A Better Way to Zone

Ten Principles to Create More Livable Cities

Donald L. Elliott

• Publisher: Island Press
• Year: 2012
• ISBN:
  9781610910552
• Topic: Architecture
• Subtopic: Architecture générale

Nearly all large American cities rely on zoning to regulate land use. According to Donald L. Elliott, however, zoning often discourages the very development that bigger cities need and want. In fact, Elliott thinks that zoning has become so complex that it is often dysfunctional and in desperate need of an overhaul. A Better Way to Zone explains precisely what has gone wrong and how it can be fixed. A Better Way to Zone explores the constitutional and legal framework of zoning, its evolution over the course of the twentieth century, the reasons behind major reform efforts of the past, and the adverse impacts of most current city zoning systems. To unravel what has gone wrong, Elliott identifies several assumptions behind early zoning that no longer hold true, four new land use drivers that have emerged since zoning began, and basic elements of good urban governance that are violated by prevailing forms of zoning. With insight and clarity, Elliott then identifies ten sound principles for change that would avoid these mistakes, produce more livable cities, and make zoning simpler to understand and use. He also proposes five practical steps to get started on the road to zoning reform.While recent discussion of zoning has focused on how cities should look, A Better Way to Zone does not follow that trend. Although New Urbanist tools, form-based zoning, and the SmartCode are making headlines both within and outside the planning profession, Elliott believes that each has limitations as a general approach to big city zoning. While all three trends include innovations that the profession badly needs, they are sometimes misapplied to situations where they do not work well. In contrast, A Better Way to Zone provides a vision of the future of zoning that is not tied to a particular picture of how cities should look, but is instead based on how cities should operate.

CHAPTER 1

A Brief History of Zoning

TO UNDERSTAND HOW ZONING doesn’t work, we need to start with how it is supposed to work—or how it was supposed to work when it was shiny and new. We also need to understand how it has already been modified to deal with new problems and pressures. We need not explore the history in detail—several good books already do that¹—but we do need to understand today’s starting point so we can avoid repeating the mistakes of the past. By reviewing how we have already tried to “fix” zoning, we can identify what parts of the fix worked and what parts did not (and why).

“Euclidean” Zoning

Traditional zoning is “Euclidean” zoning, named not after the Greek mathematician Euclid nor because a zoning map looks sort of geometrical, but after the town of Euclid, Ohio, which won the first lawsuit over the legality of zoning. But before Euclid there was New York City, which adopted the first major zoning ordinance, in 1916.² If you can imagine the chaos of New York during the early twentieth century, when it was bursting at the seams with more poor immigrants arriving daily, you can understand why the city leaders felt they needed to do something. They had already adopted a partial building code (the Old and New Tenement Laws) requiring that apartment houses be built so that a certain amount of light and air could reach the inner rooms. But now those tenements were sprouting up everywhere, even near factories—and, worse yet, near mansions.

So New York City adopted a “zoning plan” restricting the uses to which land could be put. Basically, they divided the city into “zones” for residential, commercial, and industrial uses. While they were at it, they also established an imaginary box within which buildings had to be built on each site. Some parts of the box—for example, requirements for access and light—were simply concepts carried over from the Tenement Laws of 1867, 1879, and 1901. The sides of the box were defined by distances that the building had to be “set back” from the front, rear, and side lot lines, and the top of the box was a maximum height limit on the overall structure. That was it. There were no parking controls (not many cars back then), no sign controls, no landscaping requirements, and no design controls. This was the backbone of zoning: a division of the city into use zones and boxes into which buildings had to fit in order to avoid crowding their neighbors; to avoid creating public health, safety, or fire hazards; and to prevent land use combinations likely to cause nuisances in the future.

