FedEx Updates Demand Surcharges for U.S. International Services

The Analysis

On 22 June 2026 FedEx updated its Demand Surcharge schedules for U.S. international services. The fees apply during periods of elevated volume, high capacity demand and increased operating costs. The carrier publishes separate export and import tables, and it reassesses both of them regularly, based on network shipment volume and capacity. That is what the announcement says, and if you read it like every other rate announcement you will file it under price increase, note the date, and move on.

I do not read it that way, and the reason is four words: it reassesses them regularly. A normal rate increase is an event with a date on it. This is not an event. This is a standing line item that is allowed to move, on a clock I do not control, against a variable I cannot see. From where I sit those are two completely different animals, and they do not belong in the same place in my budget.

Here is the buyer problem in one sentence. A fixed increase I can price. I can put it into the bill of materials, into the customer quotation, into the margin plan, and I never have to think about it again. A line that gets reassessed against somebody else's network volume cannot be priced in, because I do not know where it lands and I do not know when. So I have to hold room for it instead. And room I hold on one line is room I cannot spend on another, which means this surcharge is quietly competing with everything else in my budget for the same money.

Start by taking the thing apart, because the shape of it decides how I have to manage it. A demand surcharge is not base freight. Base freight is what I negotiate, what my discount applies to, and what I can compare across two carriers on a single page. A demand surcharge is an additional charge assessed on top, triggered by conditions in the network rather than by anything in my contract, and it moves when the network moves. That distinction sounds academic until you look at where it sits in my accounts. Base freight goes into the costed bill of materials for every SKU I sell. Surcharges, in most organisations I have worked in, go into a bucket labelled freight and other, which is another way of saying a bucket nobody has to explain.

The next piece of the shape is that the export table and the import table are published separately. On the buy side that is not a footnote, it is two different problems. Export surcharge lands in my cost of goods sold on the orders I ship out. Import surcharge lands in my landed cost on the components and finished goods I bring in. They hit different product lines, they hit different margin structures, and they are negotiated with different people on the other side of the table. Most finance teams I have sat with run inbound and outbound freight as one logistics line, and that is exactly how a change to one table gets noticed three months late, when the other table has also moved and the two movements have cancelled out on the report nobody reads.

Now the timing, and this is the part that makes it a budget problem rather than a pricing problem. The announcement says the surcharge applies during periods of elevated volume, high capacity demand and increased operating costs, and that it is reassessed based on network shipment volume and capacity. Both of those point the same direction at this time of year. Network volume goes up into the traditional peak season. Capacity demand goes up with it. So the mechanism that drives this charge is biased upward precisely in the quarter when my volume is also going up. I am not going to sit here and predict whether it rises; I do not have the network data and neither do you. What I am saying is narrower and I think more useful: I would rather build my budget on the assumption that this line drifts up than on the assumption that it stays where it is today, because one of those two mistakes is survivable and the other one is not.

This is where I part company with how most freight cost gets managed. The reflex is to take today's number and annualise it. I do not trust that number. I do not trust it because it was produced by a mechanism, not by a negotiation, and a mechanism does not care about my fiscal year. If a line is explicitly reassessed on a recurring basis, then today's value of that line is a reading, not an agreement. Writing a reading into a twelve month budget and calling it a plan is how a 4,950 dollar quarter turns into a 19,800 dollar year, and nobody can point to the week it happened.

So let me do the arithmetic properly, and let me be explicit about which numbers are real. From the announcement I have the date, 22 June 2026, the existence of separate export and import tables, the fact that both are reassessed regularly on network shipment volume and capacity, and the three conditions that bring the charge on: elevated volume, high capacity demand, increased operating costs. That is all I have. There is no dollar figure anywhere in it. So every dollar below is mine, an assumed magnitude, and I want it labelled as one, because if you take my 4.50 and repeat it as a FedEx rate you have misunderstood the whole exercise.

