Transpacific air cargo jumps 30-36% YoY as tech and e-commerce fill freighters

The Analysis

Xeneta puts Northeast Asia to North America spot air freight at $5.76/kg in August, 36% above late February levels, while TAC Index shows China to US rates up 30% year on year and China to Europe up 12%. Freighters are running at roughly 65% load factor, with semiconductor, AI-hardware and cross-border e-commerce cargo taking the space, and capacity is being pulled off European lanes to serve the Pacific. Transatlantic westbound traffic is squeezed as a result.

The timing is what makes this urgent. Rates have already moved, capacity has already been reallocated, and the fourth-quarter technology and holiday build is still ahead. The window to secure October to December space at a workable number is open now and closes through September, because the carriers that allocate allotments are doing it before the peak rather than during it. Shippers holding no committed allocation are the ones exposed to offloads, rollovers and rate-on-the-day pricing.

The five briefs below address different decisions. Exporters need quote validity and a mode rule. Import e-commerce sellers need a per-unit first-leg cost and a SKU split. Factories need guaranteed space for line-stopping components. Brands need to know which delivery promises they can still keep. Procurement needs the blocked-space to spot ratio and the clauses that price it.

For Exporters

• By 12 September: sort export SKUs by value density, with air above $40/kg, a split between $15 and $40/kg, and sea below $15/kg.
• Tender 70% of October to December space as a fixed-rate allotment and keep 30% on spot by 15 September.
• Cut quote validity from 30 days to 14 days on every US and Europe lane by 22 September, with freight shown separately on FOB and CIF offers.
• Confirm in writing the space held for 1 to 21 December and name one backup carrier per lane.
• Stop quoting air below $15/kg and check volumetric weight at 167 kg per cubic metre before choosing a mode.
• Never change mode from air to sea on a letter-of-credit shipment without amending the credit first.

Xeneta puts Northeast Asia to North America spot at $5.76/kg in August, 36% above late February, and TAC Index has China to US rates up 30% year on year with China to Europe up 12%. Freighters are running at roughly 65% load factor as semiconductor, AI-hardware and e-commerce cargo take the space, and capacity has been pulled off European lanes to serve the Pacific. For an exporter this is a quote-validity problem before it is a cost problem. A price list still carrying 30-day rate validity is selling into a market that repriced 36% in six months, and the gap between your quoted rate and the market rate lands on your margin at the moment of shipment.

Make the mode decision on value density rather than on habit. Assume a monthly electronics consignment of 6 tonnes to the United States. By air at $5.76/kg that is $34,560 in air freight, or $414,720 a year. The same 6 tonnes fits one 40-foot high cube at an assumed 167 kg per cubic metre, so call it one container a month; at an assumed $3,400 per FEU all-in on the water that is $3,400 a month, about $0.57 per kg. The air premium is therefore $5.19 per kg, and what it buys is roughly 20 days. Test that against margin. A 1 kg device selling at $60 with a 35% gross margin carries $21 of margin, and $5.19 is a quarter of it. Air is the right call only where 20 days costs you more than $5.19 per kg in lost sales or penalties.

By 12 September, sort the export catalogue by value density. Above $40 per kg, book air. Between $15 and $40 per kg, split the flow and fly only the units tied to a launch or a replenishment deadline. Below $15 per kg, move to the water and stop quoting air. By 15 September, tender your October to December requirement to two carriers: 70% as a blocked-space allotment at a fixed rate and 30% left on spot, because the 30-36% rise is a spot phenomenon and the allotment is the only hedge against it. By 22 September, reissue price lists with a 14-day rate validity rather than 30 days, and state the freight component separately on every FOB or CIF offer. By 30 September, confirm in writing the space you hold for the weeks of 1 to 21 December, and name one backup carrier per lane.

Three alternatives, each with a price attached. Sea plus rail through Europe removes the air premium but adds 18 to 25 days and will not support a three-day promise. A second Chinese gateway can relieve congestion, but only if the booking stays on the same security-screened routing; a changed routing means re-screening at origin and a new cut-off. Converting the slow tail of the range to sea is usually right, provided the stock-out risk sits on lines you can afford to run short. The pitfalls are specific. Offloads and rollovers hit shippers who hold no committed allocation. Air billing runs on volumetric weight at 167 kg per cubic metre, so a light but bulky product can be charged at several times its real weight and wreck the arithmetic above. And switching a shipment from air to sea changes the document from an air waybill to a bill of lading, which breaks a letter of credit that names one of them.

