At Discovering Urbanism, Daniel Nairn offers an interesting summary of Edward Murray Bassett’s 1922 defense of zoning (available as a free e-book). Bassett faced opponents who were against a new type of land use regulation, many arguing that zoning was unconstitutional. In retrospect, some of his arguments defending zoning are comical. He asserts that zoning would never go so far as to direct aesthetics because the courts would protect us from the overreach. It would be interesting to hear what he’d have to say about a planning commission meeting today. Nairn’s entire analysis is interesting, but I was particularly intrigued by Bassett’s assertion that zoning fosters cooperation. As Nairn summarizes:
Cooperation yields overall larger return on investment for all property owners. This was Bassett’s primary concern, one that he underscored with a number of prisoners’ dilemma scenarios. For example, “In some of the larger cities a landowner in the business district is almost compelled to put up a skyscraper because if he put up a low building, his next neighbor would put up a higher one that would take advantage of his light and air.” He asserted that skyscrapers were probably not a sound investment in their own right, but they were built anyway in a virtual arms race for public goods of light, air, privacy, and scenery. Zoning was the truce that made everyone better off.
I’m not sure that I follow Bassett’s logic here. If light and air are only available on floors that are higher than the floors of the neighboring buildings, then only the top few floors of any building would typically have this asset. It’s almost as if he’s talking about a race to the highest roof deck here. Aside from the problems with how he makes this argument, it is worth a look to determine whether or not zoning takes a positive step toward cooperation in the land market. Whether or not an institution fosters cooperation is a key factor in determining its success or failure. With cooperation, trade becomes a positive sum game rather than a negative sum game. For example, property rights is an institution that clearly fosters cooperation; when they are not well-defined, as in black markets, trade is often accompanied with violence.
Restrictions on land use, whether they come from public sources (zoning, height limits) or private sources (deed restrictions, HOAs) face trade offs between providing clear expectations of future development and permitting flexibility as land’s highest-value use evolves over time. On the far end, deed restrictions make it difficult to impossible to change land use restrictions, while HOAs and BIDs can often change restrictions with super majority votes from their members. Of course HOA rules often veer toward the draconian, but they are easier to overturn than other types of regulations. HOAs and BIDs also lack the stability of government entities. Since they are not likely to be around as long as cities, the time horizon of their rules may be indeterminate in some cases.
According to Bassett, zoning represents the best of both worlds, a compromise between permanent deed restrictions and rules that can be overturned too easily. On the one hand, it allows a landowner not to worry about his neighbor “taking advantage of his light and air” by prohibiting buildings taller than what zoning permits. Bassett writes from the perspective of developers and suggests that building skyscrapers is much like an arms race. He asserts that because elevators take up square footage that cannot be leased, skyscrapers are less profitable than lower buildings. He suggests that the only reason for building skyscrapers is to prevent the next door building from casting a shadow on a wasted setback. Of course it is the case now, as then, that when developers build tall buildings is because they think the net present value is greater than that of a shorter building. Only the exceedingly rare developer who doesn’t want to make money would pick a building design for the purpose of not allowing his neighbor to take advantage of light.
Many people have made the argument that tall buildings produce externalities, but he doesn’t quite identify these externalities correctly; they do not fall on the developer who wants to prevent his neighbor from taking free light and air, but rather on the owners of shorter buildings and their tenants. Bassett’s argument is instead more reminiscent of the fallacious Marxist argument that competition among firms hurts welfare. So, in my estimation, he does not make a strong argument that zoning produces cooperation by preventing a race to the bottom, or top, as the case may be. Collusion among building owners to restrict building supply may be a form of cooperation, but it’s not the type that benefits society.
As Nairn summarizes, Bassett also support zoning because:
Zoning stabilizes building and property values, by signaling to investors what they can expect from a certain district. Markets work when people know what they are buying, and zoning creates some assurance that the product will not change fundamentally. This reason is why housing developers were among the most ardent supporters of zoning in the early stages.
For much of the history of Euclidean zoning, this may have been its most important quality. Once cities introduced comprehensive plans, it was clear what types of buildings could be built on each land parcel, providing rule of law in the land market so that buyers and sellers both knew land’s potential uses, and all buyers and sellers were equal before the law.
