Financing the Building From the Lease, Not the Project

Nothing in this paper constitutes an offer to sell, or a solicitation of an offer to buy, any interest in a Woodfine direct-hold solution; any offering is made only by the applicable Private Placement Memorandum. Statements marked "planned," "intended," "targeted," "may," or "expected" are forward-looking and subject to change. Full disclosures appear at the end of this paper.

Working Paper WCP-013 · v0.1.0 · CC BY 4.0

Our thesis is that in development finance the order of the money matters more than the price of it. The conventional sequence borrows first and earns later: a developer arranges a construction loan sized against what the finished project is projected to cost, builds, and then hopes the completed building earns enough to service what was borrowed to build it. The sequence argued for here runs the other way. The first buildings are paid for with equity and carry no construction debt at all. Only once they are finished, leased and producing rent does the investment vehicle itself borrow — issuing its own mortgage-secured bonds against the income those buildings are already earning — and that borrowing pays for the next phase.

The reason for inverting the order is not a general preference for low debt. It is a specific, local obstacle. In the smaller regional markets these buildings are planned for, a commercial lender will generally finance the ground-floor retail against a signed lease with a national tenant, and two or three floors of offices above it — and, where none of the office space can be pre-let at all, perhaps only one or two — and then stop, because the office space higher up is genuinely hard to pre-lease in those markets, where a professional firm typically wants a quarter of a floor rather than three floors, and where large institutional investors are not present to stand behind the project. The practical consequence, repeated across those markets, is a two- or three-storey building. The same amount of pre-leasing in a dense, transit-served city supports a ten- to fifteen-storey one. Taking the lender out of the first stage takes that ceiling out with it.

One boundary before we start, because at a glance two different questions can look like the same one. This paper is about whether and how well a building gets financed. A separate paper of ours, on the site that covers how these buildings are actually designed, is about what physically limits a building's height on a given piece of land — which turns out to be the geometry of its parking rather than anything financial. Those are not competing answers. Geometry sets the ceiling a parcel can physically support. Financing decides whether, and how well, a project reaches it.

1. The thesis

Sizing debt against income a building has already earned, rather than against a cost estimate for a building that does not yet exist, is a fundamentally different — and, we argue, safer — sequence than conventional construction financing. It is also, over a full development cycle, a more productive one: the company's own financial model calculates that the same starting capital ends up having built over forty per cent more square feet this way than through a conventional joint venture. That is a calculation, not a result, and Section 6 is about exactly how much weight it can bear.

Two things make this a real claim rather than a slogan.

The first is that the borrowing is done by the vehicle that owns the buildings, against those buildings, rather than by a development company against its own balance sheet and reputation. The debt is secured directly on the physical buildings and on nothing else. If a debenture were ever to default, the transfer of the building behind it extinguishes that debt entirely; the rest of the portfolio carries on; and investors cannot be called on for more money to cover it, because the liability was never theirs personally in the first place.

The second is that the income has to exist before the debt does. That is the whole content of the covenant described in Section 5, and it is a self-imposed constraint rather than a lender's. A conventional lender is perfectly willing to lend against projections, because the lender holds security and is paid for the risk. The constraint here does not come from a bank. It is written into the vehicle's own governing agreement, which is a different thing entirely: a bank's requirement can be renegotiated with the bank, while this one requires the investors to vote.

2. The problem, in the reader's terms

Nothing in this section requires any knowledge of commercial property. It requires knowing how a bank thinks about lending on a building, which is the same way a bank thinks about lending on anything.

A lender's question is always some version of: what is paying this back, and what do I get if it doesn't? For a finished, rented building the answers are easy. The rent pays it back; the building is the security; the rent can be counted, because tenants have signed leases and money is arriving. This is why a mortgage on a fully leased office building is one of the more ordinary transactions in finance.

