Maersk Opens Week 46 at $3,000/FEU, Killing the $4,000 November GRI

The Analysis

Maersk opened Week 46, 9 to 15 November, at USD 3,000 per FEU on Shanghai to Rotterdam. That is USD 200 below the week before, and Week 45 landed at USD 3,200, so this is the second consecutive cut. On the same wires, the November increases carriers filed are still sitting there: CMA CGM at USD 4,400 a FEU and Maersk at USD 4,000. Three thousand against four thousand is a gap of 25%. Against four thousand four hundred it is 31.8%. The November GRI is not delayed. It is dead, killed by the company that filed it, in the same week it filed it.

The reason on the record is short: loadings are soft late in the month, nothing has closed out, lines are trading price for volume. The reason that got one line at the bottom of the story is the one that actually sets the price. Twenty-four of the forty-six direct Asia-Europe services now route through the Red Sea rather than the Cape, and Maersk wants every loop back through the canal by March 2027. Fifty-two per cent is already back. That is not a routing statistic. It is a capacity statistic in a routing costume, and we are going to take it apart.

Show me the number, not the feeling. The number here is not the three thousand. It is the fact that four thousand is still being published by the same company on the same day, and nobody has asked which of those two prices your January contract will be measured against.

Start with the unit, because three thousand is not a price, it is an opening. Week 46 opens at USD 3,000 per FEU on Shanghai to Rotterdam, and the same report says nothing has closed out. That second clause is the important one. An opening rate is the first price a carrier puts on the board for a sailing week; it is not the clearing price. If the week has not closed out, the vessel is not full, the marginal box is still worth chasing, and the desk has room to concede. Read it that way and USD 3,000 is a ceiling on what you should pay this week, not a floor under the market. Every week I see shippers treat the opening number as the outcome and book against it, and every week the desk treats it as the opening bid.

Now the sequence. Week 45 landed at USD 3,200. Week 46 opens at USD 3,000. That is minus USD 200, or 6.25%, and it is the second consecutive cut, which means Week 44 was above 3,200. If the second cut was the same size as the first, Week 44 was around 3,400 and the two-week move is roughly 11.8%. I do not have Week 44 in front of me, so treat that as conditional: what is certain is 200 down on the week and two cuts in a row. Two cuts in a row at the end of a month is not a market drifting lower. It is a desk with a number to hit before the month closes.

Then the comparison that kills the increases. CMA CGM filed USD 4,400. Maersk filed USD 4,000. Maersk opened at USD 3,000. Against its own filing that is USD 1,000 short, a 25% miss. Against CMA CGM's filing it is USD 1,400 short, or 31.8%. CMA CGM's ask was itself 10% above Maersk's, so the two filings do not even agree with each other. Three numbers, two filings, one opening, and the opening is between a quarter and a third below both. A GRI that lands 25% below the filing is not a GRI that failed in the market. It is a GRI that never left the building.

What is the filing for, then, if nobody expects it to hold? Hold that thought; it is the part nobody printed, and it is worth more than the 3,000.

Now the Suez number, and this is where the calibre question bites. Twenty-four of forty-six direct Asia-Europe services now use the Red Sea. Twenty-four divided by forty-six is 52.2%. That figure counts services. It does not count capacity. A service is a string of port calls with vessels assigned to it, and the vessels on those forty-six strings are not the same size: the Asia-Europe fleet runs from roughly 8,000 TEU up to 24,000 TEU. Take the two extremes as a bracket, since nobody published the vessel mix. If the twenty-four loops that came back are the smaller ones, averaging say 14,000 TEU, and the twenty-two still going round the Cape are the larger ones at 20,000 TEU, the TEU-weighted share back through the canal is 336,000 over 776,000, or 43.3%. Flip the assumption and the same arithmetic gives 60.9%. The 52.2% is only correct if the two groups happen to have the same average vessel size, which is an assumption nobody stated and nobody checked.

There is a second cut to the same denominator. Forty-six direct services excludes anything moving via a transhipment hub, so the denominator is truncated before you start. And a service that has returned to the canal is not a service that returns the same weekly capacity, which brings us to the actual mechanism.