The structure of the zoning document was straightforward. It included one chapter for each zone, with a narrative list of permitted uses and a list of setbacks and height limits. A separate chapter described the procedure for amending the zoning text or the zoning map. Importantly, we could change the text without changing the map (everyone in the affected chapter now had new rules for the future), and we could change the map without changing the text (owners in the recolored area now just followed a different set of rules). This is important, because this book is about changing the text of the zoning ordinance—not the map that applies it to specific parcels of land in the city.

Of course, even back in 1916, people were smart enough to know that no rule will work perfectly. For example, there may be parcels of such a weird shape that they physically cannot meet the setbacks, or instances where the height limits make no sense because the property is in a valley and the surrounding buildings would tower over it. So they allowed for “variances,” which required an appearance before a “board of adjustment” to show that yours was a unique situation created by history, or by the land itself, and not by you yourself. You also had to show that, because of that uniqueness, applying the rules rigidly would produce a result that was unnecessarily harsh on you and that flexing the rules for you would not create significant problems for your neighbors. If you created the problem yourself (for example, by selling off some buildable parcels in a way that left you with an unbuildable one, or because you just paid too much for the land and you needed a taller building to break even on the deal), then tough luck. You couldn’t create a problem and then ask your neighbors to bear the burden of fixing a problem you could have avoided. This was all very logical and straightforward—a set of simple rules and a way to make exceptions in unique circumstances.

These basic Euclidean zoning tools did not change much until after World War II. The structure of the system was robust. We could do a lot with it. The original New York system had three colors—residential, commercial, and industrial—but there was no reason it could not have six colors, or twelve, or twenty. Instead of a “business” zone, you could have “business 1” and “business 2” zones to cover different types and scales of building. If you wanted a residential zone where the uses and boxes would allow only single-family houses, just create one and recolor a part of the zoning map to the new color. If you wanted a residential zone that makes each box sit far from the others—in effect requiring larger lots and encouraging more expensive houses—just create one. If you wanted to have one zone for heavy (smelly and smoky) industry and another for light (clean and quiet) industry, create them.

So Euclidean zoning remained largely unchanged in concept, but it became more complicated in practice. Denver adopted its first zoning ordinance in 1923 and its second ordinance in 1957. Table 1.1 shows how the menu of districts within the city changed over time. Not only were the basic categories (residential, commercial) being sliced into more varieties, but entirely new types of zones appeared.

Zones typically are shown on maps in different colors—lots of them. Residential zones are often shown in yellow, so as residential zones multiplied, we could use dark yellow and then light yellow and then lemon yellow. Anyone with a computer monitor knows that there is no shortage of different colors out there—but there is a limit to how many variations of yellow the human eye can distinguish when they appear next to one another on a map. Multifamily residential is often shown as somewhere between yellow and brown, but not all of us can spot the difference between dark orange, rust, and burnt sienna without constantly referring back to the map legend. So even though each new zone could have its own color, both the public and the planners sometimes had trouble keeping track of what the different colors meant. This foreshadows one structural problem in thinking of each zone district as a world unto itself.

Table 1.1: Denver Zoning Districts

1923	1957	1994
5 residential	5 residential	16 residential
3 business	6 business	12 business
3 commercial	1 office/institutional	7 mixed-use
2 industrial	1 park	2 office
	3 special	1 park
	3 industrial	3 industrial
		1 planned unit development (PUD)
Total = 13	Total = 19	Total = 42

As the number of Denver’s zone districts expanded, the number of topics covered also expanded. The 1957 ordinance covered new topics like parking; loading; the emission of heat, glare, radiation, fumes, and vibration; zone lots (for cases where zoning was applied to something other than a single platted parcel); and special zone lot plans.³ By 1994, the list of topics was too long and complicated to summarize. To its credit, Denver is now engaged in an ambitious effort to simplify the ordinance. Similar changes were taking place in many other cities over the same time period.

