DHL, FedEx and UPS set weekly air express fuel surcharges for 6-12 July

The Analysis

Forty point seven five per cent. That is DHL for the week of 6 to 12 July, mainland and Hong Kong rates, and it is the number that turns a happy customer into a phone call, because it is never the number anybody quoted. FedEx is at thirty-eight point two five for the same seven days. UPS is at thirty-nine point zero. Three carriers, same week, two and a half points between the top and the bottom of that list.

Here is the thing I want you to hold onto before anything else. Those are percentages, and a percentage is only as honest as the number it is taken on. The published rates feed straight into parcel and document pricing and they move weekly. What nobody prints next to them is the base. Discounted freight or list freight. Freight only, or freight plus the other charges on the waybill. That one line decides more of your bill than the difference between the three carriers, and I am going to show you by how much.

I am working from the three rates for that week, the fact that they are updated weekly, and one extra thing the same source page carried: DHL's rate for the following week, 13 to 19 July, already published at thirty-nine point seven. Everything else in this piece is my own assumption and I will label it as such. The scenarios are generic. There is no client story here, because I am not going to invent one.

Start with where the number lands on the invoice, because that is where the argument starts. A fuel surcharge is a percentage applied to something. In express, the something is usually the transportation charge, and whether that means the transportation charge before your discount or after it changes the answer by a third. Some carriers apply it to the discounted freight. Some apply it to the full freight and then discount the result. Some take it on freight and on certain accessorials at the same time. I am not going to tell you which of the three does which, because I have not read their current tariff wording and I am not going to guess in print. What I am telling you is that this is a question with a document behind it, the answer is in writing somewhere in your agreement or in the published tariff, and until somebody in your company has read it out loud, every fuel number you quote is a guess wearing a decimal point.

Now put money on the three rates, using assumptions I am going to state rather than hide. Assume one cross-border parcel where the freight before fuel is one hundred dollars at list. Assume your contracted discount is thirty per cent, so the freight you actually pay is seventy dollars. Assume you ship a thousand parcels a week, fifty-two weeks a year. None of those come from the source; they are mine, and you should swap in your own.

On the list base, the week of 6 to 12 July gives you forty dollars seventy-five of fuel on that parcel at DHL, thirty-eight dollars twenty-five at FedEx and thirty-nine dollars at UPS. The spread between the most expensive and the cheapest is two dollars fifty per parcel. At a thousand parcels that is two thousand five hundred dollars in one week, and one hundred and thirty thousand dollars across a year, for the same box going to the same place. That is the number people argue about, and it is worth arguing about.

On the discounted base of seventy dollars, the same three rates give twenty-eight dollars fifty-three, twenty-six dollars seventy-eight and twenty-seven dollars thirty. The spread narrows to one dollar seventy-five per parcel, one thousand seven hundred and fifty dollars a week, ninety-one thousand a year. So the carrier choice is worth two fifty a parcel on one base and one seventy-five on the other.

Now look at what the base itself is worth. On the same DHL parcel, forty dollars seventy-five on the list base against twenty-eight dollars fifty-three on the discounted base is a gap of twelve dollars twenty-two per parcel. That is twelve thousand two hundred and twenty dollars a week at your volume, and about six hundred and thirty-five thousand dollars a year. Compare it with the carrier spread: twelve twenty-two against two fifty is roughly five to one. The base is worth about five times what the carrier choice is worth on this parcel. If you spend your afternoon negotiating the carrier and never ask what the percentage is applied to, you have spent the afternoon on the smaller number. That is the whole point of this piece, and it fits in one line: find the base before you argue about the rate.

Next, the weekly clock, because this is the part that bites people in the gap between a quote and a bill. These rates update weekly. The source page published the 6 to 12 July numbers, and right next to them the DHL number for 13 to 19 July, at thirty-nine point seven. That is a drop of one point zero five in a single week, and it was published before the week it applied to even started. Two things follow. One, a weekly table can move more in seven days than the entire gap between one carrier and another moves in a quarter. Two, at least for DHL on that page, next week was knowable on Monday of this week. If your quotation process assumes you cannot know the rate for a parcel shipping next Tuesday, that assumption is wrong for at least one of your carriers, and it is worth checking for all three.

