EIA data shows utility-scale solar levelized power purchase agreement (PPA) prices fell to roughly $24-$26 per MWh in 2024, compressing contracted margins and pushing more developers toward partial-merchant or shorter-tenor structures. That single price signal reshapes merchant solar project finance underwriting: lenders now model hub prices, basis risk, and curtailment as first-order variables rather than downside stress cases. This piece maps the 2026 framework debt arrangers apply to uncontracted megawatt-hours across the Electric Reliability Council of Texas (ERCOT) and the California Independent System Operator (CAISO).
What distinguishes merchant solar project finance underwriting from contracted PPA cash flows
Contracted solar is a fixed-price annuity: an offtaker pays a set $/MWh for 15 to 25 years, with debt sized at 1.25x DSCR (debt service coverage ratio). Merchant solar project finance underwriting replaces that annuity with hourly wholesale settlement, capacity payments, and REC (renewable energy certificate) monetization modeled on P50 (50th-percentile) and P90 (90th-percentile) production bands over a 25-year horizon.
The consequence for lenders is that revenue certainty collapses. A contracted asset has a single revenue line; a merchant asset has hundreds of hourly price paths. Merchant solar project finance underwriting therefore builds around three axes: hub price forecasts published by consultancies such as Wood Mackenzie, basis differential between the project node and the trading hub, and curtailment probability driven by grid congestion. Each axis needs its own sensitivity band and its own reserve.
The final distinction is tenor. Contracted PPA debt often matches the offtake, running to 18 years. Merchant debt rarely extends past 7 years without a hedge overlay because refinancing risk absorbs the tail. Recent EIA Electricity Annual data confirms that PPA pricing compression has pushed more projects into this shorter-tenor debt structure. See our C&I PPA bankability guide for the parallel logic in commercial deals.
Hedging tools reshape merchant solar project finance underwriting
Proxy revenue swaps and fixed-for-floating contracts are the two instruments carrying merchant solar project finance underwriting into 2026. A proxy revenue swap pays the project a fixed $/MWh in exchange for actual settled revenue, transferring hourly price and shape risk to a bank counterparty. A fixed-for-floating swap does the same on price only, leaving shape and volume risk with the project.
For a lender, the hedge converts merchant exposure into synthetic contracted cash flow for the hedged portion. Typical structures cover 60-80% of P50 generation over 7-10 years, leaving the remainder as merchant tail. Reports available through the FERC electric power markets page track the growth of these instruments across utility-scale solar transactions. Basis risk is the residual that neither instrument fully eliminates. A West Texas solar project settles at a local ERCOT node that can diverge from the North Hub by $10-$20/MWh during congestion events, particularly during summer peak demand or transmission outage windows. Merchant solar project finance underwriting requires a discrete basis reserve or a bespoke basis swap layered on top of the price hedge. In a 2024 West Texas transaction we arranged at SunRaise Capital, the project node settled $17/MWh below the North Hub for three consecutive quarters, exhausting the initial basis reserve before month 18 of commercial operation. That outcome reshaped our credit model: we now require a 24-month lookback of nodal divergence as the reserve sizing floor, not the trailing 12-month average that was standard practice through 2022. The combined hedging and reserve cost typically absorbs 150 to 300 basis points of gross project IRR.
ERCOT and CAISO market dynamics drive revenue forecasts
CAISO market operations reports show solar curtailment at 2.8 million MWh in 2023, roughly 6% of California generation, a figure now standard in merchant revenue downside models. ERCOT's energy-only structure delivers scarcity-event upside with no capacity payment; revenue variance runs wide. CAISO adds resource adequacy revenue but exposes projects to negative pricing as midday solar output floods intervals.
The curtailment trajectory has moved unfavorably: installed solar capacity in California's Central Valley and Los Angeles basin continues to outpace transmission upgrades in successive interconnection queues. Utility Dive and CAISO quarterly reports have tracked this pattern across multiple years. Lenders in 2026 treat 5-10% curtailment as the base-case assumption for California projects, versus 3-6% for ERCOT West Texas nodes.
NREL's 2025 Annual Technology Baseline, available at the NREL research portal, projects merchant capacity factors between 21% and 29% depending on region. This band is the primary P50/P90 input for revenue sensitivity models. A West Texas project sits at the top; a Northeast project sits at the bottom. The gap drives regional risk premia in debt pricing and reserve sizing.
DSCR floors and reserve structures in merchant solar project finance underwriting
Debt service coverage ratio floors on merchant deals sit meaningfully above contracted norms. Where a contracted PPA deal clears at 1.25x P50 DSCR, merchant solar project finance underwriting requires 1.40x-1.60x on a P90 case, with some lenders demanding 1.75x on fully unhedged structures. The uplift compensates for revenue variance, basis risk, and refinancing risk in the merchant tail.
Reserve structures scale in parallel. A typical merchant deal carries a 6-month debt service reserve, a 6-12 month O&M reserve, a major maintenance reserve funded from cash flow, and often a dedicated basis differential reserve. For projects using proxy revenue swaps, a hedge collateral posting requirement adds another cash sink that lenders size against mark-to-market volatility.
Cash sweeps become mandatory below trigger DSCRs (commonly 1.30x). This mechanism captures upside during strong pricing years to prepay principal and lower refinancing exposure. Lender models tested against SEIA solar market data now assume a 40-60% cash sweep whenever the trigger is breached. See our construction bridge financing guide for how these reserves layer with pre-COD debt.
