Capital sectors arrive in the order their constraints clear

Demand is rarely the binding constraint in capital infrastructure. It is usually the first condition satisfied and the least informative one. Capital appetite, tenant interest, and policy enthusiasm can all be in place for years while nothing gets built.

What decides whether a sector is buildable is narrower: whether the specific constraint that gates it has been resolved, and whether the mechanism for resolving it is repeatable. Sectors reach that state in sequence, and the sequence is legible in advance.

Sequence is set by constraint clearing, not by appetite

Every sector is gated by something. For some it is a physical input — power, water, land with the right characteristics. For others it is a procurement route, a regulatory precedent, or a standard that does not yet exist. The gate is what determines whether a well-capitalized owner with a real requirement can move from intent to construction.

The useful distinction is not how large a sector is, but what kind of gate stands in front of it. Gates fall into roughly three kinds, and they clear at very different speeds.

An engineering gate clears through work. Someone studies the network, sizes the upgrade, and prices it. The path is known even when the answer is unwelcome.

A policy gate clears through decision. Someone with authority commits to a route, a standard, or a funding mechanism. No amount of engineering advances it, and the timing is not the owner's to control.

A consent gate clears through precedent. A first project is completed, examined, and found acceptable — or it is not, and the sector waits.

The first wave is gated by engineering

Digital infrastructure and data centers, power and grid, heavy civil, semiconductor facilities, and healthcare construction share a structural feature. Their constraints are severe but they are known, and the process for resolving them exists.

Interconnection is hard, slow, and largely outside the owner's control, but it is a defined process with a queue, a study, and an answer. Long-lead electrical equipment is scarce, but it is orderable. Cleanroom and tool-install requirements are demanding, but they are specified. Healthcare carries dense regulatory review, but the review has a route and precedent behind it.

None of that makes these projects easy. It makes them tractable. An owner can determine what has to happen, who has to do it, and roughly in what order — which is the condition under which capital will actually commit.

This is also why the failures in these sectors are execution failures rather than viability failures. The projects are buildable. Whether a particular one gets built well is a separate question, and it turns on definition, governance, and procurement discipline rather than on whether the sector was ready.

The second wave is gated by decision

Defense-related facilities, decarbonization and adaptive reuse, bio-manufacturing, and modular and industrialized housing are not waiting on engineering answers. In most cases the technical questions are answerable with existing knowledge.

They are waiting on commitments that only institutions can make: a procurement route that rewards the approach, a standard that lets a design be replicated across jurisdictions, a funding mechanism that survives more than one budget cycle, an approval pathway that treats a repeatable product as a product rather than as a series of unrelated one-offs.

A sector waiting on a decision cannot be advanced by diligence. Diligence tells you what the decision is worth; it does not tell you when it will be made.

For an owner, this changes what preparation means. In an engineering-gated sector, preparation is study work. In a decision-gated sector, preparation is optionality — holding a position that can be converted quickly if the gate opens, without carrying the cost of a program that assumes it already has.

The characteristic error here is building a schedule around an expected policy outcome. The schedule is then hostage to something no one on the project team can influence, and the float that should have been reserved for it gets consumed by activities that were never the problem.

The third wave is gated by consent

Water infrastructure and desalination, nuclear and small modular reactors, and climate adaptation works are gated by something slower than either engineering or procurement. They require public acceptance of a specific consequence: a rate increase, a brine discharge, a reactor sited within view, a defense that protects one area and exposes another.

That acceptance is not won by argument. It is won by precedent — a completed project that behaved as described. Until a credible first case exists in a comparable jurisdiction, each project carries the full weight of proving the category, and the cost of that proof lands on whoever moves first.

These sectors also carry first-of-a-kind risk in its purest form. Cost bases are thin, supply chains are shallow, and the regulatory review is genuinely novel rather than merely demanding. Contingency built on comparable projects is contingency built on nothing.

What this changes for an owner

The framing matters more than the taxonomy. Any sector can be placed in this structure, and the phases are descriptive rather than fixed — a policy gate that clears moves a sector forward, and a failed first project can move one back.

The questions worth asking about a sector are the same in every case. What specifically has to be true before a project here can be built. Who resolves it. Has anyone resolved it before, in a comparable jurisdiction, with a documented outcome. What does the resolution cost, and who bears it.

Where those questions have documented answers, the sector is buildable and the risk is execution risk, which an owner can govern.

Where they do not, the project is not a project yet. It is a position in a sector that has not opened, and it should be carried, funded, and reported as one.