Grid Access Is the Moat

Last edition, we asked: what happens when the binding constraint shifts from firming at the generation layer to deliverability at the transmission layer? Deliverability is not one problem. It is three layers that must simultaneously align — and the layer most invisible to capital allocators is the one that fails most quietly and most expensively.

In 2024, the IEA tracked 1,650 GW of solar and wind projects in advanced development globally that were waiting for grid connection. That figure is not a planning horizon estimate. It is capacity in advanced development that cannot yet dispatch at scale — because the infrastructure between the asset and the load is the binding constraint. But "infrastructure" is doing too much work. The constraint is not only wires. It is wires, contracts and rules — and they fail independently.

Generation capacity is being built faster than the transmission system can absorb it. This is not a technology problem. It is a sequencing failure — across three distinct layers.

Most infrastructure investment addresses Layer 1. Most project finance addresses Layer 2. Layer 3 is where the gap lives.

1. 1
   Curtailment is non-linear on IRR.

   For projects with market-price exposure, curtailment during peak solar hours creates a non-linear revenue drag — lost generation volume at peak disproportionately reduces revenue because peak hours carry the highest spot prices.

   Curtailment assumptions must be stress-tested at state or sub-state level, not national averages. India's national figure of 0.12% masks Rajasthan's 51.5% peak-hour reality.
2. 2
   Transmission delay risk is now a first-order bankability risk.

   Think about what an 18-to-24 month construction-to-revenue gap actually means in a project finance stack: extended interest-only periods, reserve accounts eaten before the asset earns a dollar, or sponsor equity bridges that were never in the original model.

   Lenders who don't stress-test this explicitly may be mispricing risk. Most aren't stress-testing it explicitly.
3. 3
   Co-located structures change the debt structure.

   A co-located generation + BESS + grid access project is a single infrastructure asset with multiple revenue streams: capacity payments, energy arbitrage, grid services, non-curtailable delivery value.

   The appropriate structure is a unitised project with blended debt tenor — longer on generation and storage, shorter on grid access if it carries regulatory risk.
4. 4
   The aggregator lane requires capital across all three layers.

   Generation alone is commoditised. Firming alone does not solve the delivery problem. Grid access alone is a toll road without traffic.

   The capital structure of a genuine aggregator looks less like a renewable energy fund and more like a vertically integrated utility with a project finance wrapper.
5. 5
   Vietnam's DPPA mechanism opens a new revenue architecture, with caveats.

   Decree 57 (March 2025) allows direct sales to large industrial consumers outside EVN offtake. Projects with BESS or flexible dispatch may receive priority grid access.

   Curtailment risk remains under Virtual DPPA structures and compensation terms are not yet clear. Structural improvement — not yet a physical grid fix.
6. 6
   RE support mechanisms in SEA carry policy interruption risk.

   Philippines ERC temporarily suspended GEA-AII collection (₱0.0371/kWh) for May–June 2026 citing consumer affordability pressure. Fund balance ~₱466.49M as of 5 May was sufficient to cover RE facility payments.

   Government-administered RE levies are politically exposed to consumer price pressure — a revenue risk standard bankability models rarely stress-test explicitly.
7. 7
   Layer 3 risk is now a front-of-model bankability condition.

   I've heard this framed as a documentation issue — just add a curtailment clause and move on. It isn't. In practice, lenders in SEA increasingly look for either a buyer who contractually pays through curtailment events, or an explicit curtailment compensation mechanism from day one — not as a fallback, but as a pre-condition for debt sizing.

   Mid-market APAC data centre operators competing against hyperscalers for grid access face the same structural dynamic as Lake Tahoe's customers: scale confers governance-access advantage, not just procurement advantage.

   The entity that navigates Layer 3 on behalf of mid-market operators — securing dispatch priority, curtailment compensation and direct contracting rights — is not just an energy intermediary. It is a governance intermediary. That is the aggregator lane, properly defined.

The aggregator thesis running through this newsletter since Edition 2 has arrived at its structural answer. It is not a PPA aggregation play. It is a grid access architecture play — built on three stacked layers, each requiring separate capital allocation, but collectively creating something that individual project finance cannot replicate: a non-curtailable, firmable, clean energy delivery path to load.

Vietnam's DPPA reform is the most advanced policy signal. Australia's REZ framework is the most structured procurement architecture. India's scale creates the largest addressable market. Malaysia is the cross-border logistics layer. No single market has all three conditions simultaneously — which is why the aggregator entity, if it emerges, will likely be regional rather than country-specific.

That is where multilateral development finance institutions, green bonds and blended finance structures come in. And that is where Edition 8 goes.