Here is where the two clocks come apart. There is the moment you quote a customer, and there is the moment the carrier rates the shipment, and those are not the same moment. The rate that applies is the one in force at whatever event the carrier's tariff calls the rating event. For most express products that is acceptance, meaning the scan when the parcel is taken in, not the day you asked for a price and not the day the invoice is cut. But which scan, in which system, at which cut-off, is exactly the sort of thing that varies, and I am not going to pretend otherwise. So the instruction is narrow: get it in writing, per carrier, from your account contact, in one sentence. Which event fixes the fuel surcharge percentage. If you cannot get that sentence, treat every multi-day quote as exposed.

Let me walk one parcel through, because the exposure lives in the hours. Your sales desk quotes at ten in the morning on Monday 6 July, using the table for that week: DHL at forty point seven five. The parcel is booked that afternoon. Pickup is cut off at five and the driver scans it at six forty that evening. It reaches the gateway and leaves on the night sort. It is rated on acceptance, so the percentage that lands on the invoice is the one in force at that six forty scan, which is the week of 6 to 12 July and matches your quote. Good. Now shift the scan by eight hours, into the small hours of Tuesday 13 July, because the customer packed late and the driver came back the next day.

The rate in force is now thirty-nine point seven. You quoted forty point seven five and the bill carries thirty-nine point seven, so you have over-quoted by one dollar five per parcel on a hundred dollars of freight. At a thousand parcels that is one thousand and fifty dollars, gone in one direction, because of eight hours. Reverse the direction of the table and it is one thousand and fifty dollars you did not recover. Same eight hours. That is the whole problem in one paragraph, and it is measured in hours, not in quarters.

Multiply it across a quotation window. Assume you give customers a delivered price valid for thirty days. A thirty-day window contains about four weekly tables. If each moves by about one point zero five, which is the size of the move actually published between those two DHL weeks, the cumulative drift across the window is four point two points, or four dollars twenty per parcel on a hundred dollars of freight. At a thousand parcels a week that is four thousand two hundred dollars a week and about two hundred and eighteen thousand dollars a year sitting inside one clause of your quotation terms. Assume instead that each week moves half that, and the number halves. The point is not the figure. The point is that a thirty-day fixed price against a weekly table is four coin flips in a row, and nobody has written down who pays when they come up tails.

So build the Monday routine, and keep it small enough that it actually happens. One named person. Within two hours of the table appearing, that person records the three numbers, the week they apply to, and the movement against last week. One sheet, one row per week, four columns per carrier. If DHL, FedEx or UPS moves by half a point or more in either direction, that person sends one message to whoever issues quotes, and quotes already out for shipments that will be accepted in the new week get reissued within sixty minutes. Set the threshold at half a point and you will act roughly every other week, which is often enough to matter and rare enough that nobody starts ignoring it. Anything that needs a meeting will not survive a quarter. Keep it to a message.

Then fix the quotation itself, which is where most of the money is recovered. Put the fuel surcharge on its own line. Separate from the freight line, separate from the discount, with the percentage printed and the week it belongs to printed next to it. A quote that says freight one hundred, discount thirty per cent, fuel forty point seven five for the week of 6 to 12 July tells the customer three true things and leaves you one defensible position when the invoice arrives. A quote that says all-in delivered price tells the customer nothing and leaves you the entire gap. If your customer insists on an all-in number, then the quote needs an expiry measured in days and a sentence saying which party carries a surcharge change between acceptance of the order and acceptance of the parcel by the carrier. One sentence. It is the sentence that decides who eats the one dollar five.

There is a version of this for documents too, and it is worth thirty seconds. The published rates feed into both parcel and document pricing. A document at forty dollars of freight carries sixteen dollars thirty at DHL, fifteen dollars thirty at FedEx and fifteen dollars sixty at UPS. The spread is a dollar per document, forty cents between the two closest. Small per piece, and documents travel in batches, so at five hundred a week that is five hundred dollars a week between the extremes. If your documents go out under a contract with a fuel cap, check whether the cap is on the percentage or on the dollars, because those two behave very differently when freight moves.

Scale it down and the shape does not change, which is the useful part. Assume a small parcel where freight before fuel is twenty-five dollars instead of a hundred. The three rates give ten nineteen at DHL, nine fifty-six at FedEx and nine seventy-five at UPS. The gap between top and bottom is sixty-three cents a parcel, six hundred and thirty dollars a week at a thousand pieces, about thirty-two thousand seven hundred and sixty dollars a year. Notice what happened. The dollar gap shrank with the freight, but the percentage gap did not move at all, because a percentage of a smaller base is a smaller number hanging off the same two and a half points. So argue in points and in base, not in dollars per parcel. Dollars per parcel move every time your weight breaks move. Points and base do not.