Today, one of the biggest problems with zoning is that this rule of law has been eroded in some major cities. Rather than introducing broad changes, such as widespread upzoning, to meet cities’ evolving needs, planners in cities such as New York and DC have taken the approach of relying on variances and Planned Unit Developments as well as requiring projects to achieve neighborhood approval. Going into the approval process, developers often don’t know whether their plan will be approved or denied, and this makes buying and selling properties highly speculative. These tools permit flexibility in land use, but at the great expense of losing the rule of law that is key to allowing market participants to form expectations for the future. Without the rule of law, planning approval processes that don’t rely on as-of-right development are the worst of all worlds; they limit potential opportunities for trade while introducing uncertainty into the market.
“Today, one of the biggest problems with zoning is that this rule of law has been eroded in (all) cities…. planners in (every city) have taken the approach of relying on variances…”
Once one believes that they can determine the appropriate shape of development across a city and is given the power to do so, how precise (how many degrees of differentiation) would all the zones have to be set up to make variances inappropriate?
Houston has two zones “urban”(still pretty suburban by larger cities’ standards) and “suburban”, and hands out variances like candy on halloween.
Apparently even cities with more differentiated zones still must (or are paid to) hand out variances.
Sure, it would be hard to imagine any zoning plan that could facilitate development without allowing variances. It seems, though, that we know there is a problem when as-of-right development almost never takes place. DC Planning is currently in the process of creating a new zoning code with the objective of updating zoning to better meet neighborhood needs. However, it’s hard to imagine that city planners and ANCs will be willing to let go of micro control of projets now that that has become the norm.
Having a rigid and unchanging zoning code certainly hasn’t stabilized property values in Detroit. In fact, given zoning’s tendency to segregate uses I would imagine that it contributes to property values declining over time rather than remaining stable because increased infrastructure costs so people can drive between their widely seperated places of living, working and shopping drive up property taxes and so people invest less in maintaining their properties and keep moving further and further out, leaving housing that can only be sold for lower than the price paid for it new.
It will be interesting to see if Detroit’s policy of stopping some municipal infrastructure services in parts of the city will succeed in getting people to move closer in.
if as-of-right development was common that would mean that the zoning generally isn’t binding, and that we are obviously paying our planners for nothing.
as opposed to the current situation where we are paying our planners to restrict growth and development.
Variances are important for zoning, because without them, the entire zoning regulation could be thrown out as the result of any court case going against a zoning regulation due to ‘hardship’ which would create chaos.
The Bassett defense of zoining is a very interesting topic, and I’m hoping to write more extensive comments about it later if I have the time. But for the moment I’d like to address the issue that’s been brought up of variances.
I’m not sure if I understand the positions that are being taken.
– – – – – – – – –
a) Emily Washington originally wrote:
Today, one of the biggest problems with zoning is that this rule of law has been eroded in some major cities. Rather than introducing broad changes, such as widespread upzoning, to meet cities’ evolving needs, planners in cities such as New York and DC have taken the approach of relying on variances . . . .as well as requiring projects to achieve neighborhood approval.
b) and AWP then wrote:
Apparently even cities with more differentiated zones still must (or are paid to) hand out variances.
c) and then Emily Washington wrote:
Sure, it would be hard to imagine any zoning plan that could facilitate development without allowing variances. It seems, though, that we know there is a problem when as-of-right development almost never takes place.
d) and then AWP wrote [added text within brackets is mine]:
If as-of-right development was common that would mean that the zoning generally isn’t binding [and therefore not really doing its job], and that we are obviously paying our planners for nothing . . . . as opposed to the current situation where we are paying our planners to restrict growth and development.
e) and then Metalslyrunner wrote:
Variances are important for zoning, because without them, the entire zoning regulation could be thrown out as the result of any court case going against a zoning regulation due to ‘hardship’ which would create chaos.
– – – – – – – – – – –
If I’m understanding the above correctly (and perhaps I’m not), people seem to be conceding that a zoning code has to have lots of variances in order to work; and if a zoning code doesn’t require a lot of variances, it’s not much of a zoning code to begin with (although it’s also conceded that some zoning codes require more variances than others); and that a zoning codes needs to be dependent on lots of variances, otherwise it would be thrown out in court. If this is what is being said, I’m inclined to be skeptical.
