Lease runs to year twelve, the power arrangement resets in year seven
Looking at an LP slice in a 9 MW single-tenant facility in a secondary market. The lease is twelve years with two five-year renewals at market. The power side is a seven-year block arrangement with the utility, plus a second feed that has a study letter and no energization date. So the rent in years eight through twelve depends on re-contracting electricity at a price nobody in this deal will put in writing.
The sponsor's answer is that power is a pass-through, the tenant carries the cost, and a price move in year eight lands on the tenant's P&L rather than mine. Add that the grid is backordered everywhere and the tenant has no cheaper place to move nine megawatts to, so renewal is close to automatic and the term mismatch is theoretical.
The other reading is that a pass-through only protects you while the tenant can absorb it. If delivered power gets expensive and the interconnection caps them below the density they want by year ten, they leave at expiry and I own a shell in a market whose queue may have moved on. In that world the twelve-year lease was really a seven-year lease with an option.
I can build a model that supports either. What I want to know is which term the people here actually anchor the hold period to, because that choice sets the exit assumption and everything downstream of it.
Which term do you anchor a passive data center hold to?
21 votes