Here is why a routing decision is a capacity decision. I am going to use my own assumptions and I will say so: Shanghai to Rotterdam via Suez at roughly 30 days, via the Cape at roughly 42. A full round trip,Asia and Europe calls included, runs somewhere near 70 days through the canal and near 84 around the Cape. Fourteen days. Fourteen over seventy is 20%. The same ship, the same string, the same ports, produces about 20% more sailings a year once it stops going the long way. That is capacity created without a single newbuild, a single charter or a single order. It is the cheapest capacity injection in this market and nobody puts it in the supply forecasts, because it does not show up as tonnage.

Put the two together and the shape of the next five months appears. Take the all-Cape fleet as capacity index 1.0. Twenty-four loops back to the canal at 1.2 each and twenty-two still at 1.0 gives (28.8 plus 22) over 46, which is 1.104. So the return already completed has added about 10.4% of nominal capacity to the trade. If all forty-six come back by March 2027, the index goes to 1.2, which is 20% above the all-Cape baseline. The remaining twenty-two loops are therefore worth another 8.7%, and they are scheduled to arrive between now and March. That is five months, so roughly 4.4 loops a month, which works out at something like 1.7% of extra nominal capacity landing every month for five straight months, into a market that is heading into the weakest demand quarter of the year.

Now test the cost story against the price move, because this is where people get it backwards. Assume an 18,000 TEU ship, twelve extra days round the Cape, 140 tonnes a day of fuel at USD 600 a tonne. That is 12 times 140 times 600, or USD 1,008,000 of extra bunkers per voyage. Assume a Suez transit fee of about USD 600,000. Net saving from going back through the canal is roughly USD 408,000 a voyage. At 85% utilisation that ship is carrying about 15,300 paying boxes, so the saving works out near USD 27 a FEU. The rate fell USD 1,000 from the filing to the opening. Cost explains USD 27 of it, which is 2.7%. The other 97.3% is capacity and load factor. So stop reading the Suez return as a cost story. It is a turnover story, and turnover becomes supply.

Why this week, specifically. Late-month loadings are soft and nothing has closed out. Late in the month is when a carrier's own booking desk knows whether the ship sails full, and a week that has not closed out by the time it opens is a week the desk has to sell. Selling it means price. The calendar is unhelpful on top of that: November sits between the autumn peak and the pre-Christmas restocking, and the January contract reset is built from the weekly numbers printed in November and December. So a carrier has two reasons to chase volume now and one reason to keep a high number on the page. Both are visible in the same story.

Transmission, week by week. Week 46 at 9 to 15 November is the opening at 3,000, ten days out from today. Week 47, 16 to 22 November, is where the other lines answer; a filing that goes unanswered for a second week is a filing that has been withdrawn in practice even if it is never withdrawn on paper. Around the middle of the month the November GRI's nominal effective date passes with nothing behind it. December is where carriers try again, usually with a peak season surcharge and a blank sailing programme running alongside, and blanks are the tool that actually works when the spot market is soft. Then the January reset, and the March 2027 full return, which puts the last 8.7% of capacity into the trade.

Who this touches, and by how much, splits three ways. A shipper on pure spot sees the 3,000 immediately and it is the whole story: a USD 1,000 gap on a box is real money this week. A shipper on an annual contract running to 31 December does not see 3,000 at all this year; what they see is that 3,000 becomes the anchor their carrier's competitor quotes when the renewal arrives. A shipper on a quarterly or short-term contract resetting in December or January gets the full effect on the reset date, and the reset is where the month's work has to happen. NVOCCs and forwarders sit in the least comfortable chair of the three, because they sell fixed-rate product against a buying price that is moving, and a 25% gap between the published increase and the cleared spot is exactly the spread that eats a box reseller.

Here is the part the story did not print, and I have not seen it in the coverage anywhere else. The GRI announcement and the spot opening are not two attempts at the same thing. They are two different products aimed at two different readers. The 4,000 and the 4,400 are aimed at contract desks and at index providers that sample published tariffs. The 3,000 is aimed at the booking desk on a ship that is not full. Maersk published both in the same week and neither is a mistake. Which means the number that matters to you in January was never the 3,000.

It is whatever number your contract is indexed against, and if that contract says published tariff, or says an index that samples published tariffs, then the 4,000 is doing its job right now, because it is sitting in the sample while you are being told the market fell. A shipper who locks a January rate off a tariff-based index in a week when the clearing price is 25% below the tariff has paid for an increase that never happened, and paid it for twelve months. Ask what your index samples before you sign anything.

