Pre-Tender GRIs Lift Asia-North Europe 6% to $3,442/FEU as 2027 Contract Talks Open

The Analysis

Freightos put its weekly numbers out on 20 October and the headline wrote itself: Asia to North Europe spot freight up 6 per cent in a week, closing at 3,442 dollars a FEU. Asia to the Mediterranean went the same direction at half the pace, up 3 per cent to 3,927 dollars. The update itself calls it the first decent bounce the lane has managed in months.

The timing is the part I care about, and the part nobody put in a headline. That same week, the 2027 annual contract round opened on the Asia-Europe headhaul, and carriers pushed their general rate increases into exactly that window. Nobody files an increase for the week after tenders close. It has to be standing on the table while the paperwork is open, because what everyone negotiates around is whatever number is on the screen that week. This is not a recovery that happens to coincide with tender season. It is tender season, priced as a recovery. Sentiment content running high. Arithmetic content still to be established.

Three jobs for this piece. We take the 6 per cent apart and count how many dollars it actually is. We put all four lanes side by side, because only one of them got a headline. And we locate the number that really gets written into a 2027 contract, which is not the percentage anybody printed on Monday.

Start with the base, because percentages hide it. North Europe finished at 3,442 after rising 6 per cent, so the previous print sat around 3,247: divide 3,442 by 1.06 and you get 3,247.17. The move the whole trade spent Monday arguing about is 195 dollars on a forty-foot box. That is the entire rebound, expressed in the only unit a contract understands.

Run the same arithmetic on the Mediterranean. 3,927 divided by 1.03 is 3,812.62, a gap of about 114 dollars. Half the percentage on a lane that was already four hundred dollars dearer, delivering roughly sixty per cent of the move. Nothing surprising there, but it does tell you how much of a headline is an artefact of which lane it landed on.

Now the number nobody read carefully. Transatlantic rose 200 dollars to 2,850, and the coverage files that as only 200 dollars. Only. Two hundred on a base of 2,650 is 7.5 per cent. That is a larger move than the 6 per cent that carried the headline, and it is a larger move in dollars too: 200 against 195. The lane described as barely moving actually travelled further than the lane everyone is watching, on both measures at once. This number has a problem, and the problem is one small adverb.

Before anyone builds a thesis on cross-lane comparisons, one honest qualifier. These are index readings on different baskets, with different port pairs and different services behind them, published by Freightos on each lane's own terms. Comparing them is comparing levels, not identical cargo. What you can compare without any apology is money. This week a North Europe shipper paid 195 dollars more per box and a transatlantic shipper paid 200 dollars more, and both of those are real dollars on real invoices.

The edges of the measurement are worth knowing because they cost nothing to learn. A spot index reads what was actually transacted, or what holding carriers quoted and the market accepted, for cargo loading inside that week. It is not the whole market, it is not every service on the trade, it is not your account, and it is emphatically not the specific price one carrier gave one customer with a volume commitment behind it. Two buyers can stare at the same 3,442 and be looking at quite different quotes. None of that makes the index worthless. It makes it a ruler: good for direction of travel, and good for the arithmetic you build on top of it. If you are about to write an index into a contract clause, read the methodology note first, because you are signing that methodology and not merely the number.

Four lanes going four different directions in the same week is a shape, and shapes carry information. The Pacific needed no rescue: it was already sitting at 8,126 dollars to the US West Coast and 9,700 to the East Coast, which is 2.36 and 2.82 times the North Europe print. Nobody announced anything there because there was nothing worth announcing. That multiple matters more than any of this week's percentages, because every operator deciding where to put a ship this winter looks at the Pacific first. And the lane that took the headline was the one with a 2027 tender attached to it.

Split the shippers instead of the lanes and one number turns into four arrival dates. Shippers with everything in the spot market on Asia-North Europe pay it this week. There is no buffer and no argument left to win; their bill moved 195 dollars a box and that is the end of it. Anyone holding a deal that runs to 31 December pays nothing this year at all, and their exposure begins on 1 January, decided by a tender they are about to run, with this week's print reaching them only through the anchor. Shippers on index-linked paper get it passed through mechanically at whatever speed their settlement window specifies: monthly reset means the 195 arrives inside the month, quarterly reset means it arrives in January, by which point the number underneath may already be somewhere else. And there is a fourth group, whose tender closed before 20 October, that pays nothing this cycle whatever happens, because their 2027 was priced off a day nobody is writing about.

