Supply Chain Action Points
Read this first — the conclusion, and the moves to make:
- Ask your carrier in writing, per lane, whether this week's $480 on Shanghai-Los Angeles came off base freight or off a PSS, BAF or congestion line; if it came off surcharges, book it as deferred cost, not saved cost, before you reprice Q1.
- Price the roll risk on the six blank sailings before you take the dip: at an assumed $60,000 per FEU and 20% annual carrying cost a seven-day roll costs about $230, so at a 25% roll chance the expected cost is $58 against $480 saved on average cargo, but $1,667 against date-certain seasonal cargo.
- Set two lock lines today: two Shanghai-Los Angeles prints above $7,993.62 and you lock 60% of Q1; a print below $7,000 while blanks hold at six or more and you lock 50%, because that is where the blanking maths starts working for the carrier.
- Track three weekly numbers through December, Drewry Shanghai-Los Angeles, the announced blank-sailing count and Suez transits; if blanks fall to three or fewer and Los Angeles prints flat to up twice, treat the dip as over and lock the same week.
- Leave Asia-Europe out of this round: Shanghai-Rotterdam's $36.50 rise sits inside the index noise band, so hold your Europe repricing until that lane moves 3% or more in two consecutive weeks.
- Hold the 0.55 rate-to-volume elasticity as your own forward model: if December import volumes fall a further 8%, expect about 4.4% more off the rate, taking Shanghai-Los Angeles to roughly $7,183, and re-tender against that number rather than against the 3% headline.
Drewry's World Container Index fell 3% to $4,429 per 40ft container on 29 October, ending two weeks of gains: Shanghai-Los Angeles dropped 6% to $7,514, Shanghai-New York 5% to $9,631, while Shanghai-Rotterdam edged up 1% to $3,686. Six blank sailings are booked for next week, up from five this week, but the NRF expects US ocean import volumes to fall 11% month on month in November, and Drewry counts 62 Suez transits, the highest weekly tally of the year. It expects rates to fall again next week.
The Analysis
Drewry's World Container Index fell 3% to $4,429 per 40ft container on 29 October, ending two weeks of gains. That sentence is in every headline this week, and it is the sentence that will cost you money if you stop reading at the comma. The index is an average of eight lanes. The lane you book is one of them.
Underneath the 3% sit three numbers pointing in different directions. Shanghai to Los Angeles fell 6% to $7,514. Shanghai to New York fell 5% to $9,631. Shanghai to Rotterdam went the other way, up 1% to $3,686. We will come back to Rotterdam, because a 1% move is not a trend, it is rounding. Around them sit four more figures: six blank sailings booked for next week against five this week, an NRF forecast of US ocean import volumes down 11% month on month in November, 62 Suez transits in the week, the highest weekly tally of the year, and Drewry's own expectation that rates fall again next week.
Let us do the one calculation nobody in the headlines did. Six percent off $7,514 means the previous print was $7,993.62, so the fall is $479.62 a box. Call it $480. The composite moved $136.02. If you saw 3%, decided it was noise and did not re-tender this week, you left $343.60 a box on the table. That is the entire argument for reading the components instead of the average, and it takes eleven seconds.
The composite is eight lanes. Drewry does not print the lane weights in the weekly note, so I will use the plain average as a working assumption and say so out loud: eight lanes, equal weight, simple mean. Now take the three components everyone quoted. Los Angeles at $7,514 after a 6% fall implies $7,993.62 before, a fall of $479.62. New York at $9,631 after a 5% fall implies $10,137.89 before, a fall of $506.89. Rotterdam at $3,686 after a 1% rise implies $3,649.50 before, a rise of $36.50. Add them: minus $479.62, minus $506.89, plus $36.50, which is minus $950.01. Divide by eight, minus $118.75. Set that against the prior composite of $4,565.98 and you get minus 2.6%, which Drewry publishes as 3%.
That reconciles, and the reconciliation is the useful part. The other five lanes in the index moved about $19 between them, roughly $4 each. This is not a world event. It is a transpacific event wearing a world index as a costume. If you ship Asia to North America, the number for your model this week is 6%, or 5% if you land on the east coast. If you ship Asia to North Europe, the number is 1%, and I would not model it at all yet.
Look at the two coasts against each other while you are in there. New York at $9,631 over Los Angeles at $7,514 is a ratio of 1.282. A week earlier it was $10,137.89 over $7,993.62, a ratio of 1.268. The east coast premium did not narrow when both lanes fell, it widened by 1.4 points. If your network can take west coast plus inland rail, this week made that routing marginally more attractive than it was, and the size of the change is about $117 per box on the spread. Not a reason to rebuild a network. A reason to re-run the intermodal quote you last ran in August.
