Project finance has its own version of a quiet market correction. It does not show up as a headline or a rate announcement. It shows up when a deal that should have closed does not, or when the terms that come back from a lender look nothing like the term sheet from eighteen months ago. That is where a meaningful share of the clean energy infrastructure market finds itself in mid-2026.
The credit standards governing project finance shifted at least three times in the past two years, each time in ways that narrowed the field of bankable projects. Most of those shifts happened fast enough that financing models built in 2022 or 2023 never got updated to reflect them. For companies that locked in energy infrastructure commitments during the Inflation Reduction Act (IRA) enthusiasm cycle and are now moving toward construction or debt closing, the gap between what they modeled and what the market currently requires is becoming difficult to ignore.
How Tax Credit Eligibility Became a Lender Requirement, Not Just a Developer Concern
The One Big Beautiful Bill Act (OBBBA), signed in July 2025, narrowed the tax credit base for wind and solar projects in ways that changed lender behavior directly. Under the updated rules, projects must begin construction by July 4, 2026, to qualify for the production tax credit (PTC) or investment tax credit (ITC) under the new Section 45Y and 48E framework. Internal Revenue Service (IRS) guidance issued after the OBBBA also clarified that preliminary activities including permitting, site work, and financing arrangements no longer count toward the construction start. Developers now need to demonstrate substantial physical work or meet a 5% safe harbor for smaller facilities.
That guidance change landed directly on lenders. Tax equity is a core component of project finance for wind and solar. If tax credit eligibility is uncertain, tax equity does not close, and if tax equity does not close, the rest of the capital stack does not close either. According to analysis from Mintz, lenders responded by tightening collateral requirements and shifting preference toward borrowers with diversified portfolios and revenue-generating assets already in operation. Pre-notice-to-proceed (pre-NTP) financing, which funds the development activities before construction begins, surged past $7 billion in the first half of 2025 as developers raced to meet the new eligibility thresholds. The pricing on that capital reflected the risk: spread levels of SOFR plus 350 to 650 basis points, with borrowers carrying stronger balance sheets receiving the lower end of that range.
The Credit Box Shrank. Most Infrastructure Pro Formas Did Not Adjust.
The tax credit issue is one part of the picture. The broader shift is in what lenders will accept as sufficient certainty to fund a project. Rocky Mountain Institute (RMI), which convened 70 practitioners from green banks, commercial lenders, and institutional investors in late 2025, puts the core problem plainly: most climate finance work is about moving assets into the credit box, and the credit box has gotten smaller. Projects fall outside it for three reasons: insufficient cash flow predictability, counterparty credit quality that does not meet lender thresholds, or deal structures that are too small or fragmented to fit standard underwriting.
Federal support programs were designed to address all three. With those programs receding, the gap between a project that works on paper and one that a lender will actually fund has widened considerably for anything outside the dominant thesis around large-scale, anchor-tenant power demand. Mid-sized clean energy deals, distributed projects, and anything carrying meaningful permitting exposure are competing for a thinner pool of capital from private credit funds, ESG-oriented lenders, and family offices. That market grew in 2025, but it prices risk more conservatively than the institutional capital it partially replaced.
Permitting Exposure Is Now Priced as Credit Risk, Not Schedule Risk
A February 2026 Crux survey of 50 clean energy developers found that federal permitting materially affected every respondent's project portfolio over the prior twelve months, with roughly 11 gigawatts (GW) of capacity affected in aggregate. The most common cost impact was a 6-10% increase in total project development costs, reported by 58% of respondents. For a 100-megawatt (MW) solar project, Crux estimates that range translates to $10 to $14 million in additional costs.
Lenders have started pricing that exposure differently. Permitting timelines that cannot be bounded introduce schedule risk that is genuinely difficult to underwrite, because an open-ended process means construction financing could be drawn and carrying interest costs while the project sits in regulatory review. That dynamic now shows up in deal structures as tighter construction period covenants, more conservative draw schedules, and in some cases lender requirements for permitting risk coverage before debt commitments are made. The project that assumed a standard permitting timeline in a 2022 financial model is a different credit today, even if nothing else about the project changed.
What Finance Teams Need to Look at Before the Capital Planning Window Closes
The companies most exposed are those that made infrastructure commitments during the IRA enthusiasm period, modeled financing on the tax credit structures and federal support programs in place at the time, and have not revisited those assumptions since. The credit market those models were built for has moved on three axes at once: tax credit eligibility narrowed, collateral and documentation requirements tightened, and permitting exposure is now a credit variable with direct pricing implications rather than a scheduling footnote.
Revisiting the financing model is not a concession that the project is in trouble. It is recognizing that the market a project was designed for and the market where it will actually close are now different. Companies that have that conversation before reaching the debt closing table have options. The ones that discover the gap at closing have considerably fewer of them.
