How Gas Power Plants Actually Get Financed

Most guides on this topic open with electricity demand charts. Lenders do not. The first thing a project-finance desk or an SBA-backed lender wants to see is how the capital stack is structured, who carries the fuel and power-price risk, and whether the cash flows can cover debt service even in a weak year. This template is built around that question, because a gas plant is one of the most capital-intensive businesses a founder can attempt and the financing structure decides whether it ever breaks ground.

Two financing worlds matter here, and they rarely overlap. Utility-scale plants raise project finance, where a special-purpose vehicle borrows against the contracted cash flows of the plant itself, typically at 70% to 90% debt with the balance as sponsor equity. Smaller behind-the-meter gas gensets and modular peaker projects can sometimes be financed more conventionally, and in the United States projects under the SBA size thresholds occasionally use SBA 7(a) lending for the development company or for equipment.

The SBA route is worth understanding precisely because it is so often misapplied to this niche. SBA 7(a) loans are capped at $5 million and are designed for small businesses; they will not finance a 400 MW combined-cycle plant. They can, however, finance a development company, a behind-the-meter generation set serving a single commercial site, an equipment purchase, or the working capital of an operations-and-maintenance firm serving plant owners. The SBA reports its 7(a) approval data publicly through the agency's lending reports (U.S. Small Business Administration, 2025), and a plan that asks for SBA money should match the loan size, use of proceeds, and collateral to what the program actually funds rather than treating it as generic startup capital.

For everything above SBA scale, the realistic equity providers are infrastructure funds, energy-focused private equity, strategic utility partners, and increasingly large industrial or data-centre offtakers who will anchor a plant with a long-term power purchase agreement. The plan's job is to make the cash flows legible to those audiences and to show that the spark spread, capacity payments, and ancillary-services revenue together service the debt with margin to spare.

The Gas Generation Market in 2026

The natural gas power generation market was valued at roughly $96.95 billion in 2025 and is projected to reach about $122.49 billion by 2030, a compound annual growth rate of 4.8% (MarketsandMarkets, 2025). Within that, the combined-cycle gas turbine segment alone was estimated at $28.5 billion in 2025 and is forecast to reach $45.2 billion by 2034 at a 5.7% CAGR (Intel Market Research, 2025).

What sits behind that growth is a structural shift in how grids use gas. As variable renewable capacity has expanded, the value of gas has moved away from steady baseload generation and toward flexibility: the ability to ramp up quickly when the wind drops or demand spikes, and to provide the balancing and reserve services that keep a renewables-heavy grid stable. That has two consequences for a business plan. First, the most attractive new gas assets are often flexible peakers and fast-start units rather than always-on baseload plants. Second, a growing share of revenue comes from capacity markets and ancillary services rather than bulk energy sales, which changes the entire financial model.

Ownership tells the same story. In the United States, electric utilities own about 53% of the natural gas generation fleet while independent power producers own roughly 39% (U.S. Energy Information Administration, 2025). New entrants almost always position as IPPs, selling into wholesale markets or under contract, rather than as regulated utilities. In the UK, the equivalent demand driver is the Capacity Market, which pays generators to be available at times of system stress, and the rise of flexible gas plant alongside battery storage to firm up an increasingly renewable grid.

The headline figure that matters most to a UK or US founder is not the global market size at all. It is the regional wholesale power price, the prevailing gas price, and the resulting spark spread in the specific market the plant will sell into, whether that is ERCOT or PJM in the United States or the GB wholesale market in Britain. A plan that quotes a global billion-dollar figure but cannot state the spark spread in its own market is not yet financeable.

Capital Costs by Plant Type

Gas plant economics begin with capital cost per kilowatt, and that number has moved sharply. BloombergNEF reported that the all-in cost of building a new combined-cycle gas plant reached about $2,157 per kW in 2025, a rise of roughly 66% from under $1,500 per kW in 2023, driven by turbine supply shortages, longer order backlogs, and construction inflation (BloombergNEF, 2026). Simple-cycle combustion-turbine peakers carry lower build costs per kW, with EIA generator cost data and Gas Turbine World benchmarks placing them in the region of $728 to $1,544 per kW, up from around $562 per kW in 2023 (U.S. Energy Information Administration, generator cost data).

Translating per-kW figures into a project budget is where founders most often go wrong. At $2,157 per kW, a 400 MW CCGT is a project of roughly $863 million before financing costs, while a 49 MW flexible peaker at the lower combustion-turbine range is a far more approachable figure for a first-time developer. The template forces you to state the plant rating in MW, the technology, and the per-kW source, then derives the total rather than starting from a guessed lump sum.