Although I can’t say I’ve done a lot of research on the topic, it seems to me, at least from general reading about the history of NYC, that a proliferation of variances, at least in NYC, might actually be a relatively “recent” phenomenon (say, from the 1970s or so). (And, even given this relatively “recent” proliferation of variances, Josh Barro, the author of the article that was linked to, still seems to believe, at least from a quick skimming of his article, that even the current situation in NYC is much, much better than the one in Washington, D.C.)
But it also seems to me (again mostly from general reading and, admittedly, not from heavy research) that NYC’s original 1916 zoning code (which is the one that Bassett is touting) didn’t generate variances to the same degree as today’s code currently does — and that this code was also strict enough to satisfy Bassett’s belief in controlled growth. As an example of the code’s “looseness,” if I remember correctly from my readings, Rockefeller Center, for example, was built entirely “as-of-right” in terms of zoning. And buildings, like the McGraw-Hill Building and the Daily News Building, were also famously built “as-of-right” too. Because they contained manufacturing — printing presses — they were built just outside the mid-town office district. (All of these projects, by the way, had the business minded Raymond Hood as lead architect.) It also seems to me that a number of other significant projects in NYC were also built without recourse to variances. For instance, at least as far as I know, the wonderful 27-story (?) art deco One Fifth Avenue apartment house / hotel in Greenwich Village, which is adjacent to a row of townhouse and a mews of carriage houses, didn’t require any variances in order for it to be built.
My guess is that part of the reason for the relative dearth of variances (if this was actually true) was due to the “wisdom” of the zoning code itself (in other words, it was pretty loose in some ways), and part was due to the “culture” of the time (e.g., people less commonly thought they “deserved” variances, etc.).
From my, admittedly limited knowledge of the 1916 code, I think the 1916 might have been even better if it were stricter about certain things (e.g., as Jane Jacobs points out, regulating horizontal scale) and looser about others (permitting more mixed uses). But, nevertheles, it seems to me that it was a good starting point — better than the current 1961 zoning code — and a good illustration of how a code can be control growth and be loose at the same time.
Benjamin Hemric
Sat., June 2, 2012, 9:25 p.m.
Yes, Josh Barro makes a distinction between NYC having the rule of law in development as opposed to DC which does not. Josh may very well be right that DC, where almost no projects are built as-of-right has a weaker rule of law than NYC. But I think it’s hard to argue that either city has the sort of stable policies regarding land that would allow the market cooperation Bassett advocates.
I have not done historical research, either, about the use of variances over time, but it does seem that both cities, along with many others, have moved away from a situation where clear rules govern development to a situation in which rules are created on an individual basis.
It seems you are agreeing with me.
to rephrase;
Variances are important for zoning, because without them, no feasible zoning regime could be created that is also reasonable.
Essentially my point is that without variances zoning would have to be extremely precise (down to the lot, block, or sub-neighborhood) in order to be reasonable.
to rephrase again;
We have to give certain people permission to break the rules, because the rules are illegal.
I agree.
Also, the end of zoning would not lead to anarchy.
“He suggests that the only reason for building skyscrapers is to prevent
the next door building from casting a shadow on a wasted setback. Of
course it is the case now, as then, that when developers build tall
buildings is because they think the net present value is greater than
that of a shorter building.”
Your second sentence does not directly refute the suggestion in the first sentence. If, hypothetically, buildings in a shadow were extremely undesirable and had near zero value, then the net present value would indeed be increased simply by building high. I understand in fact shadowed buildings are still quite valuable, but that doesn’t mean that the “prisoner’s dilemma” is incorrect, only that its negative effects are smaller than they could otherwise be.
“NYC’s original 1916 zoning code (which is the one that Bassett is touting) didn’t generate variances to the same degree as today’s (from 1961) code currently does (today)”
1)zoning codes generally restrict density according to some function of the density present when the code was written2)redevelopment (almost always) tends to higher density as the city continues to growTherefore with the passage of enough time and the continuing growth of a city, the “market density” becomes tightly bounded by the zoning code, and continued redevelopment will only be allowed ifa)there is a general upzoning, orb)variances are grantedEmily’s seems to be arguing for A as opposed to B, in order to satisfy order and the “rule of law”. This also explains why variances would become more common with the passage of time if A does not occur. Also, without more historical knowledge (admittedly I have none) we can not be sure if the 1916 ordinance was looser or if more growth has occurred since the 1961 ordinance.If anyone is willing to do the research I would be willing to bet that (variances/redevelopment project) increases the longer it has been since a change in zoning.
awp,
Hi! I hope you don’t mind that I am posting a reply to your comment as an independent comment — I hate “nested” comments.
