Twelve candidate regional unions, scored on filed customs data by the same solver the rest of this platform uses. Three of them already exist and are included as controls: a method that cannot recognise USMCA and the EU core is not measuring union-ness. The other nine are proposals — some geographic, some built from factor complementarity, some organised along a single supply chain.
Can the group meet each member's critical demand from inside? Scored as the worst member's interior coverage, never the average — a bloc that serves four members well and a fifth not at all is not a union, and an average hides exactly the member who would leave.
Does every member gain, and are the gains distributed? A bloc whose surplus accrues to one member is a hub with satellites. It can be perfectly efficient and still fail as a union, because the satellites have no reason to stay.
Each member is removed in turn and the network re-solved. The one whose departure costs most is what the arrangement actually rests on. Weaker members are entitled to know that number before signing anything.
Solving twelve networks…
USMCA is a trade instrument. It sets rules of origin, dispute panels and tariff schedules, and says nothing about the three things that decide whether neighbours actually prosper together: whether they build each other's capability, whether they defend each other, and whether the largest of them agrees to bound its own power. A trade agreement between parties of grossly unequal size is a contract the larger party can rewrite at will — not an accusation, a structural fact — and the remedy is to write the bound into the instrument.
Each clause below is borrowed from an instrument that already uses it, carries the threshold that instrument actually sets, and is tested against filed data. A pledge that cannot be measured cannot be breached, and a pledge that cannot be breached is decoration.
Testing clauses…