Supply Chain Action Points
Read this first — the conclusion, and the moves to make:
- By 31 July, measure the actual round-voyage day count on each string from proforma schedules rather than from the market average; if it is above twelve days, re-run the annual freight budget at the top of the 25-40% band.
- Do not budget at 25%: budget at 35% and negotiate a 40% ceiling with a written re-opener, because the gap between the ends of the band is 15 points, about 12% of the low end.
- Cap contract tenor at one selling cycle and no more than four quarters; add a written re-opener triggered by restoration of the Cape routing on your specific string.
- Add 13 to 19 FEU-equivalents of in-transit inventory per 40 FEU a month to the working capital plan, and recompute it at your own cargo value per FEU rather than at 10,000 dollars.
- By 15 August, ask each carrier for the vessel count per string before and after the diversion; if the count rose by less than 15%, the capacity figure being quoted to you is not the one on your service.
With Houthi attacks persisting, most major carriers keep Cape of Good Hope routing as default, adding 10-14 days per round voyage and removing an estimated 1.3-1.8 million TEU of effective capacity. Analysts expect diversions to continue through at least 2027, keeping Asia-Europe rates 25-40% above pre-crisis levels.
The Analysis
One point three to one point eight million TEU of effective capacity. Ten to fourteen days per round voyage. Twenty-five to forty percent above pre-crisis on Asia-Europe. Three numbers, all of them ranges, and one of them is not independent of the other two.
Most major carriers are still treating the Cape of Good Hope as the default. The attacks have not stopped, and the analysts now put the diversion at least through 2027. So the trade is being planned for another eighteen months of a longer voyage, on a rate band fifteen points wide.
Before that band goes into a tender, we should take it apart. Ranges that wide are usually two estimates wearing one coat, and the coat is where the money hides.
Ten to fourteen days on a round voyage is a cycle-time increase, and cycle time converts into vessels. Take a weekly Asia-Europe string and assume a seventy-day round voyage before the diversion, which is my assumption and not a number in the report. At seventy days, a weekly service needs ten vessels. At eighty-two days it needs twelve. Two extra hulls for the same weekly slot is a 20% increase in the vessel requirement for that string. Do that arithmetic across the strings on the trade and you arrive at the 1.3 to 1.8 million TEU. The capacity figure is not a second shock. It is the day count, counted a different way.
Which means the two numbers have to agree, and the width of the capacity range is really a question about inputs. Twelve days with 1.3 million TEU removed implies a different vessel count and average size than fourteen days with 1.8 million. That 500,000 TEU spread is not uncertainty about the attacks. It is uncertainty about which ships, how many, and how full. When a broker hands you one end of the range as fact, ask which of the three inputs moved. Usually nobody knows, and the number is a midpoint in a costume.
The 2027 end date deserves the same treatment. A diversion lasts as long as the routing does, and the routing lasts as long as the capacity can be kept on the water. Hulls are not printed on demand; the orderbook delivers on a schedule set years ago. A forecast that diversions persist through 2027 is therefore, in arithmetic terms, a statement that the tonnage needed to restore the old routing will not be available before then. That is a different claim from a forecast about the security situation in the Bab el-Mandeb, and it is the more robust of the two. Security can improve in a week. A shipyard slot cannot.
Now the 25 to 40. Put the pre-crisis rate at 100 so we are not inventing an absolute. Twenty-five percent is 125. Forty percent is 140. The gap between the ends is 15, which is 12% of the low end and 10.7% of the high end. On a 40-FEU monthly programme, budgeting at the low end and paying at the high end is a 15% miss on the annual freight line, not on a single booking. A procurement team that puts 25% in the budget and signs a contract with a 40% ceiling has hedged nothing. It has written a fifteen-point spread into its own P&L and called it a forecast.
Who absorbs what. An importer with a fixed retail price and a sixty-day shipping cycle eats the full spread, because there is no point in the calendar where the price can move. An importer who re-prices quarterly passes roughly half of it and eats the other half in volume. An exporter selling FOB hands the freight to the buyer and loses only on competitiveness, which is a slower and quieter kind of loss. A high-value, low-freight-share shipper barely notices the rate and gets hit by the days instead, through inventory in transit. A low-value bulky shipper gets both, and usually has the least room to renegotiate either.
The days are the part that never appears on a freight invoice. Ten to fourteen extra days of transit on a 40-FEU monthly programme is roughly 13 to 19 FEU-equivalents of cargo in the water at any moment that would not have been there before, on my assumption of a thirty-day month. At 10,000 dollars of cargo value per FEU, that is 130,000 to 190,000 dollars of additional working capital, and it scales linearly, so put your own unit value in. Nobody budgets this line because nobody invoices it. It shows up as a higher borrowing base and a wider forecast error, months after the decision.
When it lands. The transit days hit the first shipment that sails after the diversion is confirmed on your string, and you can date that from the proforma rather than from a news story. The rate hits at the next booking. The contract hits at renewal, and renewal is where a band becomes a number that gets defended for twelve months. If your contract renews in Q4, the number that matters is not today's spot print. It is the average of the last eight weeks plus whatever the carrier believes 2027 looks like.
The coverage treated the capacity figure, the day count and the rate band as three separate findings. They are one finding with three faces. The capacity removed is the day count divided into vessel turns. The rate band is that capacity number marked up by whatever the market will bear. The 2027 date is the delivery schedule of the ships that would undo it. Once you see it that way, the actionable number is none of the three. It is the day count on your own string, which you can measure, and the vessel count on your own service, which your carrier will tell you if you ask about the proforma rather than about the rate.
The counterfactual is not a small one. A durable ceasefire, an insurance market that reopens, and the Cape becomes optional again inside a quarter. Capacity comes back faster than demand, spot falls straight through the 2027 forecast, and anyone who signed twelve months at 40% above pre-crisis is underwater for the remaining term with no clause to reopen. The asymmetry is not symmetric. Waiting costs you a rate cut you might have captured; locking costs you a year you cannot exit. That argues about tenor, not about direction, and the two get confused constantly in procurement meetings.
The trap is treating a range as a point estimate and picking the friendly end. Twenty-five percent is not the forecast. It is the bottom of the forecast, and it is the number that will be used to justify a budget that the 40% will then break. The second trap is counting the days as a service problem and the rate as a cost problem, when they are one problem on two clocks. Lock the rate for the length of your selling cycle, not the length of the disruption. If your selling cycle is a quarter, a quarter is the right tenor, and 2027 is somebody else's spreadsheet. We will know which forecast was right from the shipyard delivery schedule, not from the news.
↑ Back to the key points