The line items lenders examine first

• Turbine and generating equipment: $728 to $2,157 per kW depending on whether the plant is simple cycle or combined cycle, and on the original equipment manufacturer. Lead times from major OEMs such as GE Vernova have lengthened, so a credible plan addresses turbine availability, not just price.
• Engineering, procurement and construction (EPC): typically 25% to 35% of the project, usually delivered under a fixed-price turnkey contract that lenders treat as central to the risk story.
• Grid interconnection: the study itself plus any network upgrades the grid operator requires, ranging from around half a million dollars to tens of millions for a poorly located site. This is frequently the single biggest controllable risk in the budget.
• Gas supply connection: a pipeline lateral and metering, with the connection itself requiring FERC approval under the Natural Gas Act in the US or a connection to the National Transmission System in the UK.
• Environmental permitting and air-quality controls: from $1 million to well over $10 million when selective catalytic reduction or other emissions controls are required.

Beyond the build, the model needs an operating reserve and a fuel working-capital line, because a gas plant buys fuel before it sells power and the timing gap has to be funded. Lenders will also expect a debt-service reserve account, usually six months of debt service held in escrow, which sits on the funding requirement even though it never gets spent in a healthy year.

Revenue: Spark Spread and Capacity Payments

A gas plant earns money in three distinct ways, and a plan that models only the first will not survive lender scrutiny. The streams are energy sales priced off the spark spread, capacity-market payments for being available, and ancillary or balancing services that the grid operator pays for fast, flexible response.

The spark spread is the foundation. It is the wholesale electricity price the plant receives, minus the cost of the natural gas needed to generate that electricity, adjusted for the plant's heat rate (its fuel efficiency). When the spread is wide, the plant runs and earns; when it is narrow or negative, an economic operator does not dispatch. Because the spread moves with both power and gas markets, lenders want to see the revenue model stress-tested across a range of spreads rather than a single optimistic point estimate.

A worked example

Consider a 50 MW flexible peaker running at a 15% capacity factor, which is typical for a plant designed to cover demand spikes rather than baseload. At 15%, the plant generates roughly 65,700 MWh per year. At a $35 per MWh spark spread, that is about $2.3 million of gross energy margin before fixed operations and maintenance. Crucially, that energy margin is not the whole story: the same plant earns separate capacity-market payments simply for being available during system-stress periods, plus ancillary-services revenue for providing reserve and frequency response. For a peaker, those availability payments often exceed the energy margin.

Now contrast a 400 MW combined-cycle plant running at a 55% capacity factor as mid-merit generation. That plant produces roughly 1.93 million MWh per year, so even a modest per-MWh spread produces a very different revenue base, but it also carries far higher fuel exposure and a much larger capital cost to service. The two business models are not variations on a theme; they are different businesses with different risk profiles, and the template keeps them separate.

Net margins for merchant gas plants commonly fall in the 5% to 21% range, but that band hides enormous variation depending on whether the plant sells merchant power or operates under a contract. A tolling agreement, where a creditworthy counterparty pays a fixed capacity fee for the right to dispatch the plant and takes on the fuel and power-price risk, converts a volatile merchant revenue line into a contracted one. Lenders love tolls because they make the cash flows predictable, and a plan that can present a credible tolling counterparty or a long-term power purchase agreement will be financed on materially better terms than a pure merchant project.

The revenue section should therefore do four things: state the market and its current spark spread, define the run profile and capacity factor, layer in capacity and ancillary revenue explicitly, and show the debt-service coverage ratio under a downside case. A model that does only the first of those is the single most common reason a gas plant plan stalls at the term-sheet stage.

CCGT vs Peaker vs Behind-the-Meter

Before writing a single financial assumption, decide which of three business models the plant represents. Each has a different capital intensity, a different revenue mix, and a different financing route, and conflating them is the fastest way to produce a plan that no lender can underwrite.

The behind-the-meter model deserves particular attention in 2026 because surging electricity demand from data centres has created a class of large, creditworthy private offtakers willing to contract for dedicated on-site generation. That demand has made smaller, single-customer gas projects financeable in ways that pure merchant projects are not, because the revenue is contracted to a named counterparty rather than exposed to wholesale price swings. A first-time developer is far more likely to reach financial close on a contracted behind-the-meter or tolled peaker than on a speculative merchant baseload plant.