For a building that does not exist yet, both answers get harder. There is no rent, because there are no tenants, because there is no building. So the lender asks for something that stands in for the rent: leases signed in advance of construction, on space that does not yet exist. That is pre-leasing, and it is the hinge of this whole argument. A developer who arrives with signed pre-leases has converted a projection into a contract, and a lender will lend against a contract.

Now put that mechanism into two different places.

In a large city with a rail station nearby, the ground-floor retail space can be pre-leased to a substantial national tenant, and several floors of offices above it to two or three more substantial tenants, before a shovel goes in the ground. With that much contracted income in hand, and often a large institutional investor standing behind the project as well, a lender will fund a ten- to fifteen-storey building.

In a town of forty thousand people with a large-format retail node at its edge and no rail at all, the ground-floor retail can still be pre-leased — perhaps not to a national grocery chain but to a local optometrist. The local law firm, engineering practice and accountancy will sign too. They want a quarter of a floor each, and quite often they would rather see the building finished before committing at all. So the developer arrives at the lender with the ground floor let and very little above it, and the lender funds the ground floor and two or three storeys, which is a perfectly rational decision on the lender's part and produces a three-storey building.

It is worth being careful about the strength of that claim. It is not that no bank will finance a taller building in a smaller market. It is that conventional financing becomes materially harder, and the market's answer is systematically shorter buildings than the land or the demand would support. That is the difficulty this sequence is built around. It removes the lender from the stage at which the lender's question cannot be answered, and brings the lender back in at the stage where it can be answered easily — after the building is standing and the rent is arriving.

The everyday analogy is close to exact. A buyer who wants a rental property and has no cash must persuade a lender to finance a purchase on the strength of rent nobody has yet paid. A buyer who pays cash, lets the property, waits until the tenancy is established, and then mortgages it, is asking the lender a much easier question, and will usually get a better answer. The second buyer has not avoided debt. They have changed when the debt arrives, and therefore what it is priced against.

3. What "narrow banking" means, and what it does not mean here

The name attached to this sequence is borrowed, and the borrowing should be explained rather than assumed, because the original meaning is both real and narrower than the use made of it here.

Narrow banking is a long-running idea in banking reform. An ordinary bank takes deposits that can be withdrawn on demand and lends them out for years — borrowing short and lending long, which is where most of the fragility in banking comes from, because a bank that must repay today cannot get its money back from a twenty-year mortgage today. A narrow bank is an institution that deliberately does not do this: its liabilities are demandable, and its assets are correspondingly safe and short, so the two sides match. Proposals along these lines have appeared after nearly every major financial crisis since the 1930s, and a substantial review of the theoretical and empirical literature finds the evidence largely supportive of narrow-banking reforms, chiefly because such a system contains the moral hazard that deposit insurance and lender-of-last-resort support otherwise create [pennacchi-2012-narrow-banking].

What is borrowed here is the matching discipline, not the banking licence. These vehicles are not banks, take no deposits, and make no loans. What they take from the idea is the refusal to fund a long asset with a short, fragile claim. Applied to property development, the long asset is a building with a fifty-year life, and the short fragile claim is a construction loan with a four- or five-year term that must be refinanced at whatever rates and lending appetite happen to exist on its maturity date. The sequence described here funds the long asset with permanent equity — capital with no maturity date at all — and only issues term debt once there is contracted income to match it against.

There is a second borrowed idea worth naming for the same reason. A conventional development joint venture of the kind the group has used historically carried debt at roughly two and a half times equity, with the cost of a construction loan running for four to five years before the building earned anything. That is not a criticism of the joint-venture form, which built a great many buildings. It is the specific cost this sequence is designed to remove: not the interest rate, but the years of paying interest on a building that is not yet earning.

4. The equity phase: building with no debt at all

In the first phase, equity capital raised through the vehicle funds construction outright. No pre-leasing is required before construction begins. No construction loan is drawn. The completed buildings are, in the model's own term, unlevered square feet — finished, rentable space carrying no debt, and therefore imposing no interest payment on the vehicle during the months when the building is being leased up and is not yet full.