Perhaps the most important addition to Euclidean zoning between 1920 and 1960 concerned the automobile. Particularly after World War II, it became clear that even if the building uses and boxes were compatible with those around them, some of those uses attracted lots of cars that clogged the streets and annoyed the neighbors. Addressing this issue did not require the creation of a new zone but instead required the addition of a chapter that was stapled to each zone through cross-references. Now you had to provide a certain number of parking spaces that varied depending on your land use, and those provisions applied in nearly every zone district. This was an important change. Now zoning regulations did not vary just by zone district; some regulations varied by specific use rather than by the zone in which it was located. When sign controls came along, they were handled the same way.

Something similar happened with floodplains. Planners and engineers have always known that some places flood and they worry about it. Engineers try to keep it from happening and planners try not to have lots of people where it happens. But every type of zone district may have some parcels of land that are located in floodplains and others that are not. One solution might be to create more new zone districts—for example, for every “business 3” zone, there would also be a “business 3 floodplain” zone with fewer allowed uses and more prohibited types of buildings to protect public safety. But that would double the number of zone districts (and colors). To avoid that result, some planners decided to create a single new zone called floodplain and then “lay it over” all of the mapped flood areas. This was like drawing the floodplain boundaries on a clear sheet of plastic and taping it over the zoning map: you could see right through the plastic to determine the color of the zone where your land was located, and you could see whether the additional restrictions for the floodplain applied. There was no need to double the number of districts, but the practical effect of an “overlay zone” was the same. Again, this was an important change. These regulations varied based not on the underlying zone or the specific use but on geography or some other factor.

So now we had lots of zone district chapters describing uses and boxes, some overlay zone districts that might apply to all or none or some of the underlying districts, and some freestanding chapters (like parking and signs) that applied pretty much everywhere. The document was becoming more complex. There was a lot of overlap between chapters—you had to look in more than one place in the ordinance to find out what you could do on your property. But that was better than repeating identical parking and floodplain text in every zone district chapter, which would really have made the book an exercise in heavy lifting.

Along the way, planners and lawyers were discovering one of the wonderful loopholes of zoning law. Although state enabling laws often required that every property within a zone be treated the same, there was nothing to prevent you from creating as many zones as you wanted. You could create a new zone even if you suspected that it would be used only once, and you didn’t have to admit to that when you adopted the text of the new zone; you just mapped one piece of land, and in the future you could color other parts of the city into the new zone if you wanted. But if tomorrow never came, so what? As far as the law was concerned, everyone whose land was in that zone (all one of them) was being treated the same. A legal doctrine on “spot zoning” emerged to stop unfair use of this loophole if the rules that were applied to one parcel differed too dramatically from the rules applied to the neighboring parcels (see chapters 4 and 5), but it was used rarely and it did not address the proliferation of almost-but-not-quite-identical zone districts.

Although zoning was now in common use, not everyone was happy with it. In addition to generalized angst by those who thought there should be no land use controls, criticism of Euclidean zoning came from two other quarters. First, there were those who wondered whether the system was getting too complex. In theory, there was no limit to the number of zone districts that could be created. Didn’t that start to erode the idea of uniformity behind the use of zoning districts? How much time and effort did the city want to put into operating this zoning system, and how hard would it be for the public to understand the regulations that applied to their property?

Second, some folks—lawyers mostly—wondered whether the move from a few general districts affecting large classes of people to many smaller and more specialized zones opened up possibilities for unfairness and abuse. Changing a widely used zone district, such as “business 3,” could not be done in the dark. Text changes usually required newspaper notice—not individual notice to every property owner—but that works well only for large districts with many owners. Many “B-3” landowners would see the newspaper notice, and some inevitably would object. A debate would follow and city council would at least be informed about the objections when it decided whether to approve the amendment.

But as zoning districts proliferate, the more specialized zones often apply to smaller groups of landowners. A general “B-3” zone district might be refined so that a new variation called, for example, “B-3.2” applies only to properties along collector streets, and then further refined so that a variation called “B-3.2.5” applied to properties along minor collectors with significant industrial traffic. Amending the text of the “B-3.2.5” district would affect only a small number of properties and might not get much attention from the media or industry groups. It might be harder for the landowners to learn about the proposed change, and if they objected to the amendment their few voices might sound like only minimal opposition.

