FedEx and DHL Express Set 5.9% 2027 Increases as US Peak Surcharges Stack Up

The Analysis

Everybody is passing around the same number: 5.9%. It is also the least useful figure in this story, because nothing about 5.9% bites until 1 January, and the money you are losing right now started moving on 26 September.

I ran regional parcel operations for years before I started writing about them, and I know this film frame by frame. A general rate increase gets announced with months of runway: finance diarises it, the pricing team re-forecasts in December, everybody feels prepared. A demand surcharge gets announced with roughly the notice of a rain squall, attaches itself to a scan event rather than to a contract date, and is fully priced into your invoices before your own rate card has heard about it.

So I am going to walk this the way I was trained to, from cut-off to delivery, one hour at a time, and at every node I will ask the same rude question: who pays for this hour, and does the piece still make the promise we sold.

Here is the slice that matters, and I will keep it short. DHL Express and FedEx have both set an average general rate increase of 5.9% for 2027, with DHL's applying to US account holders from 1 January. That is the clean, quotable part, and it is the part everybody shared. Underneath it, demand surcharges are already running: OnTrac from 26 September, UPS from 27 September, FedEx from 28 September. Later in October a second wave lands, residential charges and service charges. And fuel sits on top of all of it: UPS Ground at 30.25%, FedEx Ground at 29.75%. UPS is expected to publish its own 2027 pricing later this month, which means you can price two of your carriers today and guess at the third.

Why now, and why in that order? Because peak season is the one window in the year when a carrier can put a surcharge on the table and have the market shrug. Capacity is tight, half the industry has promised the same delivery dates to the same shoppers, and nobody re-tenders their parcel business in the last ten days of September. The general rate increase gets announced early and loudly because it needs to look planned. The demand surcharge gets announced late and quietly because it only works if you cannot redesign around it. Same quarter, opposite communication strategies, and the quiet one costs you more.

Start the clock where the money starts: the acceptance scan. Not your order time, not your manifest time, not the minute your warehouse guy prints the label. The minute the carrier's device reads it. Every surcharge in this story hangs off that one event. A parcel your system timestamps at 23:50 on 25 September and the carrier scans at 00:10 on 26 September is priced on the carrier's clock, not yours. Twenty minutes, that is the whole gap. And in a busy cut-off those twenty minutes are not the exception, they are the normal condition: the trailer gets loaded before the driver finishes the paperwork, and the scan batch goes up after the vehicle has already gone.

Then there is your own dock. If your published cut-off is 17:00 and the driver actually pulls in at 18:30 twice a week, you have been selling a same-day departure you do not have. It never shows up as a surcharge line, it shows up as twelve hours of transit you quietly stopped promising and never told anybody about. Fixing it costs nothing beyond an honest conversation with your account manager and a five-minute change to your pick-up window. It is the cheapest hour in this whole piece, and almost nobody takes it.

Six to twelve hours in, the trailer is closed and the piece is on the linehaul. Nothing about pricing changes here, but this is where you lose the cheap fix. Once a parcel is inside the network, turning a residential delivery into a pick-up at a commercial address, or splitting a multi-piece order so it does not trip an oversize rule, is no longer a data edit. It is a re-attempt, and a re-attempt costs you twenty-four to forty-eight hours and a second trip in a van. Every hour you spend arguing about whether a surcharge applies is an hour you are not spending on the piece that is still inside your four walls.

Eighteen to thirty-six hours later the parcel reaches the delivery station, and the last layer attaches: the residential charge. This is the wave due in late October, and it is the one that actually wrecks margins, for a reason that has nothing to do with logistics. Sellers lock their peak-season pages four to six weeks out, so the lock happens in mid-October, and the charge lands after it. You will be selling at a price set against a cost that no longer exists. There is no clever fix for that. There is only doing the arithmetic before the lock instead of after it, and the arithmetic takes an afternoon.

OnTrac starting on 26 September, a day before UPS and two days before FedEx, tells you something about how this market prices itself. The last-mile specialists move first because their constraint is the scarcest and their contracts are the shortest. Watch the regional carriers for the direction of travel; the globals follow with better lawyers.

