Rates No Longer Follow Supply and Demand: How Carriers Rewrite Pricing Logic with Blank Sailings

The Analysis

The number that should have broken this market did not break it. Capacity came in 8 to 10 percent higher this year. Demand grew 3 to 4 percent. Put those two figures through the pricing logic we all learned, the one where more ships than cargo pushes price down until the surplus clears, and freight should have gone through the floor. It did not. Spot held. Contract held. In several lanes the rate came back up.

So one of three things is true. The demand figure is wrong, the supply figure is wrong, or the way the rate gets made has changed. I have spent enough mornings reading blank-sailing notices to think it is the third.

What follows is an attempt to price that change, because it decides what a shipper should buy and when to sign. Carriers did not discover a new economics. They built a routine of cancellations, slow steaming and rerouting that lets them decide, week by week, how much capacity exists. The rate you book is an output of that decision, not of how much cargo is in the market. Most procurement templates were written for the other market, and the gap between the two is where money disappears quietly.

The arithmetic is where the story gets uncomfortable. Supply up 8 to 10 percent, demand up 3 to 4 percent, and the difference is four to seven points of surplus dropped onto the water inside twelve months. A market that clears should fall until enough tonnage is idled or scrapped to close that gap. What happened instead is a flat line and, in several weeks, a rise. A market that absorbs eight percent more capacity without falling is not clearing. Something is holding the price. That something has a schedule, and the schedule is published.

From February 9 to March 15 this year, carriers announced 125 blank sailings. Those cancellations removed roughly 18 percent of the voyages on the main east-west services. Transpacific eastbound carried 63 percent of the cancellations. Read the three together and the operating logic shows. A hundred and twenty-five is not a reaction to an emergency; it is a programme with a calendar. Eighteen percent is not a trim; it is very nearly the exact size of the gap between a falling market and a flat one. Sixty-three percent in one direction tells you where the discipline was aimed: the headhaul, where the contract rate sits and the revenue is, not the backhaul where neither does.

There is a second lever that never shows up in a cancellation tally. Speed cuts of 5 to 10 percent absorbed another 5 to 8 percent of capacity. A ship steaming slower carries the same boxes over more days, so the same fleet produces fewer voyages a year. Nobody announces slow steaming as a capacity cut, because on paper it is not one. In the water it is. Add the two levers and a fleet that looks oversized on the orderbook looks close to matched on the sailing calendar.

Network design pulls the same lever with a longer arm. Asia-Europe services have been routed the long way, around the Cape of Good Hope, as a structural decision rather than a weekly one. A longer rotation eats capacity inside the voyage. Same ship, same crew, same daily fuel bill, fewer round trips a year, therefore fewer slots to sell. A carrier that reroutes is not withholding capacity. It is consuming it, and it can keep consuming it for as long as the route stays long.

What makes this a change rather than a phase is that the tools stopped being improvised. Blank sailings used to be what a carrier did when a crisis hit and it needed to stop the bleeding for a month. What the February-to-March window shows is a cancellation schedule issued in advance, spread across a defined set of services, sized to a target. That is management in the ordinary sense of the word: a plan, a cadence, a review. Carriers moved from taking the price the market gave them to setting the quantity that produces the price. The floor is not where the market put it. The floor is where they are defending it, and defending is an active verb.

Here is where I stop trusting the summary numbers. Take the five figures that get quoted inside the same paragraph. Supply growth 8 to 10 percent. Demand growth 3 to 4 percent. Slow steaming absorbing 5 to 8 percent. Diversions absorbing about 10 percent. Effective overcapacity of 3.5 to 4 percent. Those cannot all be true on one denominator. Supply minus demand is four to seven points. Slow steaming alone claims five to eight. Diversions claim ten. Add the absorbers and you have taken out fifteen to eighteen points against a four-to-seven-point surplus, which would leave the market short of ships, not three points long. At least three of those figures are therefore measured against different baselines: gross fleet against active fleet, teu-miles against teu, announced against realised.

My reading is that capacity growth is gross of scrapping and idling, and that the absorption percentages are quoted against nominal deployed capacity rather than against last year's effective supply. This matters for a procurement decision, because it means you cannot subtract one number from another and get a forecast. Any model telling you that a tenth of capacity is parked in the Red Sea and therefore rates must fall fifteen percent on reopening is doing arithmetic across two rulers. Trust the direction. Do not trust a point estimate assembled by adding percentages.

