Friday, August 7, 2015

In defense of by-right zoning: the dangers of arbitrary planning

An argument I've sometimes heard when strict zoning is criticized is that it's not so bad, because developers can simply ask for a zoning change or variances, which is certainly true. This is hard to get when you are in a residential area, but in most downtowns, which are primarily offices and commercial areas, the local opposition will be much less, so developers have more chances to get it.

Some urbanists and architects also like this approach because it gives a way for urban planners to review any given project and extract concessions from the developer in return for approving the zoning change they request. Since developers are often seen by urban planners and architects as greedy people who have no sense of aesthetics, subjecting their projects to review by "proper" architects and planners is viewed as a good thing, and the zoning change provides a lever to force developers to change their plans. They refuse, they don't get their zoning change.

So, is this actually a good approach? To have zoning limits so strict that almost all projects have to demand variances and changes and reviewed by urban planners and architects?

Uncertainty and land prices

I've already made the argument that land prices are extremely useful in shaping cities. Urban development is an economic enterprise, and as such is extremely reactive to price signals, the most important one here is the price of land. Any person wishing to develop an area must first buy the land on which the development will occur. The more expensive the land, the higher the density of what developers will build. If land is too expensive, then low-density projects are nonviable economically and so are excluded. If land is cheap, then low-density projects are viable and may occur.

But what determines the value of land? The land owner wants to sell for the highest possible price, the land buyer wants to buy for the lowest possible price, but what yardstick helps to determine the proper value of land?

Well, my theory, which I believe is pretty sound, is that the potential profit value of development is what decides how much the land will sell for. In essence, a development generates revenues that are the difference between the market value per square foot and the construction cost, times the number of square feet that can be built.

Land owners know this and will set their expectations for the price of the land they sell accordingly. It's important to understand that, generally speaking, only the building itself generates revenue. Parking (unless it is tolled) doesn't generate revenues, neither do trees or grass (green buffer space), etc... It's the building that generates revenues, that is either rented or sold to people who want to use it. The rarer the land, the more land owners can negotiate a higher share of these expected revenues, the more abundant the land, the lower the share they can negotiate.

So what does that mean for urban developments? The market value is largely an input from the real estate market, it's relatively easy to get from looking at other developments. Construction costs can also often be approximated based on local construction costs. So the big question here is how many square feet of revenue-generating floor area can you build? That will determine what the land is worth.

So we come back here to the Floor-Area Ratio. I once said that it was not a proper vector for urban regulations because there is no direct negative externality of FAR that justifies its direct regulatory control. However, I may have jumped to conclusions too soon. The point is that controlling FAR may actually help the market determine with a greater amount of certainty the total potential revenue of development on land.

For example, if the value of office space is, say 300$ per square foot (not the rent, the actual value), and the construction cost is 200$ per square foot, that's a total revenue of 100$ per square foot. If allowed FAR is 200%, then that means that development revenue is limited to 2 square feet of revenue-generating floor area per square foot of land. So there is a ((300$/sq ft - 200$/sq ft) X 2 sq ft floor area/sq ft of land) 200$ revenue associated to each square foot of land, which is a hard cap on land value. If land owners get on average half the total revenue, then the land value would be about 100$ per square foot.

So when FAR is limited, either directly or through a combination of height limit and lot coverage limit (setbacks, margins, etc...)...
Front setback and minimum margins reducing the area the building can cover, here, the maximum coverage is 42% of the lot, if there is a height limit, then FAR is indirectly limited (for example, if 5 stories are allowed, maximum FAR is 210%)
...or through minimum parking requirements, then this helps clear up the uncertainty as to the amount of revenue-producing floor area that can be built on any given lot. This helps both the seller and the buyer of the land establish a potential land value, and thus helps them come to an agreement on price much faster.

But what happens when you keep zoning very strict with the expectation that you will consider requests for variances and zoning changes on a case-by-case basis?

Well, in that case, an arbitrary component is introduced, creating uncertainty. The zoning may currently limit FAR to 100%, but maybe in one case the city allowed one project to have 200% FAR based on a zoning change request. What message does that send to every land owner in the area? Even if your land is zoned for a 100% FAR, you should price your land as if it were zoned twice as high. Of course, developers would be wary to pay such high prices, because there's no guarantee they'll get a zoning change.

The result is making development less likely, as land owners price their lots well over what they are currently zoned for while developers, afraid of risk, may balk at the idea of paying land a certain price that is only justified with a zoning change.

Other impacts: corruption, blackmail and administrative bottleneck

Everywhere in the world, arbitrary power is always associated with another phenomenon: corruption. Without strict guidelines, public administrators have the freedom to make decisions based only on their own feelings without having to explain them to anyone. As a result, given the massive amount of money involved, developers have an incentive to do all they can to get administrators to come down on their side. I'm sure developers would prefer doing their thing without having to grease a few palms, but in the context of arbitrary urban planning, the temptation may be too great.