For Cross-Border E-commerce

• By 11 September: rank SKUs by units per day and by margin per unit, and keep the top 20% by velocity on air.
• Keep air only where the per-unit premium is under 3% of gross margin per unit and move the rest to sea.
• Reset safety stock by 17 September: 12 days cover on air lanes and 45 days on ocean lanes.
• Load the higher inbound cost into the free-shipping threshold and subscription prices by 24 September.
• Lock January space for hero lines by 30 September so the peak is not funded by February stock-outs.
• Check volumetric weight at 167 kg per cubic metre before pricing any light, bulky SKU.

The $5.76 per kg August spot rate, 36% above late February, is a per-unit problem for an import e-commerce seller rather than a per-shipment one. Your first-leg cost is volume times weight, and the increase arrives on every replenishment without a matching change in your selling price, your marketplace commission or your free-shipping threshold. Where the margin on a low-ticket line is thin already, a 36% move on the inbound leg is enough to turn a profitable SKU into a break-even one, and the SKUs it hurts most are the light, cheap, high-frequency lines that drive your order count.

Assume 5,000 units a month at 0.4 kg each, so 2,000 kg of first-leg air cargo. At $5.76/kg that is $11,520 a month. Before the rise, the same weight at $4.24/kg, the level implied by the 36% increase, cost $8,480. The difference is $3,040 a month and $36,480 a year, and it is not divisible or negotiable per order. Per unit the inbound cost goes from $1.70 to $2.30, up $0.60 or 36%. On a $25 item running a 40% gross margin of $10, that $0.60 takes 6% of the margin. Do that arithmetic across your top 50 SKUs by volume and the split usually becomes obvious: hero lines can carry it, long-tail lines cannot.

By 11 September, rank SKUs by units per day and by margin per unit, and mark the top 20% by velocity as hero lines that stay on air through the peak. Air is defensible where the per-unit premium is under 3% of gross margin per unit; above that, test the water. By 17 September, reset cover. An air lane running seven days door to door needs about 12 days of safety stock; an ocean lane running 32 days needs about 45. The extra 33 days at an assumed 167 units a day is roughly 5,500 units, and at an assumed $9 landed cost that is about $49,600 of working capital, which is the real price of the mode change. By 24 September, factor the higher inbound cost back into the free-shipping threshold and any subscription price. By 30 September, lock January space for hero lines so that holiday sell-through is not paid for by February stock-outs.

Three routes. Split the catalogue and keep air only for hero lines, accepting a longer promise on the tail. Move the whole catalogue to sea and hold an extra 33 days of stock, which works where the SKUs are not seasonal and fails where they are. Or change sourcing so that the fastest-moving lines are produced closer to the destination, which shortens the lane rather than cutting the rate. The pitfalls are the ones that show up on the invoice. Volumetric weight charges a bulky product at 167 kg per cubic metre regardless of actual weight, so a light SKU can cost more to fly than its weight suggests. Goods bought on DDP terms hide the freight inside the unit price, which means you cannot see the 36% move or pass on any part of it. And stocking extra cover for a slower lane ties up cash precisely in the quarter when you need it for peak buying.

For Manufacturing Plants

• By 10 September: split inbound components into A-class, where a late arrival stops a line, and B-class, where it does not.
• Put A-class on a committed allotment with a named carrier by 16 September and raise plant cover from five to ten days.
• Move all B-class components to ocean by 23 September with 45 days of cover against a 30 to 35 day transit.
• Qualify a second carrier on the same lane and a second gateway for A-class by 30 September.
• Confirm fourth-quarter maintenance windows and the spares that must fly for them by 7 October.
• Keep A-class parts out of mixed consolidations and check volumetric weight at 167 kg per cubic metre.

Your exposure sits on line-stopping components rather than on finished goods. At $5.76/kg, 36% above late February, and with capacity pulled off European lanes into the Pacific, components you used to fly at a week's notice now need booking four weeks out or they will not move when the line needs them. The 30-36% increase is the visible cost. The invisible one is that fourth-quarter air capacity is being allocated before the peak, so a factory holding no committed allotment competes for whatever is left, and on a lane where semiconductor and e-commerce cargo is taking the space, what is left goes to whoever pays the spot premium on the day.

