Why Parking Geometry Sets a Building's Height

The same stall ratio governs a Professional Centre and a Suburban Office; underground parking is what tells them apart

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-011 · v0.1.0 · CC BY 4.0

Our thesis is that on the sites we build on, a building's height is not a design decision. It is arithmetic, and it runs backward. Start from the number of parking stalls the building is obliged to provide for every thousand square feet of space it rents out. Add the furthest anyone may reasonably be asked to walk from a stall to the front door. Measure the parcel. The number of floors the building can be falls out of those three facts, and it falls out before an architect has drawn anything.

This runs against the ordinary instinct, which is that a building is as tall as the zoning allows and the market will absorb. On a site with a rail station under it, that instinct is roughly right. On the sites we buy — the large-format retail nodes in smaller regional markets, where there is no rapid transit and, by our own site-selection criteria, is not going to be — it is wrong, because the binding constraint is not the height limit in the zoning bylaw. It is the land the cars need, and the distance from the cars to the door.

This paper is about that geometry and nothing else. It is deliberately not about how a building gets financed, which is a genuinely different question with its own answer elsewhere: a companion paper on our investment-thesis site sets out the financing sequence. The two are not competing answers to the same question, and it is worth saying so plainly because at a glance they can look like it. Geometry sets the ceiling — the tallest building this parcel can physically support given the cars. Financing determines whether, and how well, a project actually reaches that ceiling. Both are true, and neither alone answers "how tall is this building."

1. The thesis

A binding commitment to a minimum number of parking stalls per thousand square feet of rentable space, combined with a maximum walk from the furthest stall to the entrance, applied to a parcel of known size, determines the building's height ceiling before any other design decision is made. The geometry comes first.

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

The first is that the parking commitment is ours, not the municipality's. Almost every town we build in has a parking minimum in its zoning bylaw, and almost every one of them would let us build less parking than we intend to. Several would actively encourage it. Our governing agreement binds us to more than the town asks, and it binds us in a way we cannot relax on a particular site even if the town offers. That inverts the usual relationship between a developer and a parking requirement, in which the requirement is an imposition to be negotiated down.

The second is that we let the constraint decide the building rather than the other way round. The conventional sequence maximises the floor area the zoning permits and then finds somewhere to put the cars. Ours starts from the cars, and where the arithmetic produces a fraction of a floor, we drop the fraction. We do not lower the parking to capture the remainder. Over-building against a fixed constraint produces a larger asset without a larger return, and a larger asset on the same land is not an improvement — it is the same income spread over more building to construct, condition, and maintain.

2. The problem, in the reader's terms

Anybody who has tried to add a second car to a suburban house knows this arithmetic already.

The question feels like it is about preference — do we want two cars — and it is not. It is about the driveway. The lot is a certain width. The driveway is a certain width. Two cars need somewhere to sit where each can leave without moving the other, and where neither blocks the footpath. Depending on the lot, the answer is yes, or yes but only in tandem, or no. No amount of wanting a second car changes it, and the homeowner who buys the car first and works out the parking afterward ends up with a car on the street and a neighbour's complaint.

A commercial building is the same problem with the numbers larger and the stakes higher. The building generates a demand for stalls in proportion to how much space it rents out. The stalls take land. The land is finite and was expensive. And there is a second constraint the homeowner does not face: the stall has to be close enough to the door that the person using it does not resent the walk. A stall four hundred metres away technically exists and practically does not.

Work forward from the building and you discover the parking problem late, when the design is committed. Work backward from the parking and the building's size is the answer rather than the assumption.

3. What a parking ratio actually is

The vocabulary here is ordinary in the property industry and opaque outside it, so it is worth a plain account.

A parking ratio is the number of stalls a building provides for each thousand square feet of space it rents out. It is written as a ratio — four to one means four stalls per thousand square feet — and it is the standard way both municipalities and tenants express how much parking a building has. The Institute of Transportation Engineers maintains the reference work the industry uses to estimate how much parking different building types actually generate demand for [ite-parking-generation-5].

The denominator matters more than people expect. It is net leasable area, not the gross size of the building. A building's gross area includes the structure, the corridors, the lobby, the stairs, the mechanical rooms, and the shared washrooms. Net leasable area is the part a tenant actually rents and pays for. Because the ratio is applied to the smaller number, how you define net leasable area directly changes how many stalls you owe — and it is where a developer wanting to build more building will push, since reclassifying a thousand square feet out of net leasable removes several stalls' worth of land obligation. Our position is the opposite of that pressure: hold the stall count and argue instead that shared amenity space — the lobby, the tenant lounge — is not net leasable in the first place, because a tenant is not renting it.

