If you’re underwriting a new gas project the old way, you’re probably underwriting fiction. The constraint is not the turbine, not EPC, not even fuel. The constraint is getting on the grid on time, at a known cost, with known deliverability.
The interconnection backlog is the market signal
Interconnection queues are not just “clogged.” They are telling you the real scarcity in U.S. power: deliverable interconnection capacity.
A few numbers investors should sit with:
- As of end of 2024, ~2,300 GW of generation and storage capacity was actively seeking interconnection in the U.S.
- That’s spread across ~10,300 active projects, including ~1,400 GW of generation and ~890 GW of storage.
- Only 13% of capacity that submitted interconnection requests from 2000-2019 reached commercial operations by the end of 2024 (most withdrew).
- Median time from interconnection request to COD has stretched to over 4 years for projects built in 2018-2024 (for regions with available data).
That is why “we’ll file and see what upgrades look like” is not a plan. It’s a lottery ticket with a large carrying cost.
And here’s the part that matters for gas investors: active natural gas capacity in queues rose to 136 GW in 2024 (+72% year over year) even while solar/storage/wind volumes fell.
That is not because gas suddenly got popular, it’s because load growth and retirements are colliding with reality.
NERC’s long-term reliability view puts numbers behind the collision:
- Summer peak demand forecast: +132 GW over the next 10 years (2024 LTRA).
- Announced retirements: 115 GW over the 10 year period (includes many not yet formalized).
When demand accelerates and dispatchable exits, the market pays for what shows up reliably. The challenge is getting it interconnected.
FERC is trying to fix the pipe, but you still have to build in the real world
Order 2023 is a meaningful attempt to move the interconnection process from a serial, first-come approach to a cluster study process, add penalties for late studies, and tighten site control and financial readiness so the queue is less speculative.
It also pushes for more transparency, including a requirement that transmission providers maintain a public visual representation (a “heatmap”) of available transmission capacity.
All good. None of it changes the investor truth:
- Queue reform helps at the margin.
- It does not magically create transmission capacity.
- It does not erase network upgrade costs already embedded in constrained areas.
So your investment edge comes from how you source and structure projects in this environment.
The investor reality: you’re not buying a “site,” you’re buying a path to COD
In a queue-constrained market, the most valuable “asset” often isn’t the dirt. It’s the right to interconnect at a defined point, with defined obligations, and a plausible timeline.
There are multiple ways to get there. Here’s how I think about them.
Comparison table: 5 routes to megawatts (and what you’re really underwriting)
| Route | What you’re buying | Typical speed vs greenfield | Main risk you’re underwriting | Best fit for |
| 1) Greenfield interconnection request | A queue position and a hope | Slowest | Network upgrades, restudies, timeline drift | Low carry cost developers, long-dated capital |
| 2) Brownfield repower (retiring coal/oil/older gas) | Existing POI, often existing lines/substation | Faster | Legacy constraints, environmental cleanup, rebuild scope | Dispatchable replacement near load |
| 3) Acquire a late-stage queued project | Studies done, deposits paid, sometimes draft/executed IA | Often fastest | Transferability, milestone defaults, hidden upgrade scope | Investors who can diligence fast |
| 4) Use surplus interconnection service | Underutilized capacity at an existing POI | Fast (if feasible) | Operating limits, curtailment rules, study scope | Add-ons: peakers, engines, storage |
| 5) Behind-the-meter / private wire (limited export) | Serving load directly, minimizing grid reliance | Potentially fast | Contract structure, backup compliance, gas deliverability | Data centers, industrial, campuses |
On surplus interconnection service specifically: Order 845 pushed transmission providers to create processes so interconnection customers can use surplus interconnection service at existing interconnection points, plus provisional agreements and other tools to use underutilized interconnections.
The key is not memorizing the label. The key is asking: what is the cheapest, fastest interconnection path that still produces bankable revenue?
What “good” looks like for a gas project in 2026
A bankable gas development thesis now needs three things to be true at the same time:
- Interconnection is real
- Clear POI
- Clear upgrade scope and cost responsibility
- Clear milestones and a timeline you can defend
- Clear POI
- Fuel is real
- Deliverability during cold snaps and peak conditions
- Redundancy plan (dual feed, storage, firm transport, or at least a credible constraint story)
- Deliverability during cold snaps and peak conditions
- Revenue is real
- Not just merchant optimism
- A path to contracted cash flow, capacity value, or reliability-driven procurement
- Not just merchant optimism
Interconnection sits at the top because it is the gate. Without it, the rest is theoretical.