Assume 800 kg a month of controllers and connectors flown in at $5.76/kg, which is $4,608 a month. Before the rise the same weight at $4.24/kg cost $3,388, so the increase is $1,220 a month and $14,640 a year. Now put that next to a line stop. Assume one line running 1,200 units a day at $18 of contribution per unit, so $21,600 a day. Two days of stoppage is $43,200, more than 35 times the monthly rate increase. That comparison settles the strategy: on A-class components the objective is guaranteed space rather than the lowest rate, because a $5.76 rate that moves is worth more than a $5.10 rate that gets rolled.

By 10 September, split inbound components into A and B. A-class is anything whose absence stops a line and for which no substitute is qualified. By 16 September, put A-class on a committed allotment with a named carrier, hold ten days of cover at the plant rather than five, and get the allotment confirmed in writing with a stated remedy for rollovers. By 23 September, move the whole B-class list to ocean, accept 30 to 35 days of transit, and set 45 days of cover against it. By 30 September, qualify a second carrier on the same lane and a second gateway for A-class, so that a single capacity reallocation does not become a line stop. By 7 October, confirm the fourth-quarter maintenance windows and the spares that have to fly for them.

Alternatives exist but each of them costs something. A second gateway can relieve congestion at the price of a longer inland leg and a second round of security screening. Regional sourcing shortens the lane, but requalifying a supplier takes longer than the peak lasts. Pre-positioning A-class stock at a destination-side warehouse converts air freight into storage cost, which is worth doing when the volume is predictable and the parts are small. The pitfalls are technical and they are expensive. Air screening cut-offs are earlier than the flight, so a booking made late in the day does not move the same night.

Lithium batteries and some chemical components need a dangerous goods declaration, and without one the shipment is offloaded rather than delayed. Volumetric weight still governs the charge, and a light bulky sub-assembly can be billed at 167 kg per cubic metre. And a weekly consolidated booking will not hold space on a lane where the spot premium is moving; if the part is A-class, it does not belong in a mixed consolidation.

For Brand Owners

• By 12 September: map every SKU to its promise and its mode, and flag the promises built on rates that no longer exist.
• Set channel priority by 18 September: dated direct-to-consumer orders take scarce air space, wholesale orders take the water.
• Publish the revised promise per tier and brief customer service and checkout copy by 25 September.
• Get written confirmation from the 3PL by 30 September on which lanes hold committed space through December.
• Tier air to the hero line at 500 kg and move 750 kg to sea, saving $3,893 a month.
• Re-quote any forwarder rate agreed before the September move before signing it.

The lane numbers decide what you can still promise. At $5.76/kg in August, 36% above late February, with a 30-36% year-on-year rise and capacity being pulled from Europe into the Pacific, the cost of a fast promise has gone up by more than a third in six months while the delivery dates you published have not changed at all. A three-day promise on 60% of the catalogue was affordable at the February rate. At the September rate the same promise carries a freight bill that nobody re-priced, and a promise you cannot pay for is a returns problem with a marketing label on it.

Assume 2,500 units a month at 0.5 kg each, so 1,250 kg of cargo. Flying all of it at $5.76/kg is $7,200 a month; moving all of it by sea at an assumed $0.57/kg is $712, so the promise is costing $6,488 a month, or $77,856 a year. Now tier it. Assume the hero line is 40% of volume, 1,000 units and 500 kg, and keeps the three-day promise: $2,880 a month by air. The remaining 750 kg goes by sea at $427. The total is $3,307 a month against $7,200, a saving of $3,893 a month and $46,716 a year, bought with a promise extension from three days to eight on 60% of the catalogue. That is the trade stated in money rather than in service language.

By 12 September, map every SKU to its promise and its mode, and mark the ones whose promise was built on a rate that no longer exists. By 18 September, set channel priority: where air space is short, direct-to-consumer orders carrying a dated promise take it first, and wholesale orders with a lead-time tolerance take the water. By 25 September, publish the revised promise per tier, brief customer service with the exact wording, and update the checkout copy so the change is visible before the order rather than after delivery. By 30 September, get your 3PL or forwarder to state in writing which lanes hold committed space through December and which are spot, because a fulfilment proposal that prices a peak lane at a non-peak rate is a proposal that will not be honoured.

