Supply Chain Action Points
Read this first — the conclusion, and the moves to make:
- By 17 July, place the June and July invoices for the same lane and same box side by side, count the surcharge lines and compute surcharges as a share of the total; carry that percentage in the quarterly cost sheet and refresh it monthly.
- Give every quote and contract a lane and an effective date. For cargo that is not transpacific eastbound, reject a blanket increase justified by the composite and ask in writing for the lane-specific basis.
- Put a cap on surcharges in the contract: combined surcharges may not exceed a stated percentage of base freight, set from your own last-quarter actual, with a fixed review window for anything above it.
- By 14 August, stress-test open spot volume at two consecutive weeks of plus five percent; if the monthly increase passes ten percent of that month's freight spend, lock a portion of the space before negotiating further.
- Each Monday read two numbers only: your lane's week-on-week move and blank sailings as a share of scheduled departures. While blanking sits near three percent, plan on supply being spent and do not schedule around a correction.
Shippabo's July newsletter notes Drewry's World Container Index climbed as much as 23% in a single week before settling into a steadier ~5% weekly gain by month-end, a 22-month high, with blank sailings falling to roughly 3% of scheduled departures. Shanghai–Los Angeles rates rose 12% and Shanghai–New York 6% week-on-week per Drewry.
The Analysis
Drewry's World Container Index is at a 22-month high. The number everybody is quoting is 23 percent, and it deserves a second look, because 23 percent is not a week-on-week figure for late June. It is the early-June spike, before the index settled into roughly five percent a week by month-end. The lane numbers attached to the same story, Shanghai to Los Angeles plus 12 percent and Shanghai to New York plus 6 percent, are late-June week-on-week moves.
Three numbers, three weeks, one headline. None of them is wrong. All of them are read as if they describe the same market.
The order they come in is the actual information: 23, then 5, then 12 and 6. That is a decelerating curve sitting on a 22-month high, and a decelerating curve at a high is what a top looks like, not a middle. So the question is not how fast rates are rising. It is what is left in the tank, and who is holding it.
The index is a weighted basket of lane assessments, averaged and published as one number. That construction is useful for a headline and close to useless for a budget, because your cost is one item in the basket and the basket is not your invoice. When the composite prints 23 and your lane prints 12, nothing is inconsistent. It means the other lanes did the work and you did not ship on them. Read your lane, read the assessment date on your lane, then decide.
Put five percent into a calculator and it stops sounding modest. Five percent a week for a quarter is 1.05 to the thirteenth, roughly an 89 percent rise. Run it for a year and the index multiplies by about twelve and a half. Nothing in container shipping compounds at five percent a week for twelve months, which tells you the print is a short burst by construction and not a trend you can push into a budget. Anyone modelling next quarter at five a week is not modelling. They are drawing.
The most important line in this release is the least quoted one. Blank sailings are down to roughly three percent of scheduled departures, among the lowest in over a year. Read that as an inventory statement: the capacity has already been deployed. Blanking is the supply-side instrument, and it has been used up. With the lever sitting at three percent, the only instruments left are price and the surcharge schedule. You are not watching a market that might tighten. You are watching one that has finished tightening.
Contract against spot is the whole distribution of pain here. The release says contract rates are more insulated and renewals are trending firmer. Insulation is another word for a widening spread, and a spread is a transfer: whoever sits on the contract is subsidised by whoever books spot, until renewal, at which point the subsidy flips direction. A BCO holding a first-quarter contract that runs to December is, at this moment, the cheapest container in the market and does not know it. An SME booking twenty boxes a month at the assessed rate is funding that difference. Both will sign in the same room in the autumn, and one of them will be surprised.
Direction splits cleanly as well. The tightness described is transpacific eastbound, with capacity concentrated there rather than on Asia-Europe or the Atlantic. If you ship Asia-Europe, the correct response to a general increase this month is a polite request for the lane-specific basis, because the constraint behind this index print is not on your water. Accepting a blanket rise because the composite moved is paying for somebody else's lane.
Two clocks, and budgets usually carry one. A July GRI has an effective date, applies to bills of lading from that date, lands on the invoice in the following billing cycle, and reaches landed cost in the quarter after. Call it two to six weeks from filing to money, on ordinary monthly terms. The booking decision that created the exposure was taken weeks before the filing, on the lead times this market now demands. By the time the index reaches your inbox the cargo is on the water and the rate is already fixed. The index is a record, not a warning.
Here is the calculation I would actually run, and it needs no rate from me. Take one July invoice and one June invoice for the same lane and the same box. Draw a line where base freight ends and surcharges begin. There will be a GRI or a peak-season line, a bunker line, and on the East Coast and Gulf strings additional weight-related charges, because carriers are now adding heavy-load surcharges and weight restrictions on those services. Count the surcharge lines, then divide surcharge total by invoice total.
That percentage is the number to manage this quarter. If it is climbing while the base rate stands still, your contract is protecting a shrinking share of what you actually pay. A document that fixes the base and leaves the surcharge schedule open has fixed the smaller half of the invoice, and a renewal conversation about firmer rates is really a conversation about the half that was never covered.
Assume a monthly transpacific spend of one million dollars for illustration. That is an example figure, not a market rate. If surcharges are 18 percent of it, 180,000 dollars a month sits outside your rate clause and outside most of your dispute mechanism. That is the exposure, and no index prints it. My numbers, plainly labelled, because the only honest way to do this arithmetic is to let you substitute yours.
The trap at a 22-month high is assuming the top is a wall. It is not a wall. It is a slope in both directions. The opposite trap is waiting for the correction: with blanking at three percent and every sailing full, waiting is a position, and it carries a cost that appears on no index. The third trap sits in the contract itself, where a firm renewal signed at the top of a move and written without a collar turns a quarterly problem into an annual one.
The index will tell you where the market went. It will not tell you that you were paying for a lane you do not ship on, that three percent blanking means the supply lever is spent, or that the number your contract protects is a declining share of your invoice. Those three are worth more than the headline. Lock when the second consecutive week prints a smaller gain than the week before it. That is the first week the market has agreed with itself, and it is the cheapest week to buy a ceiling.
↑ Back to the key points