Run the money. Assume you move 40 FEU a month on Shanghai to Rotterdam, cargo value USD 60,000 per FEU, freight on your own account. At 3,000 the monthly freight bill is USD 120,000, which is 5.0% of cargo value. At 4,000 it is 160,000, or 6.67%. At 4,400 it is 176,000, or 7.33%. So the gap between booking this week and booking against the filing is USD 40,000 a month, or USD 480,000 a year, and against the CMA CGM number it is USD 56,000 a month, USD 672,000 a year. That is the size of the thing the headline called a failed increase. It did not fail quietly, it failed by half a million dollars on a mid-sized account.

Now the transit side, which is where the Suez return quietly pays you. Forty boxes on the water, USD 60,000 each, is USD 2,400,000 of inventory in transit. Twelve days shorter transit, at an 8% annual cost of capital, is 2,400,000 times 0.08 times 12 over 365, which is about USD 6,312 released. But that only applies to boxes on the loops that have actually come back, 52.2% of them, so on a book spread across the trade the realised figure is nearer USD 3,300 a month. The rest arrives when your particular loop returns, which is a question you can ask and should: get the service code and whether it is canal or Cape.

War risk premium moves the other way as well. If your policy is charged at 0.3% of cargo value while routing round the Cape, that is USD 180 on a USD 60,000 box; at 0.1% back on the canal routing it is USD 60. Saving of USD 120 a box, USD 4,800 a month on forty boxes. Check your own policy rate, because that one is yours and not mine.

Three ways the conclusion flips. It flips if December loadings recover and the blank sailing programme bites, because blanks remove capacity faster than a soft market cuts price, and a trade that loses 10% of its sailings in December can put 3,000 back to 3,500 in three weeks. It flips if the Red Sea situation deteriorates again and loops that came back go back round the Cape, which removes capacity instantly and is the one scenario where November's 3,000 looks like a gift in hindsight. And it flips if the full canal return lands ahead of March, because then the last 8.7% of capacity arrives in a quarter when nobody wants it. Watch two numbers to tell which world you are in: the Week 47 and Week 48 openings, and the count of blank sailings announced for December. Two more cuts plus no blanks means the floor is lower. A rebound plus a heavy blank programme means 3,000 was the bottom.

What to do about it, with dates. Do not book a full year at 3,000 on the strength of one week's opening. There is 8.7% of capacity still scheduled to land before March, and the longer you lock, the more of that landing you are locking against yourself. Match the tenor to the calendar: lock 30 to 45 days at 3,000, which carries you through late November and December without committing into the capacity build. If Week 47 opens below 2,800, extend to 60 or 90 days, because by then the second cut has confirmed the first and you are buying into a market that has another five months of supply arriving. If Week 47 and Week 48 both hold or rise while December blanks are announced, the bottom is in and you lock through the first quarter, because the blanks will do what the GRI could not.

Two traps. The first is the index trap above: check what your rate index actually samples before you agree a formula, because in a week like this one the published tariff and the clearing price are 25% apart and only one of them is in the sample. The second trap is operational and it is the one that will actually cost you a day. If your loop returns to the canal, transit drops by roughly twelve days, and most planning systems carry the old transit time for weeks after the routing changes. Boxes then arrive twelve days earlier than the system expects, free time starts twelve days earlier than the budget assumed, and the warehouse has no appointment slot. A shorter transit is good news that arrives as an unbudgeted demurrage bill if nobody updates the ETA. Ask for the service code, get the new transit in writing, and re-cut your free-time assumptions before the first canal sailing.

One more thing not to do: do not book against the 4,400 because someone forwarded you the filing. The filing is a page on a website. The 3,000 is a price someone is actually willing to write this week, and the 25% between them is the difference between a freight budget and a freight apology.

A carrier filed 4,000 and opened at 3,000 in the same week, and the market called it a failed increase. It was not a failure, it was two prices for two customers, and only one of them was for you. Show me the number, not the filing. Lock 30 to 45 days at 3,000 now, extend to 90 the moment Week 47 prints below 2,800, and lock through Q1 the week December blanks show up alongside a rebound, because that is the week the cheap capacity stops being cheap.

↑ Back to the key points

— By Vivian Zhao