Now the base, and this is where the release is quiet in a way that matters. It says the fall ended two weeks of gains. It does not say how large the gains were. Assume the two up weeks were +4% and +5%, which is my assumption and not Drewry's number. Compound them: 1.04 x 1.05 equals 1.092. Now take this week off: 1.092 x 0.94 equals 1.0265. You are still 2.65% above where the rally started. The dip did not hand the rally back. If your budget assumed the market round-tripped, it did not, and that gap is your problem rather than Drewry's. Ask your carrier for the print from three weeks ago and run it yourself. Show me the number, not the feeling.
Next cut with the knife: base freight or surcharge. The WCI is an all-in spot number. It will not tell you whether the $480 came out of the ocean freight line or out of a peak season surcharge, a bunker adjustment or a congestion add-on, and those are not the same $480. Off the base, you have saved money for as long as the base holds. Off a PSS, you have saved nothing, you have deferred it, and the surcharge returns the week the carrier can justify it, which is usually the week your Q1 tender closes. Ask in writing, per lane. If the answer is some of each, ask for the split. A carrier that will not split the two is telling you the cut is reversible without saying so.
Why now. Supply side, two numbers pushed up and lost. Six blank sailings next week against five this week means the carriers pulled one more rotation out of the schedule. Sixty-two Suez transits, the highest weekly count of the year, means more hulls are taking the short route again and each of those hulls is producing more slot-miles a year on the same steel. Demand side, one number won: the NRF expects US ocean import volumes down 11% month on month in November.
Put them against each other and you get a working elasticity. Volume down 11%, rate down 6%, so the rate moved 0.55% for every 1% of volume. That is a soft market rather than a collapse, and it is the most portable number in the release because you can carry it forward. If December volumes fall a further 8% from November, my assumption and not NRF's, the same 0.55 says the rate gives up another 4.4%, taking Shanghai to Los Angeles from $7,514 to about $7,183. Write it down before the December forecast lands, so you can test whether the elasticity is holding instead of arguing about it afterwards.
The 62 transits need their own paragraph, because it is the one figure in the release that cannot carry the weight being put on it. Sixty-two is a count with no denominator. We are not told the weekly average for the year, so highest of the year gives a direction and no distance. We are not told ship sizes either, and in containers a count without a size distribution is close to useless: sixty-two 24,000 TEU hulls and sixty-two 4,000 TEU feeders are a three and a half fold difference in capacity and print identically in this number.
There is a real point buried in it, and it runs against the headlines. A ship back through Suez instead of round the Cape finishes its rotation sooner and delivers more slot-miles a year on the same hull. Assume a Far East to North Europe round trip of 72 days via the Cape and 60 days via Suez, both my figures: 5.07 rotations a year becomes 6.08, about 20% more effective capacity per hull. More effective capacity into Asia-Europe should push Rotterdam rates down. Rotterdam went up $36.50. So either the transits are small ships, or the released capacity is being swallowed by service changes, or the $36.50 is noise. My money is on the third, and until Rotterdam moves 3% or more two weeks running I am not repricing a single Europe contract on it.
Which gets us to the spread, and this is where the lane network view earns its keep. Shanghai to Los Angeles is $7,514. Shanghai to Rotterdam is $3,686. The Pacific lane pays 2.04 times the Europe lane. Owners move hulls for money, and at 2.04 times nobody is moving anything off the Pacific. Capacity relief for the transpacific through redeployment is not on the table this quarter. For it to come on the table the spread has to narrow to something like 1.4 times, which is my threshold rather than a published figure, and that puts Los Angeles near $5,160, about 31% below where it printed. That is the size of the hole the market has to dig before the network starts healing itself, and it is why I do not buy any argument that blank sailings alone will put a floor under the Pacific.
Here is what the carriers paid for their six cancellations, and I have not seen this in any write-up. Assume a cancelled sailing is a 14,000 TEU ship, which is 7,000 FEU of slots, running at 85% load, so 5,950 FEU. At $7,514 that is $44.7m of revenue on one sailing. Six of them is $268.2m of top line in a single week that the carriers chose not to earn. They gave up a quarter of a billion dollars of weekly revenue to defend a price level, and the price level fell 6% anyway. That is not a market being managed. That is a market where the seller has run out of things to pull.
One more cut at that number before I use it, because six is not a quantity. Six blank sailings is a count with no denominator and no allocation. Nobody tells you how many departures the lane runs, so you cannot turn six into a percentage without going and counting the schedule yourself. Nobody tells you which alliance cancelled, or on which trade, or whether any of the six are being replaced by an extra loader three days later, which happens more often than carriers admit. And nobody tells you the ship sizes, which is the same defect the Suez figure has. Six cancelled 15,000 TEU rotations and six cancelled 4,000 TEU rotations differ by a factor of nearly four in the boxes they strand.
None of that makes the number useless. It makes it a starting point you have to finish. Go and count your own carrier's advertised departures on your own lane for next week, write the denominator down, and keep it, because the denominator is the part that stays still while the six moves around. A blank count that goes six, seven, eight against a fixed forty departures is a very different market from six, seven, eight against sixty, and only one of those two markets will roll your box.