Who Buys the Power and How You Contract It

A gas plant does not have customers in the everyday sense; it has counterparties, and the credit quality of those counterparties shapes the financing more than almost anything else. The plan needs to name who buys the electricity, who pays for availability, and what contract binds them. Three structures dominate, and most financeable projects use a blend.

• Wholesale market sales (merchant): the plant bids into the regional power market (PJM, ERCOT, or the GB market, for example) and earns the clearing price whenever it dispatches. This is the highest-upside and highest-risk route, because revenue swings with the spark spread and lenders discount it heavily.
• Power purchase agreement (PPA): a long-term contract with a utility, a corporate buyer, or a load-serving entity to take a fixed volume of power at an agreed price. A PPA with a creditworthy buyer turns volatile merchant revenue into a bankable cash flow and is often the difference between a financeable project and a speculative one.
• Tolling agreement: a counterparty pays a fixed capacity fee for the right to dispatch the plant and supplies the fuel themselves, taking the spark-spread risk. The plant owner earns a predictable fee regardless of how often the plant runs, which is the most lender-friendly structure of all.

For a behind-the-meter project the customer is a single named offtaker, most often a large industrial site or a data centre, contracting for on-site generation to secure reliable, cost-stable power and resilience against grid outages. The surge in data-centre electricity demand through 2025 and 2026 has made these single-customer deals far easier to finance, because the revenue is anchored to one creditworthy buyer under a long-term contract rather than exposed to wholesale price swings. The plan should quantify the offtaker's load profile, the contract term, and the credit rating of the counterparty, because that is precisely what a lender will underwrite.

Whatever the structure, the offtake section should state the contracted share of revenue versus the merchant share, the weighted-average contract tenor, and the counterparty credit profile. A plan that can show 60% or more of revenue under contract with an investment-grade counterparty will be financed on far better terms than a fully merchant project of the same size.

Operations, Availability and Risk

Once a gas plant is built, its economics live or die on two operational metrics: availability and heat rate. Availability is the share of time the plant can run when called upon, and for a peaker that earns capacity payments it is the metric the contract is built around; a plant that fails to start when dispatched can face penalties that erase a season's margin. Heat rate measures how efficiently the plant converts fuel to electricity, and a small improvement compounds across every megawatt-hour the plant produces.

The operating model centres on a long-term service agreement with the turbine manufacturer, typically GE Vernova, Siemens Energy, or Mitsubishi Power for large frame machines, or Wartsila and Clarke Energy for reciprocating-engine plant. These agreements cover scheduled maintenance, major-inspection intervals, and parts, and they are a substantial fixed cost that the financial model must carry from year one. Staffing for a modern gas plant is lean, often a small core team of operators and technicians supported by remote monitoring, but the plan should still show a credible organisational structure and the key hires needed before commercial operation.

The risk section is where lenders spend their time, and it should address each major exposure with a named mitigant:

• Fuel price risk: mitigated by a gas supply agreement, a hedging policy, or a tolling structure that passes fuel risk to the counterparty.
• Power price risk: mitigated by a PPA or by the contracted capacity-market and ancillary revenue that provides a floor under merchant exposure.
• Construction risk: mitigated by a fixed-price, date-certain EPC contract with liquidated damages, plus a contingency line in the budget.
• Counterparty risk: mitigated by contracting with investment-grade buyers and by credit support such as parent guarantees or letters of credit.
• Regulatory and carbon risk: mitigated by securing permits with comfortable emissions headroom and by stress-testing the model against carbon-price scenarios.

A plan that names these risks and pairs each with a concrete mitigant reads as the work of a developer who has been through a financing before. That credibility is worth real basis points on the cost of debt.

Permits, Interconnection and Consents

Permitting is not paperwork that happens after the decision to build; it is the critical path that determines whether the project happens at all. The plan must name the specific approvals, the agencies that grant them, and realistic cost and timeline ranges for each jurisdiction it operates in.

United States

• Clean Air Act air permit (PSD or Title V): issued by the EPA and the relevant state permitting authority. Expect $1 million to over $10 million once emissions-control equipment is included, and a 12 to 24 month timeline.
• Grid interconnection agreement: governed by FERC and the regional ISO or RTO. The interconnection study plus any network upgrades can run from $0.5 million to over $50 million, and the queue process commonly takes one to four years, often the single longest item on the schedule.
• Natural gas pipeline connection approval: under the Natural Gas Act, also at FERC, with a project-specific cost and a one to three year timeline.