Which brings me to the question I ask before I believe any comparison at all: what is inside the number? FBX is an all-in spot reading, base freight plus whatever surcharges sat in that week's quote, on a defined port pair. Your 2027 contract will almost certainly not be built that way. It will carry a base freight line, a bunker formula, a peak season trigger, a low sulphur line, perhaps a currency adjustment clause, each of them free to move independently of the others. Put the two side by side without aligning units and you walk straight into the oldest mistake in this trade: a contract that looks ten per cent cheaper against the index and costs more by June, because the base came down while the bunker line went up. The workload here is dull and it pays. Ask every carrier for two things in writing: the tender offer split into base plus each named surcharge, and the spot quote split exactly the same way. Then you are comparing the same animal.

The surcharge split has a second trap, and it runs in the direction nobody watches. Base freight takes all the attention while the bunker line quietly decides the outcome. Picture the obvious victory: you push the base all the way back to 3,247, reclaim the full 195 dollars, and call it a result. Then the fuel clause nobody capped, because the base negotiation ate six weeks, adds it back across the year, and the contract you won is worth precisely what the previous one was. I am not going to attach a number to what the fuel line does, because there is none in this week's data. Count it yourself, cap it on paper, assume nothing.

Contract calendars do not run on one clock, and yours decides how much of this week reaches you. Most of the trade works to 1 January, which puts tender close somewhere in the next six or seven weeks and makes this print directly relevant. A share of European retail buying runs May to April, and for those buyers October is an early warning rather than a decision point. Some contracts reset quarterly. Before doing anything else with these four numbers, find the date your own tender closes and write it next to the 3,442, because every argument in this piece hangs off the gap between those two.

And now the sentence nobody printed, which is the one that costs money. Here it is, stated flatly: the 6 per cent does not enter your 2027 contract. Nothing called a percentage has ever entered a contract. What enters it is the level standing on the day your tender closes, and the purpose of filing GRIs in the week the tender opens is not to reach 3,442, an achievement that took seven days and 195 dollars, but to keep 3,442 standing for the four to eight weeks it takes to close. Publishing an increase costs nothing. Holding one through a tender cycle costs exactly the discount you decline to give, which is why these things reliably appear in October and rarely surface in February.

Look at the week again through that lens and it changes character. Something that moved your freight bill by 195 dollars and then had to stand still for six weeks is not a market event. It is a negotiating position with a price index attached. The headline asks whether you believe in the rebound. The only question that touches your cost is whether the 195 is still there in the week you sign.

Notice who carries the risk in that arrangement. If 3,442 survives to the closing week, every shipper running a 2027 tender is anchoring against a level manufactured inside one week. If it leaks away, nothing happens to whoever announced it. The index goes back down, the same carriers file again next month, and the cost of being wrong lands entirely on the buyer who signed early. Asymmetry is the polite word.

The forwarders quoted around this story doubt the level will hold, and both of their reasons are structural rather than emotional. Demand into Europe is soft. And the very same Freightos update that printed the rebound runs under a headline about Red Sea container traffic being set for full return in the first quarter. Take the second one seriously, because it lands on precisely the lane in question. Sailing round the Cape absorbs tonnage: a long detour needs more ships to deliver the same weekly service. Coming back through Suez releases that tonnage again, and it does not spread evenly across the network, it returns to the Asia-Europe headhaul those ships were working on. Freed capacity arrives on the same lane where somebody is trying to keep a 3,442 anchor upright through tender season. That is not a coincidence either. That is arithmetic working against the increase, one week at a time, in the one place where it matters.

There is a force running the other way, and it deserves equal weight, because I have no interest in telling you the number must fall simply because a forwarder said it might. Tonnage follows revenue per slot. US East Coast at 9,700 is 2.82 times North Europe at 3,442, and US West Coast at 8,126 is 2.36 times. Every hull kept on the Pacific is a hull not pressing against the European anchor. This is how a headhaul GRI can survive genuinely soft European demand: it does not need European demand, it needs the Pacific to stay more attractive than Europe for another eight weeks.

Layer the timing on top and it gets sharper. If full Red Sea transits do come back inside the first quarter, the released capacity arrives just after most 2027 contracts have started running. The dilution lands after signature, which means the twelve month fix you signed at 3,442 is locked in while the spot market underneath it begins to soften. That is the sequence worth holding in your head while you are being asked to commit.