And now the part that no headline priced at all. Every write-up this week treats the six blank sailings as a footnote and the 6% as the story. They are the same story, and the blank sailings are the expensive half.
Run it. One 40ft box, Shanghai to Los Angeles, $7,514 this week against $7,993.62 last week, so $480 saved. Now the other side of the trade. Six blank sailings next week. Assume your lane runs about 50 weekly departures, my assumption, check it against your own carrier's schedule: six is 12% of the week's liftings. Assume your carrier runs four of those departures and cancels one, so a quarter of that carrier's capacity for the week disappears. Assume a booking is equally likely to sit on any of them, so a 25% chance your box rolls. Assume the roll is seven days. Assume the cargo is worth $60,000 a box and your carrying cost is 20% a year: $32.88 a day, $230 for the week. Expected cost of the roll is 0.25 x $230, which is $58. Against $480 saved, you take the dip eight to one in your favour, and I would take it too.
Change one assumption and the answer inverts. Make it date-certain cargo: a seasonal SKU, a six-week selling window, $100,000 of retail in the box at 40% gross margin. A seven-day slip takes 7 days out of 42, which is 16.7% of the window, and 16.7% of $40,000 of gross margin is $6,667. Expected cost is 0.25 x $6,667, which is $1,667. Against $480 saved you are handing over three and a half to one. That is not a close call, and nobody printed it.
So one week produces two opposite answers depending on what is in the box. The rate fell and the service degraded inside the same seven days. If your cargo has a date on it, the 6% is bait. If it does not, the 6% is free money and the blank sailings are an inconvenience you can price at $58 a box. Most commentary will tell you one thing or the other. The variable that decides it is not in Drewry's release. It is in your own selling calendar.
Who actually gets the $480 is worth one paragraph, because it is not everybody. If you are on a Q4 contract signed in May, you get none of it and you should not reprice anything on the strength of this week. If you are on an index-linked clause, you get it in the next settlement cycle automatically and there is nothing to do. If you are an NVOCC selling fixed retail space, your margin just widened by $480 a box and your customers will work that out by mid-December. If you are a beneficial cargo owner buying spot, you get it, and you get it only if you re-tender. Three of those four groups have to act. One has to sit still. Working out which one you are is worth more than reading three more articles about the index.
Grace Yang would open this from a different door: she would ask what $480 does to landed cost per unit and where that leaves your leverage at the table. I will borrow the question for one line and go back to mine. On a box holding $60,000 of goods, $480 is 0.8% of cargo value, and at 40% gross margin it is 2% of your margin on that box. Small per unit. Decisive only in aggregate, across a full quarter of volume. That is precisely why the lock decision belongs in a rule and not in a mood.
So write the rule down before the next print. Two lines. Above: if Shanghai to Los Angeles prints above $7,993.62 twice in a row, the dip was a false start and I lock 60% of Q1 at whatever is on offer, because the alternative is a market that has retaken the rally and kept going. Below: if it prints under $7,000 while the blank count holds at six or more, I lock 50%, because at that level the blanking arithmetic starts working for the carriers and the next move is a GRI announcement rather than another rate cut. Between $7,000 and $7,994 I ride the spot, but only on cargo without a date attached. On date-certain cargo I am not riding anything. I am paying for guaranteed space, and in a week with six blank sailings that premium is worth more than $480.
The other side of the call, because I do not trust a one-way argument. The NRF figure is a forecast, not an actual, and forecasts of this kind have missed by five to seven points before in both directions. Suppose November lands at minus 4% rather than minus 11%. Then six blank sailings are not a defensive gesture, they are a real floor, load factors tighten, and the 0.55 runs backwards: seven points of demand surprise against capacity already pulled is worth a mid-teens rebound inside three weeks, which is roughly $8,700 on the Los Angeles lane and a brutal month for anyone who sat waiting for $6,800.
Two tripwires tell you which world you are in, and both publish weekly. One: the blank count falling from six to three or fewer. Carriers see bookings before anyone else does and they stop cancelling the week the bookings turn. Two: Shanghai to Los Angeles printing flat to up for two weeks running. If both fire, the dip is over and you lock that day, not the following Monday, because the following Monday is when everyone else works it out.
Drewry expects rates to fall again next week and I have no argument with it. The elasticity, the demand forecast and the sheer weight of a fleet this size all point the same way. What I do not accept is that a 3% move in an eight-lane average tells you what to do. Your lane moved 6%. Five other lanes did not move at all. Rotterdam moved $36.50, which is a rounding error in a decent suit. Six sailings came out of next week's schedule, and on the wrong cargo that costs more than the rate fall saves.
The market handed out a discount and took back a slice of the schedule in the same envelope. Take the discount on cargo that can wait a week. On cargo that cannot, the discount is the hook. And if Los Angeles prints above $7,994 twice, stop reading indices and sign something, because the rate sheet is never going to apologise to you later.
↑ Back to the key points