Air Rates Up 26% Year on Year as Volumes Slide and Jet Fuel Doubles

The Analysis

Volumes falling while rates rise is the combination I distrust most, because it breaks the instinct that runs almost every freight budget I have ever seen: if we ship less this year, surely we pay less. This week's numbers say otherwise, and they say it loudly. Global airfreight pricing rose about 5 percent over the seven days to 5 October and now sits roughly 26 percent above where it was a year ago, while global tonnage fell about 4 percent week on week in September's last full week. That is not a demand boom. That is a cost-push market wearing a demand slowdown as a costume, and if you budget for it the wrong way round you will spend the whole fourth quarter explaining variances you could have seen coming.

The detail is worth having in your head before you do anything else. Asia Pacific outbound chargeable weight dropped 7 percent week on week, though part of that was public holidays in Japan and China rather than real weakness. Asia-Europe volumes fell 8 percent and Asia-US traffic a gentler 2 percent, with Vietnam, Malaysia, Taiwan and Singapore still growing to the US. Middle East and South Asia flows to Europe fell 6 percent and to the US 2 percent, yet flows to the US overall remained 13 percent above year-earlier levels on the strength of India and Bangladesh.

Capacity told the other story entirely: unchanged for a third consecutive week and still around 3 percent above last year. On the pricing side, Hong Kong outbound rates gained 3 percent week on week and Shanghai 2 percent, London Heathrow rebounded sharply, Frankfurt added further gains, and Chicago outbound is now more than 77 percent above its level a year ago. The fuel line explains most of it. Jet fuel has more than doubled year on year.

Put those together and you get a market where there is room for your cargo and no discount for asking nicely. Every hour of belly space an airline sells below cost hurts twice as much as it did last year, so the discipline is not going to break voluntarily. My own read is that the risk sits on the upside for shippers between now and the end of the year, and the sensible response is planning rather than hoping.

The line in that report I keep going back to is the one about operating costs: freight rates are still not fully reflecting the increase in operating costs. Carrier-side people say versions of that sentence in every cycle, and most of the time they mean "we would like more money". The reason I believe it this time is the fuel number. When jet fuel has more than doubled year on year and base rates are only 26 percent higher, the fuel cost per kilo flown has outrun the revenue per kilo flown. Airlines do not fix that by cutting price on soft lanes. They fix it by holding yield, pulling marginal freighter hours, and waiting for the peak to do the rest. Anyone waiting for attrition to bring December rates down is betting against arithmetic.

So what does this actually mean for you as an importer or exporter, rather than as a reader of freight indices? It means three specific things, and they hit three different people in your business.

Your cost model is wrong by about a quarter. Almost every product costing sheet I have reviewed in the last decade carries an air freight assumption that was negotiated once and quietly photocopied forward. If yours was set when rates were where they were a year ago, every price you quote today is understated by roughly 26 percent on the freight component. On a product where freight is fifteen percent of landed cost, that is nearly four points of margin you are giving away without knowing it.

Your commercial contracts may be worse than wrong. Anything you sold on Delivered Duty Paid or Cost Insurance Freight terms with validity running past the end of the year puts the increase squarely on you, not on your customer. Conversely, if you sold FCA or FOB, the freight increase is your customer's problem, but they will still be furious if they discover it at booking time rather than now. Both cases argue for the same action, which is telling people early.

Your instinct about availability needs recalibrating. Capacity is around 3 percent above last year and has not moved for three weeks, while tonnage is sliding. Space is obtainable. What is not obtainable is cheap space. I have already had two calls this month from people prepared to pay a panic premium for "guaranteed allocation" on lanes where their forwarder could have bought standard bookings for less with a week's notice. Do not confuse availability with scarcity, because that confusion is where real money disappears in October.

There is something buried in those regional numbers that most summaries throw away. A 26 percent global average hides very different experiences depending on where you originate. Asia-Europe volumes fell 8 percent week on week while Asia-US traffic lost only 2 percent, and Vietnam, Malaysia, Taiwan and Singapore were still growing into the US. If your supply programme sits in one of those origins rather than in mainland China, your rate behaviour and your negotiating hand are not your neighbour's, and benchmarking against their numbers will mislead you in both directions. The Middle East and South Asia read differently again: traffic to Europe fell 6 percent and to the US 2 percent, but total flows into the US remain 13 percent above year-earlier levels on the strength of India and Bangladesh. If you source out of the subcontinent, plan against a busier, firmer lane with longer booking lead times rather than assuming everybody is floating in the same soft market.