A stall is not an abstraction either. Ours are full-size at nine feet by twenty, with separate allocation for accessible stalls, electric-vehicle charging, and motorcycles. A stall also needs a drive aisle to reach it and a footpath to leave it by, so the land a stall consumes is meaningfully more than the stall itself. This is why parking ratios convert into acres so quickly, and why the acre is the unit that ends up governing the building.

It is worth noting that the academic literature on parking runs in almost the opposite direction to our practice. The most influential modern work on the subject argues that municipal minimum parking requirements are set far too high, are rarely grounded in evidence, and impose large hidden costs on cities and on the people who do not drive [shoup-2011-high-cost-of-free-parking] [manville-shoup-2005-parking-people-cities] [willson-2013-parking-reform-made-easy]. We think that critique is largely correct about dense, transit-served cities, and we would apply it there. Our sites are the case it does not reach, for a reason given in section five, and we would rather state the tension plainly than write as though the literature agreed with us.

4. The arithmetic

Four inputs, and then the division.

The stall obligation. Our governing agreement binds the office buildings to a minimum of three stalls per thousand square feet of net leasable area. In practice the design works toward four, and four is the figure published per building class: four stalls per thousand for both office classes, with the lighter classes carrying their own lower ratios — three for the single-storey retail component, which pays more rent for it, and one for light-industrial space, which generates far less daily visiting.

Both numbers are real and they do different jobs, so it is worth being exact about which is which. Three is the floor: the binding commitment, the number below which the building may not go. Four is the target: what the national tenants these buildings are built for actually require, and what the design is drawn to. The outer ceiling — the tallest building a given parcel could support — is the one computed from the binding minimum of three, because that is the figure a reader can actually hold a building to. The published prototype heights are lower than that ceiling, because they are drawn to the four-stall target instead. The last part of this section explains why that direction is the right way round and not a mistake.

The walk. No stall sits more than seventy-five metres from its building entrance. This is the constraint that surprises people, and it is often the one that actually binds. Parking extends outward from the building in tiers, and each tier is further away; past a certain point the land is still there and the useful parking is not. It is also why the buildings are spread across a site and blended into the parking field rather than clustered together — clustering puts a larger share of the stalls in the distant tiers, which users notice immediately.

The land. Our development sites run between roughly four and thirteen acres. That is not a large site by the standards of a downtown block assembly, and it is the actual constraint rather than a planning nicety.

The conversion. From our test sites and our work with parking consultants, the rough planning guide is that one acre of land supports about one floor of office, and one level of below-ground parking supports about one more. That is explicitly a rough guide for plot planning, not a formula, and it is the number a reader should treat with the most caution of anything in this paper.

Put those together and the building comes out of the division. The acreage available to the office component — which is not the whole site, since the remainder carries the other buildings and the landscaping — converts to roughly that many floors of office. Where the division leaves a fraction, the fraction is dropped and the leftover land is filled with single-storey retail or light-industrial space rather than with a partial office floor the parking cannot support. Where more height is wanted, there are only three levers: reduce the parking ratio, add below-ground parking, or let the walk get longer. Nothing else moves the answer.

That geometry is what separates our two office building classes, and it is worth being exact about what does and does not distinguish them, because it is not the stall ratio. A Professional Centre and a Suburban Office are both bound to the same governing minimum described above. What differs is whether any of the required parking is allowed to go underground. A Professional Centre takes all of its parking on the surface, and working to the four-stall target with surface parking alone, holding the seventy-five-metre walk, it proves difficult to get past five floors — which is where that class's ceiling comes from. A Suburban Office adds up to one and a half levels of below-ground parking within the building's own footprint. That below-ground allowance buys roughly its own equivalent in floors above ground, on a somewhat larger parcel, which is how the same ratio supports six to nine floors instead of three to five. The two classes are, geometrically, one calculation with a single switch thrown: surface parking only, or surface plus a below-ground allowance.

Nine is not purely a parking calculation for the Suburban Office ceiling: it is also close to where the structural and building-services consultants place a natural breakpoint, for elevator provision and emergency travel distances, and the parking geometry independently says the same class cannot realistically exceed nine or ten floors. The two constraints agree, which is part of why the ceiling is set there.

One honest gap belongs here rather than being smoothed over. Exactly where a Professional Centre's range ends and a Suburban Office's begins — whether the practical cutoff sits at four floors or five — is not a fixed rule we can point to. It is a live design question, still being worked out against the parking geometry above and against the building codes of each jurisdiction we build in, and we would rather say so than publish a boundary that reads more settled than it is.