That is only possible because of the shape of the equity. In a conventional private real-estate fund the capital has a deadline: the fund has a life, commonly seven to twelve years, at the end of which assets must be sold and investors repaid. A manager with a deadline cannot sensibly put money into an empty building for two years, because the clock is running. The units in these vehicles carry no redemption date, no maturity and no compulsory exit event — equity is held indefinitely, and an investor needing liquidity is expected to find a buyer rather than to be redeemed by the issuer. Capital with no deadline can be put into a building that will not earn for two years, because nothing forces the question to be answered sooner.

Two plain statements belong here rather than in a footnote. The first: there is a real cost to this, and it falls on the early investors. Building without debt means the same equity buys fewer square feet at the start than leveraged equity would — the model is explicit that this sequence initially develops fewer unlevered square feet than a conventional joint venture on the same money, and only overtakes it later in the cycle. An investor whose horizon is the first phase rather than the full cycle is, on the model's own arithmetic, worse off. The second: the perpetual structure that makes this possible is also the structure that makes these units illiquid, and the units of the one established vehicle cannot be transferred at all today, because that partnership is subject to an active cease-trade order issued by the British Columbia Securities Commission, arising from delayed continuous-disclosure filings rather than any finding of fraud or misconduct. A separate paper of ours deals with what transferability does and does not give a holder; it is named here so that a reader does not leave this section believing perpetual equity is costless.

5. The debt phase: how the borrowing is sized once it exists

Once income from completed, leased buildings has stabilised, the vehicle is planned to issue First Secured Mortgage Debentures — its own bonds, secured in first position directly against the physical buildings, to fund the next construction phase rather than to refinance what is already built.

Several features of that borrowing are worth a reader's attention, and each carries its qualification in the same breath.

The coverage test. New debentures may not be issued if issuing them would push the vehicle's interest coverage ratio below 1.20 times. Interest coverage is a simple idea: net operating income divided by the annual interest the vehicle owes. At 1.20 times, the buildings are earning one dollar and twenty cents of operating income for every dollar of interest due. The test is applied across the vehicle as a whole rather than property by property. Three honest qualifications sit on it. It is a floor on new borrowing, not a continuously monitored trigger — it is tested at the moment of issuance, against an income figure that is by then historical, and nothing about it re-tests the vehicle if income later falls. It excludes principal repayment, which makes it a more lenient measure than the debt service coverage ratio a commercial lender would normally impose. And it can be waived: the vehicle may exceed the floor with a special resolution of unit holders, which requires seventy-five per cent support. It is a high bar, deliberately, and it is not an absolute unconditional ceiling, and a reader should hold both of those facts at once.

The ceiling on quantum. The maximum debt under the model is one hundred per cent of development costs for new construction. Debentures fund the building of additional buildings; they do not fund the refinancing of buildings already owned, which is the mechanism by which leverage in conventional portfolios tends to ratchet upward without anything new being built. During the active construction phase, the model expects portfolio debt to reach no more than about sixty-three per cent of asset value.

Interest cannot be hidden. Each vehicle's governing agreement prohibits capitalised interest — the practice of adding interest owed during construction to the loan's principal rather than paying it as it accrues. Capitalising interest is common and legal and it has one consistent effect: debt grows during construction without the cost showing up in current income. Prohibiting it means the cost of the borrowing appears in the period it is incurred.

Debt comes down on its own. In the stabilised phase, ten per cent of distributable income is intended to be retained and applied to redeeming debentures — with the intent that leverage falls toward roughly half of asset value over the hold period rather than merely drifting downward. The mechanism is the vehicle's own cash flow, not a favourable refinancing market or an asset sale. The distribution rule this comes from is worth stating precisely, because it is not discretionary generosity: where a distribution is declared, at least ninety per cent of the fiscal year's distributable income is paid to limited partners within ninety days of year end, and until cumulative distributions equal the whole of what limited partners paid for their units, the retained ten per cent is held for partnership purposes; after that point it goes to redeeming debentures until they are fully redeemed.

