THS/01 · Infrastructure Capital Thesis
Power is the new capital constraint.
The binding constraint on AI, industrial reshoring, and electrification is no longer financing — it is deliverable interconnected power on a 12–36 month horizon.
Context
For two decades, capital was the scarce input in infrastructure. Today, the entire AI buildout, the reshoring of US manufacturing, and the electrification of transport and heat are colliding with a transmission system that was planned against a flat-load assumption. PJM's 2025/26 BRA print, MISO's LRTP, ERCOT's large flexible load framework, and SPP's queue dynamics all point to the same conclusion.
Positioning
- Underwrite power deliverability before underwriting capital structure.
- Treat queue position, ATC headroom, and substation capacity as primary risk variables — not site characteristics.
- Build optionality on behind-the-meter generation, co-location, and demand-flex structures.
Implications
IMP/01
Asset-level returns increasingly diverge based on grid position rather than sponsor execution.
IMP/02
Land without power optionality is a depreciating asset on an extended timeline.
IMP/03
Hyperscale, industrial, and energy-storage developers are now competing for the same scarce interconnection inventory.
Watch
- › FERC posture on behind-the-meter co-location.
- › PJM capacity auction trajectory through 2027/28.
- › MISO LRTP Tranche 2 approvals and EPC capacity to deliver them.
Pressure-test this thesis against your portfolio or pipeline.