Chokepoints

Say why a six-day closure produced months of congestion, using the Suez and Bab el-Mandeb numbers.

The Ever Given was stuck in the Suez Canal for six days in March 2021. By the time it was refloated, roughly 370 ships were waiting to get through. The queue took about a month to clear and the port congestion downstream took considerably longer. Six days of blockage, months of consequences, and the ratio between those two numbers is not in the blockage.

A canal is a queue

Ships arrive at some rate. The canal moves them at some rate. The difference between those two is the only slack in the system, and it is a much smaller number than either of them.

Close it and the backlog builds at the full arrival rate. Reopen it and the backlog drains at the spare capacity. That gives two lines of arithmetic that do all the work:

backlog = arrivals × days closed
days to clear = backlog ÷ (capacity − arrivals)

Dividing by a small number gives a large one, and spare capacity is the first thing an efficient system gets rid of.

Ships waiting at the worst

300

Days to clear once it reopens

Disruption per day of closure

5.0x

Ship-days lost waiting

5,400

6 days shut, 30 days to clear. The backlog is 50 x 6 ships and it drains at 10 a day, so every day of closure buys 5.0 days of disruption. Nothing about that ratio is in the closure. It is all in the spare capacity.

Days the strait is shut6 daysShips the strait can move a day60 against 50 arriving

Move the closure slider and the damage moves with it, in proportion, as everybody expects. Then leave the closure alone and move the capacity slider instead. The recovery doubles while the incident stays exactly the same length.

Transit and queue figures for March 2021 are in the Suez Canal Authority's own statements and in Lloyd's List reporting from the period, both public. The arithmetic above is the model on this page, which is unit tested, including a check that the simulation and the closed-form formula agree.

ConventionQueueing theory calls the arrival rate divided by the service rate the utilization, and the result that matters is that waiting time rises toward infinity as utilization approaches one. It is not a gentle curve. Going from 80% to 90% utilized roughly doubles the wait, and 90% to 95% doubles it again. Every network anybody depends on lives somewhere on that curve, and where it lives is a choice somebody made about cost.

Why states care about specific pieces of water

A chokepoint is a place where a large share of something has to pass through a small space. Suez and Panama for containers, Hormuz for oil out of the Gulf, Bab el-Mandeb for the Red Sea route, Malacca for almost everything moving between the Indian Ocean and East Asia.

What makes them matter strategically is the absence of a cheap detour. Suez has one: sail around the Cape of Good Hope and add about ten days on the Asia to Northern Europe run. That detour caps what Egypt can charge, and it is why the 2024 Red Sea diversions were expensive rather than catastrophic. Hormuz has no detour at all for most Gulf oil, which is why it appears in every serious conversation about the region.

The same logic runs on land and in factories. A small number of plants make the machines that make advanced chips. A handful of refineries process most of certain rare earths. Those are chokepoints with the same arithmetic and no ships in them.

What breaks

Reading the tonnage instead of the substitutability. The share of world trade passing through a strait tells you how much moves, not how much would be lost. What matters is what happens if it stops: whether there is another route, how much longer it is, and how much spare capacity exists on it. A strait with a detour is an expense. A strait without one is a weapon.

Treating the detour as free once it exists. Going the long way spends the same thing the queue was spending, which is ship-days. Ten extra days on a voyage removes roughly a fifth of the effective global fleet, and rates rise on routes nowhere near the disruption. The alternative route caps the damage; it does not remove it.

Assuming leverage survives being used. Closing a strait is the most effective way to make everybody who depends on it stop depending on it. Pipelines get built, alternative ports get funded, inventories get rebuilt, and navies get budgets. The threat is worth more than the act, which is why chokepoints are mostly used as threats.

What to do with this

When something breaks in a supply chain, ask two questions rather than one. How long was it broken, and how much headroom does the system have. The first is in every headline and the second is what determines the answer.

And apply it to things that are not shipping, because the arithmetic does not care. A hospital running at 95% of beds, a team with no slack in its schedule, a payment network with one clearing house. Each of those has an arrival rate, a service rate, and a recovery time that depends on the gap. Efficiency spends that gap, and the bill arrives on the day something goes wrong.

Test yourself

01A port runs at 98% of capacity and is praised for its efficiency. What have they bought and what have they sold?

They have bought lower costs per container and sold their ability to recover. At 98% utilization the spare capacity is 2%, so a backlog drains fifty times slower than it built. A one-week disruption becomes a year of congestion.

This is the general shape of it. Utilization and resilience are the same dial with two labels, and every efficiency drive in a network is a decision about how long the next incident will last. Nobody makes that decision explicitly, which is why it keeps being made badly.

02Is a chokepoint leverage for whoever controls it, or a liability?

Both, and which one dominates depends on whether a detour exists. Egypt earns billions a year in Suez tolls and cannot charge much more than the cost of going round the Cape, because that is the alternative and everybody knows the number. The detour puts a ceiling on the rent.

Where no detour exists the leverage is real and so is the target painted on it. A state that can close a strait invites everybody who depends on that strait to build a navy, fund its rivals, or spend twenty years engineering the dependency away. Leverage you use once is leverage you had.

03Name the second-order effect of shipping rerouting around the Cape of Good Hope for a year.

The world's shipping capacity effectively shrinks, because the same fleet spends longer on each voyage. Ten extra days on an Asia to Europe run is roughly 20% more time per trip, which removes about a fifth of the effective fleet without a single ship being lost. Freight rates rise everywhere, including on routes that never touched the Red Sea, which is how a regional problem becomes a global price.

A tutor that knows this lesson. It asks before it explains, and it will not tell you what to do with your own money.

Educational material, not investment or policy advice. Figures are cited where they come from a filing or a statistical series, and labelled as illustrative where they do not.