Chokepoints and the cost of cutting off access

In 2022, Ukraine asked SpaceX CEO Elon Musk to activate Starlink coverage around Sevastopol to support an attack on Russia’s Black Sea Fleet. He refused, saying SpaceX would become “explicitly complicit in a major act of war”; he later also cited US sanctions covering Crimea. At the moment that mattered, one private actor had the final say. It exposed a question hidden inside every chokepoint. We usually ask where dependence is concentrated: a strait, mineral, technology, network. But three simpler questions matter. How much depends on it? What does it take to use it? Who decides? The threshold of a chokepoint lies in the second question: what someone must risk, in blood, money, law or reputation, before leverage becomes action. China has lived with the first question for more than two decades. In 2003, then president Hu Jintao warned that “certain major powers” could control the Strait of Malacca, exposing China’s dependence on imported energy. Beijing gave the anxiety a name: the “Malacca dilemma”. Twenty years on, that fear has bought a navy. Malacca did not by itself drive Beijing’s naval build-up; Taiwan, territorial disputes and great-power competition all mattered. But vulnerability at sea helped define the problem. If you cannot control the passage, build enough capability to make anyone contemplating its closure think harder about the consequences. The Strait of Hormuz showed why geography carries such a high threshold. Iran blocked the strait after US and Israeli attacks, disrupting a waterway that normally carries about a fifth of global oil and gas shipments. Months later, traffic remains heavily constrained as Iran and Oman negotiate new shipping lanes and Tehran attaches wider political conditions to reopening. The consequences do not stop with the country being targeted. Closing a major strait hits importers, exporters, insurers, allies and neutral economies at once. It can send energy prices higher thousands of kilometres from the conflict and draw other powers into protecting the route. Geography creates enormous leverage precisely because using it can impose enormous costs on everyone, including the country pulling the trigger. The second question on chokepoints becomes sharper with technology: what does it take to use the leverage? A government can tighten an export licence. But the country or company on the receiving end can look for another supplier, build capability at home or redesign around the missing input. In 2019, Japan tightened export controls on fluorinated polyamides, photoresists and hydrogen fluoride bound for South Korea amid an escalating dispute over wartime forced-labour rulings. Korean companies responded by diversifying suppliers and localising production. Using a technological chokepoint tests whether the dependence is truly irreplaceable. If substitutes can be found, exercising the leverage gives the buyer its strongest reason to find them. If they cannot, the chokepoint becomes more valuable. Japan has since found itself on the other side. China imposed controls on several rare earths last year; by mid-2026, there had been no shipments to Japan of terbium or dysprosium oxide for months. Japanese companies and policymakers have responded by seeking alternative supply and reducing exposure to China. The larger lesson is simple: what the supplier weaponises, the buyer starts paying to replace. Using a chokepoint can strengthen leverage today while beginning to weaken it tomorrow. The third question may be the least visible: who decides? Institutions appear more predictable because decisions normally pass through laws, agencies and procedures. But formal process does not tell you how many people have effective discretion over the outcome. In 2018, the US Commerce Department imposed a denial order that cut ZTE Corp off from American suppliers and pushed the Chinese telecoms company towards collapse. Then US President Donald Trump instructed the department to help ZTE “get back into business”. A negotiated settlement worth up to US$1.4 billion eventually followed. Which brings us back to Starlink. Musk’s refusal shows that when dependence is high and effective discretion is concentrated, the outcome can hinge unusually heavily on one person’s judgment. The most dangerous chokepoint may not be the narrowest strait or rarest technology. It may be the one where the final decision rests with the fewest people. The next generation of potential chokepoints is being assembled in low Earth orbit. Starlink operates around 10,000 satellites. China is building Guowang, a state-backed constellation planned at roughly 13,000 satellites, while Amazon’s Leo has deployed nearly 400 and expects to begin service this year. Scale alone does not make a network a chokepoint. It becomes one when governments, businesses or military operations organise themselves around it, alternatives thin out and leaving becomes difficult. By then, the important question is no longer who built the network. It is who can deny access to it. This is where conventional resilience audits stop too early. Boards and governments are increasingly good at identifying critical suppliers, shipping routes, minerals and technologies. But identification is not enough. They need to map control as well as exposure: every chokepoint has a threshold, and someone ultimately decides whether to cross it. How much depends on it? What stops if access disappears, and how quickly can it be replaced? What does it take to use the leverage? A licence, commercial instruction, cabinet decision, military action? How does the time needed to deny access compare with the time needed to replace it? Who decides? Who can say yes, who can say no, and who can overrule them? The last question rarely appears on a supply-chain map. That list of names is the finding.

Chokepoints and the cost of cutting off access

.css-1c6uqr6{color:inherit;font-weight:inherit;font-size:inherit;font-family:inherit;line-height:inherit;overflow-wrap:break-word;}The most dangerous chokepoint may not be the narrowest strait or rarest tech but where the final decision rests with the fewest people

The most dangerous chokepoint may not be the narrowest strait or rarest tech but where the final decision rests with the fewest people