The governing agreement's own text contemplates the taller building going to fifteen floors. The geometry says that is not achievable under the same ratio and walk limit. We treat the fifteen as a figure the geometry has overtaken, and the working ceiling is nine.

Now the consequence of the two ratios, stated plainly because it cuts against the direction a reader might expect. Designing to four stalls per thousand rather than three means providing about a third more parking for the same rentable area — which means the same parcel supports fewer floors, not more. The published ceilings of five and nine floors are therefore where the four-stall working target lands, and they sit below the ceiling the three-stall binding minimum would permit on the same land. That is arithmetic on the two ratios rather than a measurement, and it is worth understanding correctly: the higher target is a self-imposed restriction on how much building we get from a site, not a licence for more of it.

5. Why the ratio is set above what towns require

A developer providing more parking than the municipality asks for is unusual enough to need an explanation, and ours is specific rather than philosophical.

The first half is about the markets. In the regional markets we build in, the private automobile is how nearly everyone arrives — tenants, their staff, and their customers. There is no rapid transit, and our site-selection criteria mean there is not going to be: these are towns with a large-format retail node and no rail. This is our own observation of how these places actually work, not a claim inherited from how the markets are defined. We would genuinely support lower parking ratios where rapid transit exists, and we would build to them there. We do not have that case.

The second half is about what happens when a developer gets this wrong, which we have watched. A development is built with relaxed parking — often with the town's active agreement, since the town would prefer landscaping to asphalt and so would we. The pre-leasing goes well; strong tenants, including national ones, sign. Then, around the fifth year, they do not renew. The reason given is not rent. It is that their people and their customers cannot park, or do not feel safe walking from where they parked, or cannot circulate through an undersized lot at the times of day when it matters. The development loses the prestige that its anchor tenants gave it, word of mouth in a small community does the rest, and the tenants who remain are in a position to demand a lower rent at renewal. The parking saving was realised at construction and paid for over the following decade.

That is why the ratio is set by what the national tenants say they require for their own businesses to work, rather than by the minimum a bylaw will accept. And it is why the commitment sits in the governing agreement rather than in a design guideline — so that it cannot be relaxed site by site under commercial pressure, including by us.

One thing this section is not arguing. Providing generous parking is not itself a virtue, and a correctly sized parking field laid out as bare asphalt is a worse outcome than a smaller one done well. The stall count travels with a design requirement: oversized drive lanes and footpaths through the parking field, and planting substantial enough to shade both people and cars. The count and the landscaping are one standard, not two line items, and there is a long-term ambition attached to it — if rapid transit ever arrives in one of these markets, a meaningful share of the stalls becomes reclaimable for landscaping and rooftop solar rather than remaining asphalt forever.

There is a further consequence that belongs to the same geometry. A building sized backward from its parking, with the fractional floor dropped rather than captured, is a smaller building than the zoning would allow. Less building for the same tenants means less to construct, less to heat and cool, and less to maintain. We treat that as the same discipline as the geometry of sustainability applied elsewhere in these buildings, rather than as a separate environmental initiative: the constraint that decides the height also decides how much building there is.

What this changes for the reader

The change is that a building's plausible height becomes something a reader can check for themselves, on any commercial property, with public information.

The check is: find the site's acreage, find the building's net leasable area, count the stalls, and divide. Does the building provide at least three stalls per thousand square feet of net leasable area — the binding floor? How far is the furthest stall from the door? And does the number of floors look consistent with the acreage, at roughly a floor per acre plus whatever is below ground? A building that fails those tests has either less parking than its tenants will want, or a longer walk than they will tolerate, or has resolved the tension some other way that is worth asking about.

That check works on buildings that have nothing to do with us, which is most of its value. It is also the reason this paper runs its arithmetic off the binding three rather than the working four: a reader applying our own advice to one of our own buildings should be checking against the number we are actually committed to, not the number we aim at.

The trade-offs are real. Parking at these ratios is a deliberate over-provision by the standards of modern planning practice, and the academic literature cited earlier would call it a cost imposed on the site and the town. It consumes land that could be building or landscape. It caps the buildings at heights that a taller-is-better view would call conservative. And it rests on a forecast — that the private automobile remains how people reach these places for the length of a building's life — which is a forecast and not a fact. If that forecast is wrong, we will have built parking fields that no longer earn their land, which is why the reclamation ambition above exists at all.

An open invitation

The question underneath this paper is not really a question about buildings.