The expansion of zoning to cover new topics also had consequences. One predictable outcome was that variances were needed more often. A site that was buildable before the adoption of parking or landscaping standards might not be once those requirements were added. Over time, boards of adjustment found that they had a steady, if not daunting, workload. In some cities, the board’s hearing calendar was backed up for months, even for property owners who were requesting something minor. The basic idea that land could be regulated through a few simple zones addressing a few topics, and that exceptions would be rare, was on shaky ground. But in spite of its weaknesses, Euclidean zoning was still the bedrock of most American cities’ land use controls. So many people rely on the predictability of the Euclidean zoning approach that increasing complexity and concerns about fairness and due process have still not discredited this most basic tool.

Standard Zoning and Planning Enabling Acts

When New York adopted its zoning plan in 1916, it acted alone. There was no national model that it “tweaked.” But the same problems were occurring in other American cities, and the idea caught on. Edward Bassett, chairman of the New York committee that drafted the 1916 zoning ordinance, was a true believer who visited every state in the United States between 1917 and 1927 to spread the gospel of zoning.⁴ By the 1920s, America was in the midst of the municipal reform movement, and talk about better ways to run cities was on many lips. This was the era that gave us the city manager form of government, merit-based municipal civil service systems (as opposed to patronage), and requirements for competitive bidding for city contracts. Zoning became associated with the reform package, and it spread in part because of the broader wave of urban reforms washing over America’s cities.

During his years as Secretary of Commerce and then President, Herbert Hoover helped lead the movement toward progressive urban government and took a great interest in planning as a solution to urban problems. By 1922, some states had already adopted enabling acts for zoning and some cities had adopted zoning laws without waiting for state authority. Hoover appointed an Advisory Committee on City Planning and Zoning, which produced the Standard Zoning Enabling Act in 1924 and then a revised version in 1926.

Three things about the Standard Zoning Act are notable. First, it was just a publication. This “Act” was not an “Act of Congress.” The federal government published it as an example of progressive legislation that individual states might want to adopt. No one was forcing cities to zone, but some in the federal government thought it was a good idea. If your city wanted to zone, this was a good place to start. Second, the Standard Zoning Act required that zoning be “in accordance with a comprehensive plan,” but it did not recommend what specific controls zoning should contain. It focused primarily on procedures: how to adopt and amend a zoning ordinance and how to allow variances. Third, one of the most strongly debated sections of the Zoning Act was the one authorizing variances—the rules on how much latitude the board of adjustment should have to vary the zoning rules and when they could do it. This debate foreshadowed ninety years of continuing discussions about how zoning should reconcile the need for both predictability and flexibility.⁵

In 1928, the Advisory Committee went on to publish the Standard City Planning Enabling Act, which outlined the establishment of a planning commission to draft a “master plan,” street plans, controls for the subdivision of land, and even the need for regional planning.⁶ Again, this was a publication of the Commerce Department, not an “Act of Congress.” These two publications very effectively promoted both city planning and zoning. By 1930, the U.S. Commerce Department reported that thirty-five states had adopted legislation based on the Zoning Act and that ten states had adopted statutes based on the Planning Act. These were ideas whose times had apparently come.

The sequence of these events, though, has created lots of trouble for planning and city management ever since. Most importantly, creating legal authority to regulate land before creating the authority to plan for future land uses puts the cart before the horse, at least in the minds of planners. It reflected a “ready-fire-aim” approach to land use and suggested that it was OK to set the rules for what could be done with land based on current conditions—and without thinking about the future. In publishing the two Standard Acts separately, Congress was only suggesting that states might want to do both, or one, or neither. Although most well-run cities now support both planning and zoning, the connection between the two is often ...