Assume I move 900 export shipments and 300 import shipments a month. Mine, replace with yours. Assume the export demand surcharge currently works out at 4.50 dollars a shipment on my mix and the import table at 3.20 dollars a shipment. Both are assumed magnitudes for the shape of this thing, not quoted figures. Then the baseline month is 900 times 4.50, which is 4,050 dollars, plus 300 times 3.20, which is 960 dollars, so 5,010 dollars a month. Flatlined over a quarter that is 15,030 dollars. That is the number that ends up in a budget built the usual way.

Now let the mechanism run. Assume the tables get reassessed twice inside the quarter, because the announcement says regularly and twice a quarter is my conservative reading of regularly. Assume each reassessment adds 1.50 dollars on the export table and 1.00 dollar on the import table. Mine again. Then the quarter does not look like 15,030. The first month runs at 5,010. The second month, after one reassessment, runs at 900 times 6.00 plus 300 times 4.20, which is 5,400 plus 1,260, so 6,660. The third month, after the second reassessment, runs at 900 times 7.50 plus 300 times 5.20, which is 6,750 plus 1,560, so 8,310. Add them up and the quarter costs 19,980 dollars against a budget of 15,030. The gap is 4,950 dollars, in one quarter, on a line that most companies do not even give its own code.

Work that backwards and it tells me how to budget it. If I take the baseline 5,010 a month and hold 33 percent on top of it, I get about 6,663 a month, which over three months is about 19,989 dollars. Against an actual of 19,980. So a third of headroom on this one line would have covered that quarter almost exactly. That is the number I would take to my finance director, and I would take it with the arithmetic showing, because a headroom request without arithmetic is just a bigger budget request. I would also insist the headroom sits on the surcharge line and not somewhere in general freight, otherwise it gets spent in July and I am back here in September.

Translate the gap into the language that gets a decision made. That 4,950 dollars is not a freight number, it is a margin number. Assume my gross margin on the affected product lines is 12 percent. To earn 4,950 dollars of margin back at 12 percent I have to sell another 41,250 dollars of goods. And if this repeats across four quarters, which on my assumption it does, that is roughly 19,800 dollars a year of margin that has to be recovered by selling more, on a cost that nobody in the commercial organisation caused and nobody in the commercial organisation can influence. That is the sentence I would put on the slide.

There is a second number in this that I have never once seen somebody check, and it is the one I would go after first because it is free money. Ask a single question: does my agreement discount apply to surcharges, or only to base transportation charges? Assume my negotiated discount is 32 percent, which is my number, not yours. If that 32 percent applies to base freight only and not to the demand surcharge, then on a 5,010 dollar monthly surcharge bill I am leaving about 1,603 dollars a month of discount on the table, which is 4,810 dollars a quarter. Compare that with the 4,950 dollar gap I was just complaining about. Roughly the same size, and this one is recoverable with one e-mail instead of one negotiation. Whether the discount applies is a contract question, not a network question, which makes it the one part of this whole thing I can actually win.

Then the seasonal arithmetic, which is where it stops being a nuisance. Assume my peak window runs September through November; that is my assumption, adjust it to your own calendar. Assume export volume in that window runs at 1.5 times baseline, so 1,350 shipments a month. Assume the export table has reached 7.50 dollars by then. Then the peak quarter export surcharge alone is 1,350 times 7.50 times three months, which is 30,375 dollars, against the 15,030 dollars of a baseline quarter. That is 15,345 dollars of extra cost, and it is not one thing, it is two things multiplied: the rate per shipment went up and the number of shipments went up, and they went up together. Anything that is charged per shipment and priced off network conditions has this property, and unless somebody models it that way, the peak quarter always looks like an operational failure instead of an arithmetic one.

The way the two tables behave once they reach the last mile is not mine to write about. Nadia Pryce reads that far better from the ground than I do. What I can say is only about the shape of the number in my ledger: one table moves my cost of goods sold, the other moves my landed cost, and they do not move together.

