The Capital Stack in Development Finance

Developers talk about "the loan," but projects are financed by a stack: layers of capital with different costs, different rights, and a strict order of repayment. Understanding the stack is not academic. It determines how much cash equity a project actually needs, what happens when something goes wrong, and which financing gaps have structural answers.

This guide walks the layers from the most senior position to the first loss, then covers the agreements that govern how the layers coexist. Specific structures (how much of a stack any layer can represent, and at what cost) vary by project, sponsorship, and capital source, and are confirmed during project review.

Senior debt

The senior construction or bridge loan sits at the top of the repayment order and holds the first-priority lien on the property. It is the largest layer in most stacks and the cheapest, precisely because it is the most protected: every other layer absorbs losses before the senior lender does. In exchange for that protection, senior lenders exercise the most control over the project's mechanics: the budget, the draw process, the completion obligations, and the conditions under which everyone else in the stack can act.

Senior proceeds are sized against cost and value together, which is why the loan-to-cost and loan-to-value measurements matter before any subordinate layer is discussed. The senior loan defines the space the rest of the stack must fill. Its documents define what kinds of capital are even permitted to fill it, which is why the senior term sheet should be read with the whole stack in mind, not just the loan it describes.

Stretch senior

A stretch senior facility is a single loan that reaches modestly deeper into the stack than conventional senior debt: one lender, one set of documents, covering ground that might otherwise require a separate mezzanine piece. The appeal is simplicity: no intercreditor negotiation, one draw process, one workout counterparty if things get difficult. The trade is pricing and structure: the lender is taking subordinate-layer risk inside a senior instrument and prices the blended exposure accordingly. For sponsors who value execution speed over optimized cost of capital, the trade is often worth it.

Mezzanine debt

Mezzanine debt sits behind the senior loan and is secured not by the property but by a pledge of the equity interests in the entity that owns it. That distinction is the whole design: if the mezzanine loan defaults, the remedy is a foreclosure on the ownership interests; the mezzanine lender can step into the sponsor's shoes while the senior loan remains in place on the property.

For a developer, mezzanine capital converts part of the equity requirement into fixed-cost financing. It raises the total leverage on the project and adds a second set of covenants and reporting obligations. Whether that trade strengthens or strains a project depends on the exit: mezzanine debt behind a clean, conservative takeout is a tool; behind a hopeful one, it is pressure.

Preferred equity

Preferred equity invests in the ownership entity itself, senior to the sponsor's common equity but junior to all debt. It earns a priority return (sometimes paid currently, often accrued to exit), and its remedies live in the entity's operating agreement rather than in a mortgage: failure to meet the preferred return or a capital event by an agreed date can shift control of the entity.

Because it is equity rather than debt, preferred capital can sometimes be added where loan documents prohibit further borrowing. It is a frequent answer to a capital-stack gap discovered after the senior loan is sized, and a common vehicle for a partner recapitalization when one investor needs liquidity and the project does not.

Common and joint-venture equity

Common equity is the sponsor's capital and any joint-venture partner's: the first money in, the last money out, and the layer that owns whatever the project earns beyond every senior claim. JV equity arrangements bring their own architecture: contribution schedules, promote structures that reward the sponsor for performance, and governance rights that decide who controls major decisions. None of that architecture changes the layer's position in the stack: common equity takes the first loss, which is why it demands the highest return and the most conviction.

Sponsor equity does not have to be all cash. Documented land basis, a site bought early and carried through entitlements, is regularly recognized as contribution, which is one of the quieter advantages of disciplined land acquisition. In JV structures, expect the capital partner to underwrite the sponsor as carefully as the project: the promote is compensation for performance, and the governance terms describe what happens if performance lags. Reading those terms against a bad quarter, not a good one, is the discipline that keeps partnerships intact.

Where C-PACE fits

In jurisdictions that authorize it, C-PACE financing funds qualifying energy, water, and resiliency components of a project through a special assessment on the property, an odd layer that is neither mortgage debt nor equity. Because the assessment runs with the land and collects like a tax, it changes the senior lender's collateral picture, which is why senior consent is a precondition rather than a courtesy. It can reduce the load on more expensive layers where it fits, but availability and program rules are highly jurisdiction-dependent. The C-PACE page covers where it applies and how consent gets handled.

Intercreditor agreements: the rules between the layers

When a stack has more than one capital provider, the relationships are governed by intercreditor, recognition, and subordination agreements. They read as legal plumbing and function as the project's constitution. The questions they settle: who gets paid, in what order, from which cash flows; what the subordinate layers may do when the project stumbles (cure a senior default, replace the sponsor, force or block a sale); standstill periods that keep a junior layer from acting immediately; and what happens to each layer in a foreclosure or bankruptcy.

Sponsors sometimes treat these agreements as a negotiation between their capital providers. That is a mistake. The intercreditor terms decide who the sponsor answers to in the project's worst month, and how much room exists to fix a problem before control shifts. They deserve sponsor-side attention before closing, when attention is cheap.

Who takes the first loss

The stack's order is easiest to see at the exit, when value flows down the layers in sequence: senior debt is repaid first, then any mezzanine debt, then preferred equity's capital and accrued return, and common equity keeps the remainder. A shortfall runs the same path in reverse: common equity absorbs the first dollar of loss, then preferred, then mezzanine, and senior debt is impaired only after everything beneath it is gone. Every layer's pricing is that sentence, converted to a return expectation.

Building a complete stack

The practical use of all this is simple: a complete sources-and-uses, with every layer identified and documented before construction starts. The capital-stack calculator lets you layer senior debt, subordinate capital, and equity against total cost and see the remaining gap. If a gap exists, structured capital covers how the subordinate layers get arranged in practice, and the senior, mezzanine, and preferred equity article goes deeper on the instruments themselves.

Two design principles are worth carrying into that work. First, cost of capital is not the only cost: a cheaper layer with harder remedies or a shorter fuse can be the more expensive choice for a project whose schedule might slip. Second, complexity itself has a price: every additional layer adds documents, consents, reporting, and one more party whose approval the project may someday need quickly. The strongest stack is usually the simplest one that actually closes the sources-and-uses.

A stack discovered to be incomplete at the second draw is a crisis; the same gap identified during review is a design problem with several answers.