Existing investors get first refusal on the debt. When the vehicle issues debentures, existing equity holders have a right of first refusal — the opportunity to buy into that debt before it is offered to outside purchasers. An investor who wants exposure to both halves of the same buildings' capital structure can have it.

The debentures are designed to qualify for listing, not to trade privately. The intended design meets the applicable exchange's own listing standards, with the intent to list the debentures on it, while the equity units themselves remain unlisted. Meeting an exchange's standards and a security actually trading on it are two different steps, and it is the first of those two the design targets today — no debenture has yet been issued, so no listing has yet occurred either. In the Mexican vehicle's planned design the same distinction is explicit: the debt instrument is intended to trade on that country's exchanges once issued, while the equity certificates stay off any trading board. A reader should read "exchange-qualified" here as the accurate present-tense description — the design is built to meet listing standards — and "exchange-traded" as the intended outcome once an actual issuance occurs, not as a current fact.

6. What the model claims, and what it does not prove

The capital-efficiency claim deserves its own section because it is the single number most likely to be quoted out of this paper, and it needs its qualifications attached to it rather than trailing behind.

The claim is this: by the end of a full development cycle, and starting from exactly the same initial capital, this sequence develops over forty per cent more square feet than a conventional legacy joint venture. The mechanism is compounding rather than cleverness. Because the developer's profit is taken as equity rather than drawn out as cash, and because a portion of income is retained rather than fully distributed, capital accumulates inside the vehicle and is available to build again — each cycle starting from a slightly larger base, with debt raised against income that already exists rather than against the next projection.

Now the qualifications, and they are substantial.

This is a model output. It is a forward-looking calculation from a set of stated assumptions about construction cost, lease-up timing, rent levels, interest rates and holding periods. It is not a measurement of anything that has happened. No Woodfine Building has been constructed. No First Secured Mortgage Debenture has been issued. No completed portfolio exists against which either the equity phase or the leverage ceiling has been tested. The comparison against a conventional joint venture is a comparison against a modelled alternative, not against a real competing project that was run in parallel.

The model also loses in the early years before it wins in the later ones, and the crossover point is itself a model assumption. Anyone assessing this should ask when the crossover is projected to occur and what happens to the answer if lease-up takes longer than assumed — because an unlevered building that stays empty is a pure cost with no offsetting interest saving to show for it.

Our position is that a model claim of this kind should be published with its parameters attached, so that a reader can disagree with the assumptions rather than only with the conclusion. The forecast methodology behind these vehicles is documented publicly in outline: distributions are reconciled to distributable income rather than to accounting net income, specifically excluding the non-cash revaluation gains that cannot actually be distributed, and the development-phase debenture is modelled at amortised cost so that the effective borrowing rate the forecast shows is higher than the coupon on the instrument's face. That is the right direction. It is not yet the same as publishing the parameter set behind the forty per cent figure, and until that is published the figure should be read as a claim we are making rather than one a reader can check.

What this changes for the reader

The first change is that "capital-efficient" stops being an adjective and becomes an arithmetic claim with a mechanism attached. An investor can ask three specific questions of this structure, or of any other that claims something similar. Is there construction debt on the first buildings? What is the borrowing sized against — a cost projection, or income already contracted? And what happens to the covenant if income falls after the debt is raised? Those three answers tell a reader more than any amount of language about prudence.

The second is a way of reading the pre-leasing question that generalises well beyond this company. When a reader is told that a market cannot support a taller or better building, it is worth asking whether that is a fact about the market's demand or a fact about how the market is financed. Those get conflated constantly, and they have different remedies. Demand cannot be engineered. A financing sequence can be changed.

