Project Finance and Non-Recourse Structures
Project finance funds a single asset — a toll road, a power plant, a wind farm — with debt that can only be repaid from that asset's own cash flows, so lenders analyze one project in isolation instead of the sponsor's whole balance sheet.
A utility company wants to build a $2 billion offshore wind farm. It could borrow the money on its own corporate balance sheet, the way it funds everything else — but that would put the entire company's credit at risk if the wind farm underperforms, and would show up as $2 billion of new debt weighing on every other project the company wants to finance afterward. Project finance offers a different route: create a standalone legal entity that owns nothing but the wind farm, have that entity borrow the money, and structure the loan so that lenders can only ever be repaid from the wind farm's own cash flows — never from the parent company's other assets. If the project fails, the lenders' recourse stops at the project's own balance sheet.
Why "non-recourse" changes everything about the analysis
In an ordinary corporate loan, a bank lending to a diversified company is really betting on that company's overall creditworthiness — one bad division can be offset by others, and the company has many potential sources of cash to service debt. In project finance, the special purpose vehicle (SPV) that borrows the money owns exactly one thing and has exactly one source of cash: the project itself. There is no diversification to hide behind, so lenders analyze the single asset with an intensity closer to underwriting a bond backed by one specific, immovable piece of collateral than to a normal corporate credit decision.
This concentration is precisely why project finance debt is structured around a strict cash flow waterfall: revenue comes in, and it flows through a fixed, contractually defined order — operating expenses first, then debt service (interest and scheduled principal), then reserve accounts topped up to required levels, and only after all of that is satisfied can any cash be released up to the equity sponsors as a distribution. Lenders don't just want to be paid; they want to be paid before equity gets a cent, every single period.
| Feature | Corporate finance | Project finance |
|---|---|---|
| Borrower | The whole operating company | A single-purpose entity (SPV) owning one asset |
| Recourse | Lenders can claim against the company's general assets | Non-recourse — lenders' claim stops at the project's own assets and cash flows |
| Analysis focus | Overall company creditworthiness, diversified cash flows | One project's contracted revenue, construction risk, and operating performance |
| Off-balance-sheet? | No | Often, for the sponsor — the debt sits at the SPV level |
| Typical use | General corporate purposes | Toll roads, power plants, pipelines, wind and solar farms, PPPs |
Because there is no diversification and no recourse to a parent's broader balance sheet, project finance lenders substitute contracts for diversification: long-term power purchase agreements, construction contracts with fixed prices and completion guarantees, and minimum revenue guarantees from a government counterparty all exist to remove as much uncertainty as possible from the single cash flow stream the debt depends on.
A worked scenario: the debt service coverage ratio through the project's life
A toll road SPV borrows $800 million to build and operate a highway concession, secured entirely against future toll revenue, with no recourse to the construction sponsor's other assets. The loan documents require a minimum debt service coverage ratio (DSCR) — cash available for debt service divided by the debt payment due — of 1.30x every period, meaning the project must generate 30% more cash than it strictly needs to make its loan payments, as a cushion.
Construction phase. The road doesn't exist yet, so there's no toll revenue and no DSCR to measure — the SPV draws down the loan in stages against construction milestones, and lenders rely on a completion guarantee from the construction contractor rather than project cash flow, because there isn't any yet.
Early operations, a mild recession. The road opens, but a recession reduces freight and commuter traffic below the base case forecast. Cash available for debt service comes in at 1.15x the required payment — below the 1.30x covenant. This triggers a cash trap: the waterfall automatically restricts distributions to equity sponsors, forcing extra cash to build up in a reserve account rather than being paid out, even though the project is still current on its debt. Lenders are protected before the project actually defaults.
Mature operations, traffic recovers. Several years in, traffic settles at or above the original forecast and the DSCR climbs to 1.45x. The cash trap releases, sponsors resume receiving distributions, and — if the loan structure allows it — the SPV may refinance at a lower margin now that a real operating track record exists, since the project has demonstrated it can service its debt through a downturn.
The DSCR covenant is where project finance actually earns its reputation for discipline: it acts automatically, restricting cash to equity the moment performance dips below plan, well before the project would ever miss an actual debt payment — turning what would be a lender's after-the-fact problem into a built-in, self-enforcing early warning system.
Further reading
- Yescombe, Principles of Project Finance
- Fabozzi & de Nahlik, Handbook of Structured Finance (project finance chapters)