Let me now do the arithmetic, and I will label every assumption so you can run your own version this afternoon. Assume you run eight tonnes a month from Shanghai to Frankfurt, which is a very ordinary profile for a mid-size exporter of components. Assume your all-in rate a year ago, including surcharges and excluding pickup in Europe beyond the airport gate, was 3.60 dollars per kilogram of chargeable weight. That number is my working assumption, not a published figure, so swap in your own actual air waybill average. Applying the reported year-on-year increase of 26 percent puts you at about 4.54 dollars per kilo today. A year ago that month of eight thousand kilos cost you 28,800 dollars. Today the same eight thousand kilos cost about 36,288 dollars, which is 7,488 dollars more every month and close to 90,000 dollars over a year.

The uncomfortable part comes when you adjust for the fact that your own volumes probably softened too. The reported lane numbers are week on week rather than year on year, and the report attributes part of the weakness to holidays, but assume for planning purposes that your October requirement is 7,400 kilos, roughly eight percent below your old plan. Seven thousand four hundred kilos at 4.54 dollars is about 33,566 dollars. Compare that with what the same 7,400 kilos would have cost at last year's rate, which is 26,640 dollars, and you are paying about 6,926 dollars a month more to move 600 kilos less. That is a 26 percent unit cost increase showing up as a roughly 16 percent increase in the freight line, and it is exactly the pattern that makes finance teams think the freight budget is being mismanaged when it is simply being repriced.

The weekly move itself deserves its own line, because it is the one that causes arguments inside the month. The 5 percent increase in the seven days to 5 October took the notional rate from about 4.32 to about 4.54 dollars, which is 22 cents a kilo. Assume you tender roughly a quarter of your monthly volume each week, or about 1,850 kilos. One week's move cost you roughly 400 dollars on that week's tender, and if nothing changes it is about 1,600 dollars a month, or nearly 19,000 dollars across the fourth quarter. Small enough to ignore once, large enough to notice by Christmas if you ignore it eight times.

Now the levers, because complaining about the index has never reduced anybody's freight bill. I can see exactly three that move this number on shipping lanes like this one.

Fix the density. Assume one of your regular cartons measures 60 by 40 by 40 centimetres and carries 4.2 kilos of product, with a volumetric divisor of 6,000. That carton bills at sixteen kilos of volumetric weight, so you are paying for nearly four times the physical weight. Reduce the largest face by eight centimetres and you are at 12.8 kilos billable, a saving of 3.2 kilos per carton. At 4.54 dollars that is about 14.5 dollars a box. Assume you ship three hundred of those a month and you have roughly 4,356 dollars a month, which is a little over 52,000 dollars a year, for one afternoon's work with your packaging supplier.

Move part of the volume into a deferred service tier. Assume your forwarder offers a tier with roughly three extra days of transit for a twelve percent discount; both figures are assumptions, so go and get the real ones for your lane. Assume you can comfortably put forty percent of your monthly volume on that tier. That is 2,960 kilos, and a twelve percent saving on 4.54 dollars is about 54 cents a kilo, so roughly 1,611 dollars a month and a little over 19,000 dollars a year. Real money, though clearly not transformative on its own.

Buy commitment where it is cheapest. With capacity flat and demand soft, you have a reasonable hand to play. Assume you fix seventy percent of your forecast monthly chargeable weight at today's 4.54 dollars through 31 December, and leave thirty percent to float on spot. If the index prints another five percent in a single week, as it just did, your floating thirty percent costs you about 504 dollars extra that month. Contained, predictable, and much easier to explain than an unhedged position that swung the other way.

Those two operational levers together, density and deferred transit, come to about 5,967 dollars a month, or roughly 72,000 dollars a year on this example. Set that against the roughly 6,926 dollars a month the year-on-year repricing is costing you and the point is obvious: you can claw back most of the increase with work you control, but only if you actually do the work. Nothing else I know of moves this line. There is no clever routing trick, no favourable currency swing, no relationship discount that survives contact with a market up 26 percent.