The diligence questions I care about (and why)
If you’re investing in gas development right now, this is the due diligence list that separates “projects” from assets.
Interconnection and deliverability
- What exactly is the POI and voltage level? (Substation is not a POI. The breaker position is.)
- What studies are complete and what’s still pending?
- Is there a draft or executed interconnection agreement? Berkeley Lab flagged 408 GW that already has a draft/executed IA but is not yet operating. That’s a big pool of “almost-there” assets and also a big pool of stranded capital.
- What network upgrades are assigned, and what can move?
- What triggers restudies? (Project changes, other withdrawals, cluster reshuffles)
- What are the operating constraints? (seasonal limits, curtailment rules, stability constraints)
Site control and permitting reality
- Is the land fully controlled for the full term you need (including laydown, gas lateral, water, transmission easements)?
- Is it greenfield, brownfield, or a repower? (Each has a different hidden cost profile.)
- Are there local constraints that kill schedules? (zoning fights, noise, traffic, wetlands, cultural resources)
Commercial structure
- Who is the buyer and what are you actually selling?
- Energy
- Capacity
- Reliability product
- A toll
- Energy
- What happens if COD slips? (liquidated damages, step-in rights, security postings)
- Is there a credible path to financing with this interconnection position?
A simple chart: time-to-revenue by development path
This is not a promise. It’s a practical way to visualize schedule risk.
Time to first meaningful revenue (illustrative)
Greenfield queue entry | ██████████ (4–6+ years)
Brownfield repower | ██████ (2–4 years)
Acquire late-stage queue/IA | ████ (1–3 years)
Surplus interconnection add-on | ███ (1–2 years)
Behind-the-meter (limited export)| ███ (1–2 years)
The point: in a world where median queue-to-COD is now over 4 years in many regions, speed is a feature you can’t fake.
What to do today: an investor checklist
If you want to deploy capital into gas (or gas-plus-storage) development without getting stuck in the queue swamp, here are practical steps you can take now.
- Screen markets by “interconnection pain,” not just power prices
- Where are queues longest?
- Where are upgrade costs most volatile?
- Where is load growth concentrated?
- Where are queues longest?
- Build a sourcing pipeline around existing interconnection
- Retiring thermal sites with strong POIs
- Projects with draft/executed IAs (but verify why they aren’t built)
- Underutilized interconnections where surplus capacity might exist
- Retiring thermal sites with strong POIs
- Treat interconnection diligence like title diligence
- You would never buy land without title work.
- Don’t buy a project without a full interconnection risk memo.
- You would never buy land without title work.
- Pre-negotiate control rights
- Step-in rights if milestones are missed
- Clear cure periods and reporting requirements
- Ownership of studies and queue position documentation
- Step-in rights if milestones are missed
- Decide your revenue model before you finalize the design
- Peaker vs combined cycle vs engines is a commercial decision as much as a technical one.
- It changes the value of the interconnection configuration and constraints.
- Peaker vs combined cycle vs engines is a commercial decision as much as a technical one.
Bottom line
Interconnection has become the controlling variable for new generation development. Berkeley Lab’s queue data shows a system with massive interest, high withdrawal rates, and multi-year timelines to COD.
If you want dispatchable megawatts in service on an investor timeline, you need to source projects where interconnection is already advanced or structurally advantaged, and you need to diligence it like it’s the asset. FERC’s reforms are directionally helpful, but the winners will still be the teams who buy time and certainty, not just steel and land.
What to watch next (5 bullets)
- How Order 2023 cluster studies and penalties actually change cycle times in each region.
- Whether transmission “heatmaps” become genuinely decision-useful, or just PR-grade visuals.
- The growing pool of projects with IAs but no COD (stranded development capital).
- Reliability-driven procurement and how it reshapes dispatchable investment returns (especially under tight reserve margins).
- M&A pricing for brownfield sites with strong POIs (this is becoming a distinct asset class)
Sources
- Lawrence Berkeley National Laboratory, Queued Up: 2025 Edition (data through end of 2024)
- FERC, Explainer on the Interconnection Final Rule (Order 2023 / 2023-A)
- NERC, 2024 Long-Term Reliability Assessment
FERC, Order 845 (surplus interconnection service and related reforms)