Three paths. Absorb the $6,488 a month and protect the promise, which is defensible only where the promise is what wins the order. Tier the promise, which is what the arithmetic above supports. Or move the tail to a slower but still dated service and set the customer at eight days rather than three, which costs conversion on the tail and almost nothing on the hero line. The pitfalls are the ones customers remember. A promise broken silently at the door is worse than a promise stated at eight days from the start. Assuming the old rate will come back is an assumption rather than a plan, and capacity pulled from Europe into the Pacific is not a two-week event. And a forwarder rate quoted before the September move should be re-quoted before it is signed, because the gap reappears as a surcharge after your delivery dates have already been published.

For Procurement Teams

• By 11 September: build one table of the last four quarters of lane volume, weight and rate to anchor the negotiation.
• Tender for a 70/30 blocked-space structure with a 10% shortfall tolerance and a rate valid to 31 December.
• Require a separate stated rate for 1 to 21 December and confirm it in writing by 30 September.
• Attach the surcharge schedule to the agreement by 22 September, naming what can move mid-term, on what notice and on what baseline.
• Sign only for the volume you are confident of tendering and hold the rest on spot rather than committing to a good-month number.
• Put a 30-day notice requirement on any carrier-initiated rate change and name a second carrier for peak overflow.

At $5.76/kg spot and a 30-36% year-on-year rise, the question is not whether to hedge but how much to hedge and on what terms. Blocked-space allotments for the fourth quarter are being allocated now, which means the negotiating window is September and the leverage moves to the carrier as the peak approaches. A blocked-space agreement converts a moving spot rate into a fixed one in exchange for a volume commitment. A spot-only position keeps flexibility and pays the peak premium on every kilogram. Neither is right on its own, and the ratio is the decision.

Assume 72 tonnes a year on the transpacific lane. All spot at $5.76/kg is $414,720. Now assume a 70/30 split: 50.4 tonnes on a blocked-space rate of $5.10/kg, which is $257,040, and 21.6 tonnes left on spot at $5.76/kg, which is $124,416. The total is $381,456, a saving of $33,264 a year, or 8%. The commitment is where the risk sits. If actual volume comes in 15% below the committed 50.4 tonnes, you pay for 7.56 tonnes of space you did not use; at $5.10/kg that is $38,556, more than the entire saving. So the negotiation is not about the $5.10 rate. It is about the shortfall tolerance, and 10% of tolerance is what turns this from a bet into a hedge.

By 11 September, put your last four quarters of lane volume, weight and rate into a single table so the negotiation starts from your own numbers rather than the carrier's. By 15 September, tender the fourth-quarter requirement and ask for a 70/30 structure with a 10% shortfall tolerance, a rate valid to 31 December, and a separate stated rate for the peak weeks of 1 to 21 December. By 22 September, get the surcharge schedule attached to the agreement in writing, including which surcharges can move mid-term, on what notice, and calculated on what baseline. By 26 September, sign the allotment for the volume you are confident of and hold the rest on spot rather than committing to a number you can only meet in a good month. By 30 September, put a 30-day notice requirement on any carrier-initiated rate change and name a second carrier that can take peak overflow.

The alternatives are a spot-only position, a partial charter for the largest and most predictable flow, and a shift of slower product to sea so that the air requirement shrinks before the negotiation begins. Each has a cost. Spot-only is cheapest until the peak week, when it stops being available at any price you would pay. A charter needs volume and a fixed schedule, and a partial charter for a single shipper is rarely economic. Moving product to sea cuts tonnes off the lane and is the most reliable way to improve your position, because it changes the demand you are negotiating over.

The pitfalls live in the wording. A rate without a shortfall tolerance is a take-or-pay in disguise. A surcharge list that lets the carrier revise charges on notice is a rate that is not fixed. An index-linked clause without a stated publication, a stated floor and a stated cap passes the whole move through to you. And a letter of intent signed in September to hold space in December is worth nothing unless it names the volume, the weeks and the remedy when space is not provided.

↑ Back to the key points