– – – – – – – – –
awp wrote:
1) zoning codes generally restrict density according to some function of the density present when the code was written
Benjamin Hemric writes:
Although I am not as knowledgeable about the NYC’s 1916 zoning code as I would like to be, my understanding of it from what I do know is that this code differs significantly from the subsequent 1961 code (and perhaps most other “modern day” codes too) as it was not really focused on controlling density in the first place (although its authors may, or may not, have had that in the back of their mind). The 1916 code was instead focused on restricting “light and air” by regulating heights and set backs and, thus, only restricted density indirectly — and incompletely. (It also was interested in restricting land use via other mechanisms.)
The basic set-up for the “light and air” restrictions was thus: Depending on the width of a street, a building was allowed to go up only so high before it had to be set back a certain number of feet from the street. Then it was allowed to go up only another “x” number of feet before being set back further. At a certain point, a builder was entitled to build a tower as high as he could go (and as high as was technically possible) as long as the tower didn’t take up more than “x” percentage of the building lot.
This is the zoning code that allowed for the construction of the Empire State Building, the Chyrsler Building, the GE Building (originally the RCA Building) etc. It didn’t focus on density; there is no mention (as far as I know) of floor area ratios; and, at least theoretically, there was no upper limit on density — only a technological limit.
This is very different from the current 1961 code which is, indeed, so it seems to me more interested in regulating density than preserving light and air. The concept of floor area ratio is central to the 1961 code, but sky exposure planes (correct expression?) are “violatable.” So, although the 1961 zoning code also preserves light and air to a degree, it seems to me that the priorities shifted.
– – – – – – – – –
awp wrote:
2) Redevelopment (almost always) tends to higher density as the city continues to grow. Therefore with the passage of enough time and the continuing growth of a city, the “market density” becomes tightly bounded by the zoning code, and continued redevelopment will only be allowed if A) there is a general upzoning, or B) variances are granted.
Benjamin Hemric writes:
Although I’m not positive about it, I don’t think this was true with regard to NYC and the 1916 code.
Despite perceptions, it seems to me that only a very small part Manhattan was built up to the market density of the 1916 code — since the code seems to have been rather generous and there were, in 1961, very large parts of Manhattan that were still “undeveloped” by today’s standards (e.g., mostly six-story tenements, row houses and low loft-buildings, etc). (Look at photos of NYC’s skyline in the 1950s and it looks positively deserted by today’s standards!) And that’s not even to mention vast portions of NYC outside of Manhattan.
From my readings (again, admittedly, incomplete), one of the reasons for replacing the 1916 code with the 1961 code was that, indeed, the 1916 code was considered too generous. Plus, “[sub]urbanists” were seduced by the concept of skyscraper towers on plazas and wanted to make it easier to build such structures (while still staying within the bounds of the theoretical FAR that was “generated” by the 1916 code). I’m not sure about this, but I think the theoretical basic “FAR”of the 1916 code was actually decreased by the 1961 code and only allowed up again if a builder provided a plaza or other “amenity.” So, in a sense, NYC may have been downzoned with the adoption of the 1961 code.
– – – – – – – – – –
awp wrote:
Emily’s seems to be arguing for “A” [general upzoning] as opposed to B [the granting of variances], in order to satisfy order and the “rule of law”.
Benjamin Hemric writes:
As mentioned in my original comment, I wasn’t sure what anybody was really trying to say in the original comments. But it seemed to me that people were saying something different — that it is essentially impossible for one to have a zoning code that both restricts development effectively (by some standard) and have one that is operable without a large number of variances. And if this is what people were saying, I was (and am) skeptical, as the 1916 zoning code seems to have allowed for a tremendous amount of growth in NYC between 1916 and 1961 (it didn’t seem to hamper NYC any!); it didn’t seem, as far as I know, to require many variances (e.g., you could apparently build the Daily News Building, the (original) McGraw-Hill Building, the Empire State Building, the Chrylser Building, Rockefeller Center, etc., etc. without needing variances); and it also restricted growth in a way that was important (i.e., “light and air”) and , at least originally, satisfactory to planners.