The Red Sea is the biggest single variable and it behaves backwards. Diversions absorb something around a tenth of effective capacity, and that absorption holds Asia-Europe up harder than any commercial decision the carriers have taken. Bad news in the Red Sea is good for the rate. Good news is the risk. If Suez reopens and the diversions unwind, the capacity parked on the long route returns to the short one, and the overcapacity figure now at 3.5 to 4 percent is forecast to rebound to 14 or 15 percent. That is not a drift inside a range. That is a different market with the same ships. The trigger is a security condition rather than a commercial one, and it can move in a week.

Which is why I will not treat today's level as the baseline for a two-year plan. A rate that depends on one strait staying shut is not a rate you build a factory around. It is a rate you rent. The asymmetry is what makes it expensive. You earn nothing for being right that the Red Sea stays closed, because that scenario is already inside the price. You surrender the entire benefit of a reopening if you are locked long on the day it happens.

Different shippers take this differently, and the split is not the one people assume. A large BCO moving 200 FEU a month on a contract with a committed minimum is not exposed to the rate, because the rate is signed. It is exposed to the cancellation. When eighteen percent of voyages come off, the contract slot is protected last rather than first, because the contract slot is already sold and the spot slot is where this week's money is. Volume buys price. It does not buy priority unless priority was written into the document, and mostly it was not written in. The failure looks fine in January and becomes a rolled booking in the final week of the quarter, and the cost lands not in freight but in the penalty clause on the other side of the ocean.

A shipper moving 40 FEU a month has the mirror problem. They buy spot or a quarterly mini-contract, so a cancellation is survivable: rebook, pay the new number, move on. Their exposure is the delay, and the delay runs to 14 days. Fourteen days is not an inconvenience when your buyer has a delivery window. It is the difference between a full invoice and a chargeback. Small shippers also see the notice last. They hear about a blank sailing from their forwarder, typically a week after it entered the market, by which point the substitute sailings have already repriced.

Category decides more than scale does. Furniture, appliances, seasonal goods, anything with a fixed retail launch, carry the delay risk almost in full, because the cargo loses value with the calendar. A container of patio furniture arriving three weeks late arrives in the wrong season, and the freight saved is nothing against the markdown. High-value electronics carry the rate risk instead, because freight is a thin slice of landed cost while the capital in transit is thick. A ten percent move in the all-in rate on a container of laptops is a rounding error next to a week of carrying cost. Thin-margin commodity goods carry both, which is why they are first to move to a slower and cheaper routing and first to be hurt when that routing gets blanked.

Lane and port finish the split. The transpacific eastbound took 63 percent of the cancellations, so the discipline was concentrated on the US-bound headhaul while the Asia-Europe leg was handled with the long route instead. Two mechanisms, one headline. If you ship both, you are carrying two different risks under one contract template, and one of them will be wrong. The lane with the cancellations needs a service commitment. The lane with the diversion needs a reopening clause.

Timing is where procurement decisions are made or missed, and the first wave hits the booking before it hits the rate. A blank-sailing notice lands two to four weeks ahead of the affected voyage. In that window space on the remaining sailings tightens and the substitute price moves, all of it before the index prints anything. On spot, your cost changes within days of the notice while your index reading changes three weeks later. That lag is the whole game if you buy spot.

The second wave is contract repricing and it runs on a quarter. Annual contracts reset on their anniversary, quarterly mini-contracts every thirteen weeks. In a capacity-managed market the carrier does not need the index to rise before raising your renewal. It needs only to show that it can withhold capacity, and the cancellation record already shows it. Expect renewals priced against the carrier's willingness to sail rather than against the past twelve months of prints. If your renewal falls in the first quarter of next year, you are negotiating now, in the sense that the evidence being used against you is being generated now.

The third wave is structural and it takes a season to arrive. If cancellations keep running at this cadence, peak season stops being a demand event and becomes a capacity event. Peak rates used to be driven by everybody shipping at once. In a managed market they are driven by how much capacity the carriers choose to leave in for the peak. That moves the planning calendar. You stop planning around when your customer orders and start planning around when the carrier publishes. Most shippers have no habit for that, and the ones who acquire it buy space in the quiet weeks at a price the others will envy by August.

Run a small number, because percentages hide cash. Assumptions, and they are mine rather than market quotes: 40 FEU a month, all-in ocean rate of 2,400 dollars per FEU on that lane. Substitute your own figure, the arithmetic travels. Monthly ocean spend is 96,000 dollars. Now suppose the discipline holds the rate flat while the surplus would have pushed it down 15 percent. That fifteen percent is 14,400 dollars a month and 172,800 dollars a year, on forty boxes. That is the price of the floor. The importer pays it, quietly, in the shape of a rate that did not fall, and it never appears as a line item because nothing happened.