The third is a plain statement of where the risk went. It did not disappear. In the conventional sequence, the risk of a building not leasing sits substantially with a lender, who is paid to hold it and who has security and remedies. In this sequence it sits with the equity — which is to say with the investors — during the whole of the construction and lease-up period. That is the trade. A reader who prefers the conventional arrangement is not making a mistake; they are choosing to have a professional lender in the deal, priced accordingly.

The trade-offs belong together in one place. Equity-funded construction means the earliest investors fund the least efficient phase of the cycle. The coverage covenant is waivable by a supermajority and is tested only when new debt is issued, so it constrains borrowing rather than guaranteeing solvency. The capital-efficiency claim is modelled and unverified. And the whole sequence depends on being able to raise substantial equity before any building exists, which is a real commercial constraint and not a technical one: at the date of this paper, one vehicle of four is established, and its units are subject to an active cease-trade order.

An open invitation

The central question in this paper is an empirical one that we cannot settle from inside the company.

To real-estate finance academics: the honest open question is whether income-sequenced, equity-first development genuinely outperforms conventional leveraged development across a full market cycle including a downturn. Our model says it does over a cycle it assumes. The interesting test is the cycle it does not assume: a period where rents fall while an unlevered building is in lease-up, or where the coverage covenant blocks the next phase precisely when land is cheapest. Leverage is not only a cost; in a rising market it is the mechanism by which a developer captures more of the rise. A structure that forgoes it in the first phase forgoes that too. We would rather see that tested against this model's published parameters than assume it either way.

To the banking-reform literature: we have borrowed the narrow-banking framing because the matching discipline genuinely is what we are doing, and we are aware that the original idea is about deposit-taking institutions and systemic fragility, not about property development. If the analogy breaks somewhere important — if there is a reason the matching argument works for a bank's balance sheet and does not carry across to a development vehicle's — that is the kind of correction that would change how we describe this rather than merely how we defend it.

To credit analysts and rating professionals: the interest coverage covenant in this structure is deliberately more lenient than the debt service coverage ratio a commercial lender would impose, it excludes principal repayment, it is tested at issuance rather than continuously, and it is waivable by vote. We think that combination is defensible for a vehicle with no external construction debt and no refinancing wall. We would like to be told if it is not, and specifically whether a covenant tested only at issuance is worth anything at all in a downturn.

To anyone with experience of a development programme that ran this sequence: the forty per cent figure is a calculation, and the most useful thing anyone could hand us is a real case — a programme that built unlevered and then borrowed against completed income, over a full cycle, with what actually happened to the compounding. We have not found one at this scale. If it exists, it is worth more to this argument than any refinement of our own model.

Conclusion

The sequence is the claim: equity first, with no construction debt on the earliest buildings, then debt issued by the vehicle itself and sized against income those buildings are already earning. It answers a specific, local problem — that in smaller regional markets a lender's perfectly rational question about pre-leased income produces buildings two or three storeys tall — by removing the lender from the stage where that question cannot be answered. The capital-efficiency result that follows, over forty per cent more built space from the same initial capital by the end of a development cycle, is a model output and should be read as one: nothing has been built, no debenture has been issued, and the model wins late after losing early. What a reader can hold us to is not the figure but the sequence, and the covenant, and the published parameters when they arrive.

References

Pennacchi, G. 2012. Narrow banking. Annual Review of Financial Economics 4. https://doi.org/10.1146/annurev-financial-110311-101758

Canadian Securities Administrators. National Instrument 51-102 — Continuous Disclosure Obligations. https://www.bcsc.bc.ca/securities-law/law-and-policy/instruments-and-policies/5-ongoing-requirements-for-issuers-insiders/current/51-102

Contributors

Prepared by Woodfine Management Corp.