How PPA tenor compression is changing merchant solar project finance underwriting
Corporate PPAs that ran 15 years in 2019 now clear at 7-10 years, and utility PPAs are following the same path. This shortening pushes more of the project's economic life into merchant exposure. A 10-year PPA on a 25-year asset leaves 15 years of uncontracted output that lenders must price into the base case, not the sensitivity.
The response is a two-tier debt structure. A fully amortizing tranche sized to the contracted period sits senior; a mini-perm or back-use tranche sized against merchant cash flows sits behind it. Some 2025-2026 deals split further, creating a hedge-covered middle tranche between the two. This layering makes merchant solar project finance underwriting a three-book exercise rather than a single sizing calculation. Institutional appetite for the merchant tail, priced correctly, is where the additional yield lives in a compressed rate environment.
| Feature | Contracted PPA | Merchant |
|---|---|---|
| Typical DSCR floor | 1.25x P50 | 1.40x-1.60x P90 |
| Debt tenor | 15-20 years | 7-10 years |
| Debt service reserve | 6 months | 6-12 months |
| Cash sweep trigger | Rare | Below 1.30x |
| Basis reserve | Not required | Standard |
Rating agency methodology has adjusted too. The framework outlined in our residential solar ABS rating methodology note extends to utility-scale merchant tranches: hedge coverage ratio, basis reserve size, and refinance sensitivity all drive final rating. For deals paired with batteries, our standalone storage underwriting guide details the incremental capacity revenue treatment that layers on top of the solar merchant case.
Frequently asked questions
What DSCR do lenders require for merchant solar project finance underwriting?
Merchant solar deals typically require 1.40x-1.60x DSCR on a P90 case, compared to 1.25x on P50 for contracted PPAs. Unhedged structures can push floors to 1.75x. The uplift covers hourly price variance, basis risk, and refinance risk on 7-10 year tenors that leave 15+ years of asset life uncontracted. Reserve requirements scale in parallel: 6-12 month debt service reserves, dedicated basis reserves, and cash sweeps triggered below 1.30x. FERC market reports confirm the pattern across ERCOT and CAISO markets, and lender term sheets in 2026 rarely deviate outside these bands without material hedge coverage in place.
How does a proxy revenue swap work in solar project finance?
A proxy revenue swap pays the project a fixed dollar-per-megawatt-hour amount in exchange for the actual hourly settled revenue at the project node. The bank counterparty absorbs price shape, volume, and in most structures basis risk on the hedged portion. Typical coverage is 60-80% of P50 generation for 7-10 years. For the sponsor, the hedge converts merchant exposure into synthetic contracted cash flow, allowing more debt to be sized against those hedged MWh. Wood Mackenzie tracks the growth of these instruments in its solar market service coverage.
What curtailment risk should merchant solar underwriting assume?
CAISO market operations reports document more than 2.8 million MWh of solar curtailment in 2023, roughly 6% of solar generation lost to grid conditions. Lenders now stress-test 5-10% curtailment in the base case for California projects and 3-6% for ERCOT West Texas nodes. The trajectory is unfavorable as installed capacity outpaces transmission upgrades. Basis reserves and shape hedges partially address the exposure, but a project without transmission upgrades in its interconnection queue carries residual risk that flows directly into pricing and reserve sizing decisions.
What capacity factor do lenders model for merchant solar projects?
NREL's 2025 Annual Technology Baseline projects merchant solar capacity factors between 21% and 29% across major U.S. regions. Lenders use 25% as a rough national P50 midpoint, with regional adjustments: 27-29% for West Texas and the desert Southwest, 22-24% for the Mid-Atlantic, and 21-23% for the Northeast. A P90 case typically applies a 10-12% haircut to P50 output for annual variability, plus a curtailment overlay. These inputs feed directly into revenue sensitivity models and DSCR sizing decisions in every merchant deal that reaches credit committee.
How does ERCOT differ from CAISO for merchant solar revenue?
ERCOT runs an energy-only market with high scarcity pricing but no capacity payment. Solar projects capture revenue almost entirely from hourly energy, which delivers upside during heat events but leaves wide monthly variance. CAISO offers resource adequacy payments and structured day-ahead energy but exposes projects to negative pricing when midday solar floods intervals. Underwriting models for the two markets diverge on revenue mix: ERCOT weights scarcity value; CAISO weights capacity revenue and curtailment reserves. SEIA market data details deployment patterns and clearing prices in each market at a regional level.
Are hedged merchant tranches rated similarly to contracted PPAs?
Not fully, but the gap has narrowed. Rating agency methodology, covered in depth by Asset Securitization Report, treats hedge-covered tranches with a hybrid approach: hedged MWh receive credit at contracted-equivalent DSCR levels while the uncovered tail receives merchant stress. Hedge counterparty credit quality, coverage ratio, and hedge tenor drive the notching. In practice, a 2026 deal with 70% P50 hedge coverage and a BBB-rated bank counterparty might carry a BBB- senior tranche for the hedged years and a BB back-use tranche for the merchant tail beyond hedge expiry. The pricing gap between those two tranches reflects the residual revenue variance that neither the hedge nor the reserve can fully absorb.