So what do I actually negotiate, given that I cannot negotiate the level? I negotiate the mechanism, and I go after four things. A notice period: I want written notice before a reassessment takes effect on my account, and my target is ten business days, because without notice I cannot re-quote and without a re-quote I absorb. A cap: I want a quarterly ceiling on the aggregate surcharge per shipment, and my opening ask is baseline plus 30 percent, which on my arithmetic is 6.50 dollars on the export table. A reopen right: if the ceiling is breached, I want 30 days to reopen pricing, the same way I would have a reopen right on any other indexed input. And applicability: my discount applies to total transportation charges, surcharges included, in writing, with the percentage stated. Notice and cap and reopen are hard to get. Applicability is not, and it is the one worth roughly 4,810 dollars a quarter.

On locking it before peak, the sequencing matters more than the number. By 31 July 2026 I want the surcharge treatment in writing, meaning the notice period, the ceiling and the reopen right, even if the ceiling is worse than I wanted. A bad ceiling is better than no ceiling, because a ceiling I can model and no ceiling I cannot. By 15 August I want my peak volume forecast shared with the carrier and the forwarder, and I am prepared to commit 70 percent of forecast peak volume to them in exchange for that written ceiling. Volume commitment is the only thing I have that they want, and the end of July is the last point at which my forecast is worth anything to anybody, because after that they already know what the network looks like.

Then the alternatives, priced. Dropping a service tier is the first one I look at. Assume 25 percent of my peak export shipments can move one tier slower, with about two extra days of transit, and assume the slower tier carries a surcharge of 2.00 dollars instead of 7.50. Then 25 percent of 1,350 is about 337 shipments, and 337 times the 5.50 dollar difference is about 1,856 dollars a month saved. The cost of that saving is two days of transit on a quarter of my peak volume, which is a commercial decision for the brand, not a procurement decision, and it has to be made with the people who own the customer promise in the room.

Pulling volume forward is the second. If I ship in August what I would otherwise ship in October, I pay the August surcharge instead of the October one. The cost is inventory carrying. Assume carrying cost runs at 18 percent a year on goods value, so a month of extra holding is about 1.5 percent of the value of the goods moved. Against 1,856 dollars a month of surcharge saving, that trade works on low value dense goods and fails on expensive ones, and I have watched people do it backwards for years: pull forward the expensive stuff because it is urgent, then discover the carrying cost ate the whole saving. Model it per SKU or do not do it. Changing the trade term is the third and slowest; moving the freight risk to the customer is a real commercial renegotiation, assume six to twelve months. A second carrier is the fourth, and I treat it as resilience rather than savings, because splitting volume usually makes both rate cards worse.

The traps, and these are the ones that cost real money. Putting surcharges into a general freight bucket is the first, because a line with no code never triggers a review. Assuming the discount reaches surcharges is the second, worth about 4,810 dollars a quarter on my numbers. Managing the export and import tables as one thing is the third, and it is how a move in one table gets hidden by a move in the other. Budgeting today's reading as if it were an agreement is the fourth, and it is the one I keep coming back to, because it is the error that produces the 19,800 dollar year. Quoting customers for ninety days on today's surcharge is the fifth, and it converts a variable cost into a fixed commitment without anybody signing anything. The sixth is ownership: if the surcharge line belongs to logistics rather than procurement, then the person who can negotiate applicability is not the person who is measured on the line, and nothing moves.

If I were signing this as the buyer, I would take a slightly worse base rate in exchange for a capped surcharge, every single time. A pretty base rate with an uncapped, network-driven surcharge behind it is not a price, it is an option I have written for somebody else, and I have signed enough of those to know how they end. What I want on the page is a surcharge line with its own budget code, thirty three percent of headroom on it, ten business days of written notice, a quarterly ceiling at baseline plus 30 percent with a thirty day reopen right if it is breached, my stated discount applying to total charges including surcharges, and a peak volume commitment of 70 percent given early enough to be worth something.

On my arithmetic that is 4,810 dollars a quarter recovered from a question nobody asked, and another 4,950 dollars a quarter that stops being a surprise. Both of those come from treating a moving line as a moving line. The carriers will keep reassessing. That is their mechanism and I do not get a vote. What I get a vote on is whether my budget knows it.

↑ Back to the key points

— By Grace Yang