A Framework for Locating Scarcity Value in the Network

The previous modules built the components of an interconnect thesis: the three-leg capital stack, the three prices of the wire, the subsea and fiber capital flows, the shift to hyperscaler ownership, and the fabric-and-optics supply chain. This lesson assembles them into a single framework for locating where network constraints convert into durable scarcity value, because not every constraint does. Some bottlenecks are temporary and compete away; others are structural and pay rents for decades. Distinguishing them is the whole game.

Start with a definition of durable scarcity value. A constraint produces durable value when it is both hard to duplicate and strategically necessary. Hard to duplicate means that a competitor cannot simply build around it, whether because of geography, permitting, capital intensity, or technical difficulty. Strategically necessary means that the workloads generating the most valuable demand cannot function without it. A constraint that is hard to duplicate but not necessary is a curiosity; a constraint that is necessary but easy to duplicate invites competition that erodes the rent. Only the intersection, hard and necessary, sustains returns. Every candidate interconnect asset should be tested against both conditions.

Apply the test across the three scales. At the subsea scale, raw capacity fails the durability test, because on well-served corridors capacity is abundant and duplicable. What passes is route diversity through geographic chokepoints: an independent path that avoids a congested or vulnerable corridor is hard to duplicate, because the geography that makes it independent is fixed, and it is necessary for any workload that cannot tolerate a cut. This is why, as argued earlier, the market's systematic underpricing of diversity is a recurring opportunity. Diversity is exactly the attribute that satisfies both conditions and that the market discounts because its value only appears on the day of a failure.

At the metro scale, the framework points to the neutral meeting points where many networks interconnect and to the low-latency routes between compute campuses. A neutral interconnection point is hard to duplicate because its value grows with the number of networks already present, a position that a new entrant cannot replicate quickly, and it is necessary because traffic between different networks must physically meet somewhere. Low-latency metro routes are hard to duplicate because the physics of distance is fixed and the right-of-way to build a shorter path may not exist, and they are necessary for latency-sensitive inference and interactive workloads. Both satisfy the test, and both are places where independent capital can earn scarcity rents even in a market increasingly dominated by vertically integrated giants.

At the intra-cluster scale, the framework points to the leading edge of the optics supply chain. Leading-edge transceivers, high-capacity switch silicon, and the packaging frontier are hard to duplicate because the technical difficulty concentrates capability among a few manufacturers, and they are necessary because the superlinear scaling of the fabric means the largest clusters cannot be built without them. The scarcity here is not permanent for any single component, because each generation eventually matures and its rents compress, but the leading edge itself is a durable position: whoever can consistently execute at the frontier captures the scarcity that perpetually re-forms there as cluster sizes grow.

Notice the common thread. In every case, the durable value sits not in the abundant, easily built capacity but in the specific attribute that is fixed by geography, position, or technical difficulty and that a critical workload cannot do without. This is the discipline the framework enforces: resist the temptation to invest in the leg that is growing fastest in headline terms, and instead locate the specific chokepoint within it that satisfies both conditions. The fastest-growing leg often has the most competition and the thinnest rents; the durable returns hide in the attribute that the growth depends on but cannot easily reproduce.

The framework also disciplines the question of who captures the value. Vertically integrated hyperscalers capture the value of the trunk routes and private fabrics they build for themselves, internalizing those rents rather than paying them to others. Independent capital captures the value of the neutral meeting points, the diverse secondary routes, and the leading-edge components that even the giants must buy or depend upon. The analyst's task is to map each chokepoint to its likely captor and to price accordingly, distinguishing the rents that will be internalized from those that remain available to the merchant market. Carry this two-part test forward, hard to duplicate and strategically necessary, and apply it to every interconnect asset you evaluate. It is the compact expression of the entire course, and the next lesson turns it into a working playbook of signals, divergences, and positioning.