Consolidation is the next question, and the answer is boring until it is not. A percentage is linear, so three parcels of twenty-five dollars rated separately carry the same total fuel as one waybill of seventy-five dollars rated once: thirty fifty-seven against thirty fifty-six, which is rounding. What breaks the linearity is a floor. If the tariff carries a minimum fuel charge per piece, and whether it does is something you read rather than assume, then three small pieces pay three floors and one consolidated waybill pays one, and the saving is real even though the percentage is identical. So the question for your account contact is not whether consolidating is cheaper. It is whether there is a floor under the fuel line, and how it is applied to a consolidated waybill.

When it has already gone wrong, work fast and in order. Pull the waybill and get the acceptance scan timestamp in writing, to the minute. Compare that timestamp against the week on your rate sheet, not against the invoice date. If the two disagree, raise it with the carrier with the waybill number, the timestamp and the two percentages side by side, and do it inside whatever claim window the carrier's terms set out. I am not going to put a number on that window because I have not read it. Ten minutes on the day the invoice arrives is worth an hour three weeks later, because by then the scan data is harder to pull and nobody remembers which quote the price came from. And tell the customer before they work it out for themselves. An email saying the surcharge moved one point oh five and here is the credit note is a different conversation from an invoice they have to query.

Now the traps, and they are mostly boring. One: the base. If the surcharge is taken on freight plus accessorials, then a remote-area charge or an address correction can carry fuel on top of it, and your invoice grows in a place you were not watching. Two: per shipment or per waybill. A consolidated waybill rated once is not the same as three parcels rated separately, and the difference is not rounding. Three: parcels and documents may not sit on the same table for the same carrier. Four: the mainland and China-Hong Kong designation. The published figures are for that designation; a lane booked under a different origin convention may be rated off a different table, and this is the trap most likely to cost you a month before anyone notices.

Five: the wrong week. Somebody quotes from last week's sheet because this week's was not in the shared folder yet, and the error is invisible until the credit note conversation. Six: minimums and rounding. A percentage on a very light parcel can be rounded up or floored, and over a thousand pieces the rounding is a real number.

One more thing from the same page, offered as a question rather than an answer. It also lists channels with no fuel surcharge at all: EMS global express, and one agent's dedicated line for the second half of 2026. Absence of a fuel line does not mean absence of fuel exposure. It means the exposure is either inside the base rate or absorbed by whoever is selling you the service, and those are two very different arrangements when the table moves for six weeks straight. So if you are quoted an all-in rate with no fuel line, ask what happens if the published tables rise three points by September. If the answer is nothing changes until a certain date, get the date. If the answer is nothing ever changes, ask how, and write the answer down.

Who does what, and by when. Operations owns the Monday capture and the sixty-minute reissue, and it should be one person with a named backup, because the week somebody is on leave is the week the table moves. Sales owns the quotation format, and the change is mechanical: fuel on its own line, week printed, expiry printed. Finance owns one reconciliation a week, ten minutes, taking five invoices at random and checking the percentage on the invoice against the percentage on the sheet for the week of acceptance, not the week of invoice. Any mismatch gets logged with the waybill number and raised with the carrier within five business days. Ten minutes a week is fifty hours a year, and it is the only control on this list that actually catches a carrier-side error rather than preventing your own.

Start today rather than at the next rate change. Pull the three numbers for the week of 6 to 12 July and the DHL number for the following week into one sheet before close of business. Get the rating-event sentence from each carrier in writing this week. Put the fuel line on every quote that goes out from tomorrow morning. Then, when the table moves again on Monday, the only work left is updating one row and sending one message. That is the difference between a weekly number that surprises you and a weekly number you already priced. The window on this is the week, not the month. A quote issued on Monday morning is stale by Monday fortnight, and a parcel billed at the wrong percentage is a margin you do not get back.

One last thing, because it is the one people skip. The money you lose here is not lost in the big move. It is lost in the eight-hour slip, the quote that outlived its week, the base nobody read, and the all-in price that absorbed four weekly tables in a row. None of those show up as a crisis. They show up as a margin that is quietly two points lower than it was in June and nobody can say why. Get the parcel back first, argue about the contract later.

↑ Back to the key points

— By Nadia Pryce