This may also go the other way, with either administrators or even sometimes NIMBY groups using the situation as leverage to extort favors from developers. For example, forcing developers to pay directly for the repair or replacement of local infrastructure, or forcing the developer to provide public goods like parks in exchange for building permits. This can result in much higher prices for new constructions as all the burden of providing for local infrastructure needs can be unloaded on them alone. This doesn't mean that having developers provide certain social goods is a totally bad idea, rules that establish certain fees to contribute to a local infrastructure fund to pay for the marginal cost of new developments may be a sensible approach, but this has to be set in the rules long before any project is proposed.

Finally, there is one final drawback. Setting up a case-by-case review process of every project is extremely expensive in terms of labor and time. Since cities have a limited amount of planners and architects, if there is a construction boom, then the review process might get congested as the urban planning department is overwhelmed with projects to review and comment. This can lead to delays and a limit on how many projects get the green light in cities, limiting new constructions just when demand is highest for them.

All these factors coalesce into one major final negative effect: the elimination of small-scale developers who do not have the funds and expertise to deal with all these issues and of small-scale developments that do not yield the revenues to justify facing down all that process.

In conclusion

I think what cities need is not more control by urban planners and architects, but less... arbitrary inputs in the development process need to be reduced to a minimum. Where rules exist, they must be applied upstream, based on clear, established guidelines to reduce, or even eliminate, uncertainty. When upzoning is required to deal with higher demand (and I believe rules must have a pro-redevelopment bias), this upzoning should be gradual and affect a large area, not just spot-by-spot rezoning based on specific requests. Even exceptional projects should ideally be realized in the context of specific rules, for instance through the purchase of "air rights" (buying unused FAR limits or height limits from lesser density developments) or incentive-based zoning (giving FAR or height bonuses for certain initiatives).

As in most things, in terms of urban development, the rule of law must be predominant, not the rule of men, in order to provide certainty and to level the playing field.


  1. I've heard that this is one reason housing in Vancouver is so expensive, because even if they don't require variances for projects, they still go through a lengthy review process which acts as a choke point at the building department. Even here in Cincinnati, which is not exactly a booming locale, is seeing major delays in the zoning and building department. It's especially severe at the historic conservation board because of the redevelopment happening, which just overwhelms the city offices that are still at staffing levels more appropriate for 2008.

    I've also heard of instances where overly-permissive zoning caused development to stall because it led to speculation and inflated prices. Only when the zoning was changed to be more restrictive did the lots finally start selling and development began. I would say that's a problem with property tax structures that reward speculation and penalize development, and zoning is just a tool to work within that flawed system, so both must be considered together. It may make price discovery harder than it is currently, but with the system set up to encourage development and not land banking, there will be more comparable projects and property sales to analyze.

    1. There is an issue with speculation indeed. The problem is that when density limits are nearly inexistent, the potential revenue of development is hard to pin down. You could have a 4-story building, a 10-story building or a 20-story building... these all have very variable revenue potential, and so it's hard to evaluate the appropriate land value. Land owners may be willing to sit on their land until they get the developer who wants to build a skyscraper, because selling at a much lower price to the developer who wants to build a 6-story building would have a very high opportunity cost.

      Of course, one solution is to make it extremely expensive to sit on land, notably through land value taxation. That way, speculators have an incentive to sell their lots quickly, because they are paying the taxes for it every year.

    2. Right, the land tax seems to be the way around that, but then that tax has to be calculated at some point too, which should be based on the underlying value of the property. How exactly that's achieved I'm not entirely sure, because the value of land isn't based solely on proximity to downtown but also on the desirability of the particular neighborhood, transit access, views, and stuff like that.

      Of course, wouldn't the simplest way to determine value be the way they already do it, but looking at comps within the surrounding few blocks? If the neighborhood is all large-lot single-family homes, then it's unlikely that a skyscraper will be plunked down there unless the city's real estate market is already severely distorted due to restrictive zoning.

      Historically, there weren't really any cataclysmic changes in density (or built form anyway) because supply and demand were able to respond to each other throughout the entire city, so growth and densification happened gradually but deliberately and even fairly predictably. Thus, one couldn't expect to get much more value out of a particular parcel than what was already there. It wasn't so much a choice between a 4 story building and a 20 story high rise, but a choice between a 3 story building and a 5 story building.

    3. Historically, there WERE cataclysmic changes in density and built form from large-scale urbanization, industrialization, and migration. Between 1910 and 1930, the population of Los Angeles quadrupled, growing by almost a million people! Zoning makes sense in that kind of context when land went from farms to single family homes to skyscrapers almost overnight.