One more hour deserves a name: the hour the promise breaks, which is never the hour the fee starts. If your product page says three to five business days, and you are quietly losing twelve hours at the cut-off and another twenty-four on a failed first attempt, you do not break on day five, you break on day four, and you break on the customers who live furthest out, because those are the addresses where the first attempt fails most often. Promise, surcharge and geography fail in the same place, and that place is a house at the end of a long route.

"Average" is doing enormous work in that sentence about 5.9%. A general rate increase is a weighted number across zones, services and weight breaks, so part of the book goes up more than 5.9% and part goes up less, and the distribution never gets published. I do not need to see the distribution to tell you which side you are on. Residential goes up more than commercial. Remote and extended-area delivery goes up more than metro. Oversize and additional-handling pieces go up more than anything else, and light, bulky parcels in the far zones are the worst of the lot, because the dimensional divisor is applied to the base rate and the base rate is exactly what the increase moves. If your book is mostly big cartons of light goods going to houses in low-density ZIP codes, your real increase was never 5.9%.

Twenty minutes of work settles it. Pull your last thirty days of shipments and group them by service, zone, residential flag and dimensional factor. You are looking for the buckets where the increase clears 5.9%. In most books I have seen, those buckets are also the ones carrying the volume you cannot walk away from, and that is the trap: the pieces that got expensive are the pieces your customers ordered. Dropping them costs more than eating the increase.

Let me put a number on the whole stack, with every assumption on the table where you can check it. Assume one 2 kg residential parcel, ground service, zone 5. Assume a base rate of USD 9.80 — my figure, for the sake of the arithmetic, not anyone's published tariff. Assume a peak demand surcharge of USD 1.50 per piece on residential. Assume the late-October residential charge adds USD 0.60. Assume fuel is charged as a percentage of transportation charges, base plus applicable surcharges, at the UPS Ground figure of 30.25%.

Right now, once the late-October wave has landed: base 9.80, plus demand 1.50, plus residential 0.60, gives 11.90. Fuel at 30.25% of 11.90 is 3.60. The piece costs 15.50 against a base rate of 9.80. That is 5.70 of surcharge and fuel sitting on a 9.80 rate, and none of it has anything to do with 2027.

On 1 January the base moves to 9.80 times 1.059, which is 10.38. Add the same 1.50 and 0.60 and you get 12.48. Fuel at 30.25% of 12.48 is 3.78. The piece costs 16.26.

So the general rate increase, the number in every headline, moved this parcel by 0.76. What was already on the books had moved it by 5.70. If you are negotiating over 5.9%, you are negotiating over about five percent of your problem. And look inside that 0.76: 0.58 is the rate increase, and 0.18 is fuel charging itself on the rate increase. Fuel eats the increase too. It does that every single time the base moves.

Now put the same thing on an hourly clock, because that is how the money actually leaves your account. Assume 2,000 pieces a day of this profile through an eight-hour operating window. That is 250 pieces an hour. The January step of 0.76 a piece is USD 190 an hour, USD 1,520 a day, about USD 45,600 a month on this one profile. The stack that is already on the books, the 5.70, is USD 1,425 an hour and USD 342,000 a month. Put those two numbers next to each other in the same meeting and the conversation about 5.9% ends on its own.

Here is the point I have not found in any of the coverage, and it is the one that will actually cost you money. The effective dates do not line up, and the gap is a quarter long. Demand surcharges started 26 September. The general rate increase lands 1 January. Between those two dates you are running on contracts signed at the old rate and settled against the new fees. If your agreement has a clause that carries published surcharges through to your customer, that quarter is uncomfortable and survivable. If it says rates as per schedule A and nothing else, you are funding the difference yourself for ninety-odd days, and you will find out in January when the reconciliation lands, which is the worst month of the year to discover a margin problem.

That is why the 5.9% headline is a distraction. It is a dated, announced, single event that anyone can prepare for. The ninety-day window is not an event at all, it is a condition, and conditions do not get announced. My bet, and it is a bet rather than a fact, is that most mid-sized shippers currently moving parcels into the US cannot produce, on demand, the clause in their customer contracts that governs surcharge pass-through. Go and look for yours.