Again, it seemed to me that the 1916 zoning code was changed because “planners” were enamored of towers on plazas and because they became “greedy” and/or emboldened and now wanted to micro manage growth more than they did in 1916 (in terms of FAR, etc.). But from my perspective, the 1916 code was essentially just fine, as it restricted what it seems to me “should” be restricted (i.e. “light and air”) — and, even then, in a rather loose way — and, it more or less left alone, what is better “restricted” by the marketplace (e.g., density).
– – – – – – –
awp:
This also explains why variances would become more common with the passage of time if “A” does not occur.
Benjamin Hemric writes:
While this seems to me to make sense theorectically, I don’t believe (at least from my readings so far) that this is what happened in terms of NYC’s zoning codes.
My current tentative theory about the increase in variances (at least in NYC) is that “incentive zoning” (introduced in NYC in the early or mid-1970s, I believe), like bonsued theaters, bonsued arcades, etc. were what opened a Pandora’s box of “negotiated” zoning, and everybody — buildings, planners, community groups, etc. — got drunk with the power of “let’s make a deal” zoning. Developers get more FAR than they would have originally have gotten under the 1961 zoning code; planners get credit for providing their favorite trendy “amenities”; community groups get bragging rights for knocking down some of the FAR and fine-tunning the “amenities”; etc. etc.
Benjamin Hemric
Tues., June 5, 2012, 8:00 p.m.
Post wherever you like. I, on the other hand, do prefer the nested comments.
“The 1916 code was instead focused on restricting “light and air” by regulating heights and set backs…tower as high as he could go (and as high as was technically possible) as long as the tower didn’t take up more than “x” percentage of the building lot”
http://en.wikipedia.org/wiki/1916_Zoning_Resolution
if wikipedia is correct (which also matches your description) then the 1916 was not really zoning(as most people mean it, or at least as I mean it), but instead a regulation that still had the effect of limiting density, en extrema. It is basically a setback requirement, that changes with height of the floor being built.
——————————————————————————————————————————————————————————————————————————————————————————————-
“Despite perceptions, it seems to me that only a very small part Manhattan was built up to the market density of the 1916 code”
There was not a maximum density under the 1916 code as I understand it, especially if you are right about the increase of setback ending at some height. There would have been an effect of lower density due to the loss of the available floor area at higher levels. Instead of a 400000sqft 40 story cube the 1916 code would leave the building as a 300000sqft 50 story ziggarat(numbers completely made up), leading to higher costs and lower return to large buildings. Without a binding limit on density then my point “the “market density” becomes tightly bounded by the zoning code” is not applicable.
——————————————————————————————————————————————————————————————————————————————————————————————-
Actually we seem to agree on this third point. I am okay with certain regulations. If one really does believe that their is some negative externality to tall building in CBDs then there is a roll for govt. intervention.
If we leave the 1916 regulation out of this debate and focus on more typical zoning plans, I continue to hold my earlier arguments.
Typical zoning(restrictions on what, and at what densities, can be placed where) can either be feasible or nonfeasible(previously reasonable), and binding or nonbinding.
feasible=few zones
non-feasible= many zones, which I was previously calling reasonable to imply that there was actually a chance that the zone described something semi-appropriate to all/most properties within the zone. In order to do this zones would have to be tiny and thus there would have to be too many to track, keep up with, and update (thus non-feasible).
binding=limits development
nonbinding=does not place any actual limits on development.
——————————————————————————————————————————————————————————————————————————————————————————————-
“awp:”Therefore with the passage of enough time and the continuing growth of a city, the “market density” becomes tightly bounded by the zoning code, and continued redevelopment will only be allowed if a) there is a general upzoning, or b) variances are granted.This also explains why variances would become more common with the passage of time if “A” does not occur.Benjamin Hemric writes:While this seems to me to make sense theorectically, I don’t believe (at least from my readings so far) that this is what happened in terms of NYC’s zoning codes…..My current tentative theory about the increase in variances (at least in NYC) is that “incentive zoning” (introduced in NYC in the early or mid-1970s, I believe), like bonsued theaters, bonsued arcades, etc. were what opened a Pandora’s box of “negotiated” zoning””Developers still have to pay the costs of the negotiated/incentive zoning. They will be willing to provide more bonuses and pay more as the gap between As-of-right density and “market” density continues to widen. So with the growth of NYC, under your scenario, I would expect more concessions from developers to obtain the variances.