. A slow-moving, low-margin box that ships twice a year does not deserve a seat on an expensive block-space; it belongs on the boat where cheap is the only thing that matters. I have seen teams protect the wrong parcels because a loud account manager claimed everything was urgent, and then the truly urgent one still slipped. Protect the math, not the loudest voice.

A note on that 21-day Europe rail-plus-truck number, because the headline figure hides the real risk. Twenty-one days is the happy path when the rail runs, the truck is booked, and customs on both ends cooperates. In practice the rail leg out of China is the part that buffers, and a missed consolidation window pushes you into the next week's train, not the next day's. If you do use it for genuine Europe freight, build a ten-day cushion into the promise you give the buyer, and never quote the 21-day number as a commitment. The carriers will not eat the slip; you will. Treat it as a cheaper plan with looser guarantees, not as express with a longer label.

If you only do one thing after reading this, do the contract call this week. Not a polite email, a call where you get a number and a commitment in writing. Every peak season I watch the people who 'meant to' circle the drain while the ones who picked up the phone in September sail through October calm. The lane is not getting looser; the surcharge is not getting cheaper. The only lever you control is how early you move, and early is measured in days from now, not weeks.

One more thing on timing, because I keep coming back to it. The 2026 US space filled a month before the mid-October 2025 peak, which means the people who moved early last cycle are the people with chairs this cycle. The briefings landed 20 September. If you read this and think 'I will call next week,' next week is the week the contract desks are slammed and the spot desk is closed. The advantage goes to whoever picks up the phone today. I have said this every peak season and every peak season the laggards are the ones who call me in November asking what happened.

↑ Back to the key points

— By Leo