Timing, then, and who owns what. This week, and I would want this done before 16 October, ask your forwarder for three things in writing. A fixed rate valid to 31 December for your committed volume. The exact fuel surcharge formula, naming the index it tracks, the lag, and how often it resets. And the all-in per kilo for your specific origin and destination pair, with pickup, customs clearance and delivery either included or explicitly excluded, because this is where the "cheap" quote and the real invoice part company. If the fuel line on your card floats weekly while your fuel budget was written monthly, assume there is currently a rolling gap of about 0.55 dollars a kilo between what you plan and what you pay, and fix that before it becomes a December argument.

By 23 October, re-baseline the cost model. Whoever owns pricing should take every product costing sheet that still carries a pre-2026 air freight assumption and replace it with a figure 26 percent higher, then walk through which quotes stop clearing their target contribution margin. I would put a hard target on this: one hundred percent of affected SKUs re-costed within ten working days, and any customer contract expiring after 31 December re-quoted by 30 November. Then, before the end of October, commit your allocation. My default position is seventy percent fixed and thirty percent spot, with a tolerance band of plus or minus ten percent written into the agreement so you are not paying for forecast you did not use on a lane whose volumes are falling.

There is also a trigger worth setting. September's weak tonnage partly reflects Golden Week, which ran from 1 to 7 October, and Japanese and Chinese public holidays in the affected weeks. The usual pattern after that kind of distortion is a catch-up wave out of Asian origins in the second half of October, which tightens local handling and gives carriers their next talking point. Assume that wave arrives around the third week of the month and commit your allocation before roughly 24 October rather than after. Set a standing rule that if the weekly index rises five percent or more in a single week, you pull one week of finished goods forward rather than waiting for it to settle.

Anyone shipping out of Chicago needs to treat this as a separate emergency. Rates more than 77 percent above a year ago means your entire US-origin export cost base has very nearly doubled. Every quote you have issued on an old cost card is now wrong, every landed-cost calculation feeding those quotes is wrong, and the longer you let them sit unconverted the more damage each acceptance does. Reprice those lanes first, this week, before you touch anything else.

One more point on changing modes, because I have watched people get this wrong in both directions. The saving from moving a consignment off air and onto ocean or a hybrid product is real on the freight line and frequently much smaller on the total landed cost line once you price in inland haulage at both ends, the extra days of carrying inventory, the additional safety stock a longer pipeline forces you to hold, and whatever your cost of capital is on goods sitting on the water. Run the whole chain before declaring victory.

Assume a 1,500 kilo consignment costing roughly 6,800 dollars by air at today's rate. Even if ocean brings that down by two thirds, twenty-five extra days in transit on cargo worth about ninety thousand dollars is working capital you have to price, and your customer will quietly expect more buffer stock from you. What I prefer in a market like this one is a split rather than a switch: fly the part of your range where shortage does real damage, sail the predictable part, and revisit that boundary monthly.

On alternatives and on the traps, here is my honest ranking. Do not book a charter or pay a panic premium unless you have a hard date you would otherwise miss; with capacity up around three percent year on year and tonnage falling, commercial belly space is there and your forwarder can get it. Test a sea-air or deferred multimodal product with a genuine pilot rather than a conversation, say 500 kilos, and judge it on measured door-to-door days and measured cost rather than on the routing diagram in the brochure. Look at consolidating less-than-container-load ocean for the portion of your volume that genuinely does not need to fly, though be careful about opportunity cost if your product shortages cascade.

Check with your insurer how declared value is assessed for air movement, since price movements in freight do not move your insurable exposure and some policies quietly ride on freight multiples. And check whether your booking comes with written confirmation of flight allocation, because "subject to space" language buried in an email is the most expensive sentence in this business during a peak.

My own position after reading this week's data is pretty clear. I do not think this pricing holds forever, and I do think the volume softness matters if it continues past November. What I am not prepared to do is assume that soft volumes will rescue my freight budget on their own, because the fuel arithmetic sitting behind that 26 percent does not support that conclusion. Plan for the expensive version, claw back what you can through density and tier selection, commit enough volume to sleep at night without committing so much that you pay for optimism you did not need, and tell your customers early. I will be logging my own all-in per kilo every Monday until Christmas, and if the curve bends sooner than I expect I will be glad to have been too careful.

↑ Back to the key points

— By Leo