To parking-policy and transportation researchers: the honest open question is whether a fixed, high ratio remains the right target across the fifty-year life of a building, when parking demand itself is moving. Remote and hybrid working patterns have changed weekday peak occupancy in office buildings in ways that were not modelled beforehand. Vehicle automation, if it arrives at scale, changes the relationship between a trip and a stored car more fundamentally than any policy could. We have designed for the demand we observe now, with an intention to reclaim stalls later, and we do not know whether that is the right hedge or an expensive one. We would rather have that argued with us than assume it.

To the reform literature specifically: we have cited work that argues our practice is wrong, because we think it is right about most places and we are not sure where the boundary of the exception is. The most useful contribution anyone could make is a characterisation of which markets genuinely require high ratios and which merely have them out of habit — a test, not a verdict. We claim our markets are on the exception side. We would like that claim tested.

To municipal planners in the towns we intend to build in: we are in the unusual position of building more parking than you are asking for, which may cut against your own objectives for your town centre. We would rather have that conversation early than be surprised by it at permit stage, and we would rather contribute to a transit case in these markets than simply outlast the lack of one.

And to anyone who has operated a development where parking was relaxed and tenants later left: the non-renewal pattern described in section five is an observation from our own experience, and it is doing a great deal of work in this argument. If the pattern is more complicated than we describe — if there are cases where relaxed parking worked in a comparable market — that is the finding most likely to change what we build.

To structural engineers and building-code specialists across the jurisdictions we build in: the honest gap named in section four — exactly where a Professional Centre's range gives way to a Suburban Office's — is unresolved on our side too. We would welcome help settling it against real local code, rather than treating the four-or-five-floor question as a detail to be fixed later.

Conclusion

Parking geometry sets a building's height ceiling. Three stalls per thousand square feet of net leasable area is the binding floor; four is the target the buildings are actually drawn to, because it is what the tenants they are built for require; no stall sits further than seventy-five metres from the door; and one acre of land supports about one floor of office, with a level of below-ground parking buying about one more. Divide, drop the fraction, and the building's height is the answer rather than the ambition. That is the whole of this paper's claim, and its narrowness is deliberate: it says nothing about how a building is financed, which is a separate question with its own answer. Geometry decides how tall the building can be. Something else decides whether it gets built well.

References

Institute of Transportation Engineers. Parking Generation Manual, 5th edition. 2019. https://www.ite.org/

Manville, M., and Shoup, D. 2005. Parking, people, and cities. Journal of Urban Planning and Development 131(4): 233–245.

Shoup, D. 2011. The High Cost of Free Parking, updated edition. American Planning Association. https://www.planning.org/

Willson, R. W. 2013. Parking Reform Made Easy. Island Press.

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 buildings and the site-planning discipline described in this paper; Woodfine Management Corp. employs the staff and retains the consultants who prepare this material and perform the site-planning work it describes. 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 parking commitments described are terms of a governing partnership agreement summarised here in plain language; that agreement, and not this paper, governs, and a prospective investor should read it rather than rely on this summary. Some statements above describe planned or intended future work; 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. No Woodfine Building has been constructed as at the date of this paper, and no site has been acquired, optioned, or permitted on the basis of the arithmetic described here.

Data and reproducibility

Three kinds of figure appear in this paper and they carry different weight.

The stall obligations, the seventy-five-metre walk limit, and the stall dimensions are terms of a governing agreement and of the prototype specification drawn against it. They are commitments, not measurements, and a reader relying on them should read the agreement rather than this summary. The two office parking figures are worth restating here because the difference matters and is easy to lose: three stalls per thousand square feet of net leasable area is the binding minimum the agreement imposes, and four is the ratio the buildings are designed to and published at. The arithmetic in this paper runs off the binding three, and says so where it does, because a reader checking a building should be checking against the commitment rather than the ambition.

The conversion between land and floors — one acre to about one floor of office, one below-ground level to about one more — comes from our own test-site plot planning with parking consultants. It is described in our own material as a very rough guide and should be read that way. It is the least precise figure in this paper and the one most likely to move as real sites are planned.

The published prototype ranges — three to five floors for the shorter office building, six to nine for the taller — are design conclusions reached with structural and building-services consultants, not observations of built buildings. No Woodfine Building exists, so there is no measured stall count, walk distance, floor count, or occupancy anywhere behind this paper. The non-renewal pattern at under-parked developments described in section five is our own observation from prior work in these markets rather than a study, and we have not attempted to quantify it.

The academic literature cited runs largely against our practice, which is why we have cited it. Those works are published and available through their publishers, and a reader who wants to weigh our position against the mainstream view should read them rather than our characterisation of them. No independent party has reviewed the geometry described here, verified the conversion guide, or confirmed the claims made for it.

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