United Kingdom

• Electricity Generation Licence: granted by Ofgem, with ongoing licence conditions; the application itself runs weeks to months.
• Development Consent Order (NSIP): required for plants over 50 MW in England (350 MW in Wales), decided by the Planning Inspectorate and the Secretary of State at DESNZ under National Policy Statements EN-1 and EN-2. Expect 12 to 24 months or more.
• Environmental Permit: issued by the Environment Agency (or Natural Resources Wales), covering emissions and operating conditions, with a permit fee plus compliance cost and a multi-month timeline.

India (a third jurisdiction)

For founders developing in India, the approval set differs again. A new gas plant typically needs techno-economic concurrence from the Central Electricity Authority (CEA), environmental clearance from the Ministry of Environment, Forest and Climate Change (MoEFCC), a gas allocation, and a power purchase agreement, with tariff approval through the central or state electricity regulatory commission (CERC or SERC). The fuel-allocation step is distinctive and frequently the binding constraint, because domestic gas supply is rationed and an imported-LNG plant carries a different cost base entirely.

Across all three jurisdictions the lesson is the same: permitting and interconnection routinely take longer than construction, so the plan should present a single integrated critical-path timeline that sequences development, financing, and build rather than assuming they overlap neatly.

Mistakes That Kill the Financing

Across gas-plant plans we have reviewed, the same handful of errors recur, and each one is enough to stall a deal at the term-sheet stage. They are easy to avoid once you know lenders look for them.

• Underestimating the interconnection queue. Founders model a two-year build and forget that the grid interconnection study and any network upgrades can take one to four years and cost tens of millions. The plan should treat interconnection as the critical path and budget for it explicitly.
• Modelling baseload economics for a peaker. A plant designed to run a few hundred hours a year cannot be financed on a 55% capacity-factor revenue line. Match the capacity factor to the actual operating role, then build revenue from there.
• Ignoring capacity and ancillary revenue. For a flexible plant, capacity-market payments and balancing-services income often exceed the energy margin. Leaving them out understates revenue and signals that the developer does not understand how modern gas assets earn.
• No fuel hedging or supply plan. A plant that buys gas at spot prices and sells power at spot prices is fully exposed to spark-spread volatility. Lenders want to see a fuel supply agreement, a hedging policy, or a tolling structure that transfers that risk.
• Treating permitting as a formality. Air permits, environmental consents, and the UK Development Consent Order are multi-year processes with real refusal risk. A plan that assumes automatic approval is not credible to anyone who has financed energy infrastructure.

The common thread is that gas-plant financing rewards realism. A conservative plan that models a downside spark spread, a long interconnection timeline, and a contracted revenue floor will raise money more easily than an aggressive plan that assumes everything goes right.

Sample Plan Preview

Here is an extract from the executive summary of a gas peaker plan built on this template, showing the level of specificity lenders expect.

The full template carries this specificity through every section, from the customer and offtake analysis to the five-year financial statements, so the finished plan reads as the work of an operator who understands how the asset actually earns.

What's in the Template

The gas power plant business plan template gives you a complete, investor-ready structure with prompts and worked examples tailored to power generation:

• Executive Summary – the plant, the market, the ask, and the revenue model in 60 seconds
• Company & Project Overview – SPV structure, sponsor experience, site, and technology choice
• Market & Price Analysis – regional wholesale prices, gas prices, and the spark spread
• Offtake & Revenue Strategy – merchant, PPA, or tolling, plus capacity and ancillary revenue
• Permitting & Interconnection Plan – named approvals, agencies, and the critical-path timeline
• Operations & Maintenance – staffing, OEM service agreements, and availability targets
• Risk & Mitigation – fuel, price, construction, and counterparty risk with mitigants
• Management & Sponsor Team – operating track record and key development hires

The optional Financial Forecast add-on (included in our $300/£250 and $1,000/£800 packages) provides a 5-year Excel model with income statement, cash flow, balance sheet, spark-spread sensitivity, debt-service coverage, and a startup capital requirements table. You can also explore our market research and content service, browse the full library of free business plan templates, or read about working with an Avvale business plan writer.

Frequently Asked Questions

Useful Links & Resources

Related Avvale guides and reference sources for power generation founders:

• Free business plan templates library
• Bespoke business plan service
• Avvale client case studies
• Biomass power plant business plan template
• Battery energy storage system business plan template
• U.S. EIA generator cost data