Same arithmetic at scale, same caveat on the assumption: 200 FEU a month at 2,400 dollars is 480,000 dollars a month, and fifteen points of foregone decline is 72,000 dollars a month, 864,000 a year. At that size the answer is not to argue about the rate. It is to buy the option. A second-carrier allocation with a minimum sailing frequency, or a service commitment on the primary string, costs a premium. Call it 3 to 5 percent, my assumption again, which is 14,400 to 24,000 dollars a month on that spend. Set it against 72,000 dollars a month of foregone decline plus the cost of one rolled quarter. The insurance is cheaper than the exposure. It almost always is, and it almost always gets bought after the first roll rather than before.

Demand growing 3 to 4 percent is no comfort either, and this is the part that gets read backwards. Growth of three to four percent means the market is bigger than last year, which also means your competitor's volume is up, so the surplus is not being worked off by growth. In a clearing market, three percent demand against eight percent supply still produces a fall. Growth only helps if it outruns the ships, and it is running at less than half the rate.

Put the eighteen percent on your own calendar, because that is the number you actually manage. A weekly string runs 52 sailings a year, so eighteen percent off is nine or ten voyages gone. If your cargo sat on three of them, likely enough since you book the same weekly service, you are looking at three rolls a year, and at a week each that is three weeks of schedule risk per container. Nobody puts that line in the rate comparison, and on most lanes it is bigger than the difference between the two cheapest quotes.

There is an institutional reason to expect this to persist rather than fade. The discipline is not a strategy one carrier can abandon quietly, because the cost of abandoning it is visible immediately: the first line to break ranks fills its ships and drops the rate for everyone, itself included. That is a coordination problem this industry failed at for decades and has now, for one cycle, solved. Assume it holds until the cancellation data says otherwise, and do not assume it holds forever.

Keep the forecasters' own figures on the table, because they define the stake. Going into the year, the published expectation was spot down 25 percent and contract down 10 percent across the full year. Those numbers were in somebody's procurement plan, probably yours. If contract instead comes out flat, the ten percent you budgeted as a saving has to be found somewhere else in the landed cost, and it is not there. That is the quiet half of a capacity-managed market. The savings are not taken from you in an argument. They simply fail to appear.

The counterfactual matters because it tells you when to stop waiting. Suppose you do what most shippers did last cycle: hold volume back, refuse to commit, wait for the rate to break. If the discipline holds, you spend the year buying spot at a level that never comes down, and then you pay the peak premium on top when your own demand arrives. Waiting is correct only if the cancellations stop. That is measurable, which is the good news.

The condition that flips the conclusion is specific and observable. Blank sailings are expensive to run. A cancelled voyage still has to be absorbed somewhere in the network, and the carrier eats the repositioning. If bookings fall far enough that cancellations stop being sufficient to hold the price, the discipline breaks and the market clears on supply again. Watch two series and ignore the rest: the weekly count of announced blank sailings, and whether realised cancellations match the announcements. Announcements high while realised cancellations fall means the carriers are announcing for effect and the floor is thinner than it looks. Both falling together means the market is clearing again and waiting has become the right trade.

The alternatives each come with a bill. Moving a lane to a different port buys a shorter inland leg and a different sailing string, and costs a new drayage contract plus a set of cutoffs you have not tested. Splitting volume across three carriers buys resilience and costs you the volume discount on all three. Moving to a slower service buys 10 to 15 percent on the rate and costs a week of transit that your customer notices before your finance team does. Air freight buys certainty of arrival at five to ten times the ocean rate, which is rational only for the slice of the order carrying a penalty. Carrying more inventory buys protection against the roll and ties up cash at your cost of capital. The trap is choosing one of these in the week the roll happens, which is the most expensive week in the calendar to buy anything.

Two traps belong to this market specifically. Signing a long contract at a rate that already contains the Red Sea premium is a bet that the strait stays shut for the life of the contract, and nobody is paying you to take that bet. Treating a low annual rate as a win when the contract carries no service commitment is the other. A rate eight percent under market with a thirty percent chance of a two-week roll is not eight percent cheaper. Run the chargeback through it and it is dearer than market, and you find that out in the quarter where the discovery costs the most.

None of this is a forecast. It is a description of who sets the price and what they use to set it. Carriers spent a decade being told that a commodity market cannot be managed. They have spent this year demonstrating that it can, provided you are willing to sail fewer ships and call the result a schedule. The rate is not a weather report any more. It is a quotation, and the person quoting also decides how many ships leave. Somebody ought to tell the textbooks. At this point the textbooks are running about eighteen percent behind.

↑ Back to the key points

— By Vivian Zhao