Three things could prove me wrong and I want them on the record. UPS publishes later this month and comes in below 5.9%: the anchor moves, everybody re-forecasts, and my 0.76 becomes something nearer 0.50. Fuel retreats: it is the single biggest line in the arithmetic above, 3.60 out of 15.50, which is 23% of the piece, and it is the most volatile number in the stack, capable of moving either way on a month's notice. Peak volumes come in soft: carriers quietly discount the demand surcharges away in January and the re-quote you pushed through in October leaves you looking expensive next to competitors who waited.

The condition that flips the whole conclusion is the contract clause. With a pass-through clause, everything in this piece about cost is somebody else's problem and your only job is the service promise. Without one, 5.9% is irrelevant to you and 5.70 is the number that matters, and you have roughly two weeks of usable time before the late-October wave takes away your ability to re-price in time for peak.

The effect splits three ways, and it splits by hour rather than by percentage. The marketplace seller, mostly residential, gets hit at the delivery attempt, hour eighteen to thirty-six, and their exposure is a promotion page that is already locked. The B2B spare-parts shipper, mostly commercial, gets hit at the tender scan, and their exposure is a service-level agreement with a penalty attached. The high-value small-goods shipper gets hit everywhere, because their promise is the product.

If you are triaging on Monday morning, the order is not complicated. Every piece already in the network with a delivery commitment inside forty-eight hours comes first: you cannot re-price those, you can only deliver them, so go and make sure they deliver. Next are the top twenty accounts by residential volume, because that is where the late-October wave lands and those are the ones who will ask you for a credit in January. Your own rate card comes at the end of the list, and it is the only item on the list you fully control.

What I would actually do, starting today. Re-quote now rather than in December. You have two carriers priced and you can model the third inside a plausible band, so a budget built on 5.9% for all three is a budget you will rewrite in three weeks, while a budget built on a range is a budget you can defend. When UPS publishes, you change one line, not the whole model.

Run a seventy-two-hour timestamp audit. Pull the acceptance scans for 26, 27 and 28 September, per carrier, and set them against your own manifest times. You are hunting for pieces whose internal timestamp and carrier timestamp fall on opposite sides of an effective date. Each one is either a genuine charge you should have seen coming or a billing error you can dispute, and disputes have a filing window, so find them in October rather than in January.

Get your peak-season pricing locked against the right cost. Whatever the mid-October date is on your calendar, move the pricing decision ahead of it and price against the post-late-October cost, not the current one. If the residential charge makes a promotion unprofitable, kill the promotion. Killing a bad promotion is a decision you can make in an hour. Unwinding a quarter of unprofitable volume is not.

Find the clause. Every live contract, every tariff addendum, every agreed rate schedule: search it for language that carries published surcharges to the customer. Where it exists, confirm the mechanism in writing, whether it passes through automatically or needs notice. Where it does not exist, that is your October project, and the wording you want is specific: surcharges published by the carrier and applicable to the service shall be passed through at cost, effective on the carrier's publication date rather than your invoice date. One sentence, and it is the difference between a margin question and a legal one.

Re-engineer the piece before you renegotiate the rate. Most of the increase above 5.9% is attached to attributes you control: residential versus commercial destination, carton dimensions against the dimensional divisor, and whether three items go out as three parcels or one. Two centimetres off a carton can drop a piece out of an oversize band. A commercial pick-up option at checkout moves a residential piece into a cheaper bucket and, honestly, plenty of customers prefer it. Consolidating a three-piece order into one removes two residential charges. None of that needs a contract amendment, and all of it is faster than a rate negotiation.

And when the network is already jammed, triage by promise, not by value. A parcel due in twelve hours to a customer who was told two days is a bigger problem than a high-value parcel due in five days, because the first one generates a chargeback, a bad review and a phone call, and the second one generates nothing at all.

When a customer calls about a parcel that is already late, do not open with the rate increase. Open with the hour: where the piece is, what the next scan will be, and what you are doing in the next four hours. Nobody has ever been calmed down by an explanation of weighted average pricing, and everybody has been calmed down by somebody who clearly knows where their box is.

I know how this reads: too much to do, too few days to do it in. That is accurate. But the order matters more than the volume, and the order is short. Get the pieces that are already moving out the door. Get your timestamps straight. Get the clause found. Then argue about 5.9% if you still have the energy for it. Get the parcels home first — 先把包裹救回来再说. The rate argument will still be sitting there in January, and by then you will at least know what it is really worth.

↑ Back to the key points

— By Nadia Pryce