How this paper was produced

This paper is grounded in the operating and development work the preparing staff do themselves, and in what they have learned from the people they work with routinely: the graphic designers, web developers, and software developers and engineers who build the platform alongside them, and the architects and structural, building-services, and civil engineers they develop buildings with. No outside professional reviewed or approved this paper, and nothing in it is professional advice. It was drafted and edited with AI assistance under editorial direction, and the analysis and conclusions are Woodfine's own.

Disclosures

Woodfine Capital Projects Inc. ("Woodfine") is the author of record and is the developer and promoter of the financing structure described in this paper; Woodfine Management Corp. employs the staff and retains the consultants who prepared this material. Woodfine has a direct commercial interest in the approach argued for here. This work was funded internally; no external research funding was received. Nothing in this paper constitutes an offer to sell, or a solicitation of an offer to buy, any interest in a Woodfine direct-hold solution; any offering is made only by the applicable Private Placement Memorandum, which prospective investors should review with their own professional advisors. The covenants, ceilings and retention rules described are provisions of a governing limited partnership agreement summarised here in plain language; that agreement, and not this paper, governs. Only one direct-hold vehicle is established; the United States, Spain and Mexico vehicles are planned and not yet formed. The established Canada vehicle's units are subject to an active cease-trade order issued by the British Columbia Securities Commission, arising from delayed continuous-disclosure filings rather than any finding of fraud or misconduct; it must bring its outstanding filings current and obtain the Commission's revocation before any future offering of units can close. No First Secured Mortgage Debenture has been issued and no Woodfine Building has been constructed as at the date of this paper. Some statements above describe planned or intended future activity; language such as "planned," "intended," "targeted," "may," and "expected" marks this forward-looking content, which is subject to change and does not constitute a commitment regarding future performance. This paper states no projected rate of return, distribution yield, or investment performance figure of any kind, and none should be inferred from it.

Data and reproducibility

Three kinds of statement appear in this paper and they do not carry equal weight.

The covenants and prohibitions — the 1.20-times interest coverage floor and its special-resolution waiver, the one-hundred-per-cent-of-development-cost ceiling on new debenture quantum, the prohibition on capitalised interest, the ten per cent of distributable income applied to debenture redemption after capital recovery, the at-least-ninety-per-cent distribution rule and its ninety-day timing, the ring-fencing of debt to the buildings, the prohibition on capital calls, and the right of first refusal on debentures for existing equity holders — are terms of the established Canada vehicle's own governing agreement. They are commitments rather than measurements, and they are specific to that vehicle; the three planned vehicles will be governed by separate instruments. A reader relying on any of them should read the agreement.

The construction-lending description — that a lender in these markets will typically finance ground-floor retail pre-leased to a national tenant plus two or three floors above it, against a ten- to fifteen-storey building on comparable pre-leasing in a dense transit-served market — is our own observation from prior development work in these markets rather than a study. We have not attempted to quantify it, and it is the kind of claim that varies by lender, by country and by year.

The capital-efficiency figure — over forty per cent more square feet by the end of a full development cycle on the same initial capital — is an output of the company's own financial model, calculated from stated assumptions about cost, timing, rents and rates. It is not an observed result, and it does not describe a completed project, because none exists. The same model shows this sequence developing fewer square feet than the conventional comparison in the early phase, and that qualification travels with the figure. The peak leverage of approximately sixty-three per cent during construction and the intended stabilisation near half of asset value are likewise modelled expectations, not measurements. The model's parameter set has not been published; until it is, a reader cannot reconstruct the forty per cent figure, and should treat it accordingly.

The debentures are designed to meet the applicable exchange's listing standards, with the intent to list them once issued; the equity units are not intended to be listed. No debenture has been issued yet, so no listing has occurred yet either — "exchange-qualified" describes the design today, "exchange-traded" describes the intended result once an actual issuance occurs. No independent party has reviewed the financing model described here, reproduced its figures, or verified this paper's description of the governing agreement.

Woodfine Capital Projects™ is a trademark of Woodfine Capital Projects Inc.