So define conditions instead of picking a side. Start with the branch where it holds. Assume Asia-North Europe prints at or above 3,442 in the week your tender closes. The increase worked, whatever the demand picture says, and you should plan for carriers opening 2027 at that level and conceding on the surcharge lines rather than on base. Buying twelve months near 3,442 in that world is buying certainty at the top of a manufactured range, and it is still cheaper than paying an index for twelve months from a high starting point, provided the Pacific stays warmer than Europe. The other branch is the one where it does not hold.

Assume the lane prints below 3,247, the level it started from, in two consecutive weeks before you close. The GRI has then been absorbed rather than accepted, every argument for fixing collapses, and you switch the ask to index linkage with a collar and let them prove the level. In that branch the worst possible outcome is paying the market, and the market is by definition lower than the announcement. Between those two lines there is no decision available, and manufacturing one anyway is exactly how people sign in the wrong week.

Let us put dollars on it, with every assumption written down so anyone can argue with it. Assume an importer moving 40 FEU a month on Asia-North Europe, which is 480 boxes a year, all currently on spot or tracking the contract index. Assume the increase holds and becomes the base for a calendar 2027 deal. Four hundred and eighty boxes multiplied by 195 dollars is 93,600 dollars a year. Monthly, forty boxes at 195 is 7,800 dollars. Full freight spend at 3,442 across that volume is 1,652,160 dollars for the year, and the same 480 boxes at 3,247 is 1,558,560. Ninety-three thousand six hundred dollars, six per cent, one week's work, arguing about a number nobody had heard of on Friday.

Now the version most people actually live in. Assume the same importer commits 70 per cent of volume to an annual contract and keeps 30 per cent in the spot market. That puts 336 boxes under contract, and 336 times 195 is 65,520 dollars. The remaining 144 boxes ride whatever the index does across twelve months. Splitting the volume does not abolish the money; it only changes whose guess it is, yours or the market's, and does it for twenty eight thousand dollars of exposure rather than ninety three thousand. Assume 20 FEU a month to the Mediterranean, 240 boxes a year, and that 114 dollar delta is 27,360 dollars. Assume 15 FEU a month across the Atlantic, 180 boxes a year, and the 200 dollar move nobody featured is 36,000 dollars. Stack the three and you get 156,960 dollars of annual exposure created in a single week, with 63,360 dollars of it sitting on lanes that did not get a headline.

All of that lands somewhere concrete, and it is not the carrier's ledger. Grace Yang has spent her career pricing this from the buying side, and the question she would ask — how much of this actually reaches my landed cost, and how much of it can I simply refuse to recognise — is the right one to put once the 195 dollars has been counted. From my side of the desk the answer is mechanical: it reaches whoever signs a fixed rate in the week the number is highest, and it reaches whoever let a bunker formula run without a written cap. Everyone else in that room is a spectator.

Where is the line you cross and stop arguing? Mine is built out of these four prints and nothing else. If Asia-North Europe is sitting at or above 3,442 in the week your tender closes, I stop trying to win the discussion and start managing the exposure: a twelve month deal, at least 70 per cent of volume committed, the surcharge lines capped, and the bunker formula written into the contract rather than referenced. If it has come back below 3,247 in two consecutive prints before closing, I invert the whole approach and ask for index linkage with a collar set between 3,247 and 3,442. In the middle of those two lines I do nothing, because waiting is free and the pointer that actually decides this is the closing print, not today's percentage.

The mistakes on offer here are dull ones, which is why they get made. Comparing an all-in index against your base freight line and concluding you are overpaying when you are paying exactly the going rate. Signing a twelve month fix the week after a GRI and finding out in week five that nobody ever transacted at it. Ignoring the Mediterranean because the headline said North Europe, while 114 dollars a box runs through 240 boxes a year. Tendering Europe and the Atlantic as one package and letting a concession on the lane nobody measures pay for a harder line on the lane everybody does. None of these requires poor judgement. They require one careless spreadsheet.

One last thing, for anyone minded to read this as a turning point. A lane that needed one week and 195 dollars to change direction can need one week and 195 dollars to change back, and the tonnage coming out through Suez is already arguing with the rate programme coming out through a press release. The sentiment content of this story is high. The arithmetic content is a hundred and ninety five dollars standing on a screen, waiting to find out whether anyone is still watching in six weeks. If the closing print is still there, sign the fixed rate that week. If it has come back under 3,247 twice, ask for the index. Those two are the only instructions this week's numbers will support.

↑ Back to the key points

— By Vivian Zhao