Wednesday, November 2, 2011

Park Economics

What's an urban park worth to a city?  The question is straightforward, but despite the attention lavished on parks in the urban design literature, the economics of parks their measurable costs and benefits have received less attention.  Although "the science of city park economics is still in its infancy," according to the Trust for Public Land, a number of studies over the past decade have helped establish a consensus on a few key points.

Attempts to determine the value added by parks date at least back to the 1860s, when Frederick Law Olmsted attempted to calculate the increase in Manhattan real estate values generated by the creation of Central Park.  That the park increased the value of the land adjacent to it was obvious, but Olmsted might have considered several other key questions, such as:
  • Even if a park increases values, do those increases outweigh the the value that would have been gained from leaving the land available for development?
  • Alternatively, could the same or approximately similar increases in value be achieved with a reduction in park acreage (and thus an increase in developable land)?
  • Are values better served by a small number of large parks or many small parks, bearing in mind that many small parks are more expensive to maintain than a few large ones?
  • Do certain types of park programming and design increase values more than others?
A growing literature has begun to provide answers to these questions.   The Trust for Public Land has put together an accounting of the potential benefits of parkland, which, however, does not consider the countervailing costs of maintenance and land lost to development.  Other writers have made more questionable arguments. Among these is the one that revenue-positive parks are an economically desirable option in comparison to money-losing single use residential development, a point which is not so much in favor of parks as it is an illustration of the financial unsustainability of contemporary development patterns.

One of the best resources on the topic is a 2001 MIT Department of Architecture thesis, which was guided by several urbanist eminences including Andrés Duany and Eran Ben-Joseph.  Although I'd encourage you to check out this paper on your own, some of the author's most important findings include:
  • Homes adjacent to parkland receive a 22% price premium, but the premium rapidly drops off for properties beyond 600 feet from the park.
  • Large parks do confer greater premiums than small parks, but the premium is small compared to the effect of being close to a small park.
  • Smaller lots place a higher value on park proximity than large lots.
In light of these findings, the author recommends a series of small parks, rather than a handful of large parks, as the property value benefits of the former will typically exceed the maintenance savings of the latter. 

The intended use of a park also has a large impact on the effect the park will have on surrounding properties, according to the Trust for Public Land.  Space devoted exclusively to athletics, for instance, may actually harm adjacent property values; open space designed as a wildlife corridor or for stormwater runoff can also be a disamenity where an unlit and forested area is perceived to pose safety issues.  Large parks may also suffer from safety issues resulting from lack of use.  Certain common purposes for open space, therefore, may not coincide with the objective of improving surrounding property values.

None of this, I think, would come as a surprise to Jane Jacobs, who through personal observation and study reached many of the same conclusions:
"Conventionally, neighbourhood parks or parklike open spaces are considered boons conferred on the deprived populations of cities.  Let us turn this thought around, and consider city parks deprived places that need the boon of life and appreciation conferred upon them.  This is more nearly in accord with reality, for people do confer use on parks and make them successes – or else withhold use and doom parks to rejection and failure."
Even more to the point:
"City districts with relatively large amounts of generalized park ... seldom develop intense community focus on a park and intense love for it. ... Parks intensely used in generalized public-yard fashion tend to have four elements in their design which I shall call intricacy, centering, sun and enclosure."
Jacobs, in Death and Life of Great American Cities, was reacting the common tendency of planners in her era to dedicate excessive amounts of land for open space, such that the space became an obvious disamenity, and a liability, rather than a benefit.  Planning decisions about parks are going to revolve around more than the mathematical calculations of a developer, but having the tools to think about parks in a more objective way can aid in the creation of parks that are both successful and financially beneficial.


  1. In reading your post and thinking back to Jacob's description, I'm reminded of Meridian Hill/Malcolm X park in DC.

    It has many of the features that both of you suggest make good parks, although it might be somewhat larger than necessary. It's on a hill, and features large steps and a waterfall that make it almost interactive. These features also split the park up to make it possible to have many uses at the same time.

  2. It's interesting to compare two parks I'm familiar with in New Orleans: Audubon Park and City Park. Audubon is smaller, but it's in a denser part of town and gets a great deal of use. On weekends the bike/jogging path around the park is like Bourbon Street on Mardi Gras.

    City Park is bigger, and gets much less use. Part of it is access, and part of it, I suspect, is simply that City Park is off at the edge of the city, in an area where homes have larger yards.

    Both parks have attractions: playgrounds, a zoo in Audubon and a museum in City Park, golf courses in both, a stadium in City Park. But one is an essential part of New Orleans while the other isn't.

  3. I would actually challenge or rather make a more subtle argument regarding Audubon vs. City Park (and I work in the parks field in NOLA). I am not sure of the statistics for Audubon, but City Park receives more than 10 million visits a year - hardly an unused park. While both see heavy amounts of daily use, overall City Park is more of a regional/destination park; a place you go to on the weekends for a special event or to visit an attraction. There are actually many pocket parks much smaller than Audubon that I think more accurately fit the "small" type suggested in the blog post - Alma Peters or Daneel Park, for example. These are an extension (or replacement) for people's yards, not as much an entertainment destination sort of park.

  4. @Emily: somehow I never made it to that park in two years of living in the DC area, but I agree, it looks like it has a lot to offer for its size. You don't see that many Italian Renaissance-inspired parks in American cities -- Olmsted's approach really won out for the most part.

    @Cambias & Anonymous: Interesting points. The post is really directed at the neighborhood park, but even a conclusion that more pocket parks are desirable doesn't rule out the need for a few much larger parks, since, as Anonymous notes, they often serve a different purpose.

  5. I would suggest that a city needs many small (very small) local "neighborhood" parks and one or a few larger "central" ones. Where the neighborhood parks are sort of a local public yard, the large central parks are in part a designed respite from the city as well as citywide destinations.

    In successful park systems, both are heavily utilized.

    Additionally, Olmsted's approach can be better said to pertain more to central-type parks than neighborhood-type ones. Although his wider urban design schemes gave us picturesque suburbia, which is the ideal that gave us what my friend today called "house farms".

  6. I agree, Steve. Although the 1961 NY zoning law deservedly gets a bad rap, the underlying intent, whether or not it was carried out well in the code, was to introduce pocket parks/plazas to Manhattan, which had virtually none prior to that time (widening the sidewalks for outdoor cafe seating might have been a better alternative, but the city does seem to be gradually warming up to that concept).

    As for Olmsted, it's interesting how his park plans and suburban plans were essentially at cross-purposes: very large parks rely on surrounding high-density urbanism for their patronage, and won't pencil out in low-density areas. The plan of Riverside had large areas set aside for picturesque park, for instance, but looking at it today it appears most has been subdivided.

  7. I agree that there should be many, quite small "neighborhood" parks. A 100x100 foot space is a small park indeed, but, if it were a private yard, it would be quite spacious (about 1/4 acre). Thus, it could serve very nicely as a "shared yard."

    Of course the nice thing about a private yard is that it is adjacent to the house. We should remember that parks (and private yards) are used in large part by small children. Think of how a park could be used by a mother with an infant, or a three-year-old, or a six-year-old. I think you will conclude that having a park very close by, let's say 200 meters maximum, is more important than having even a very large park like New York's Central Park that is a half mile away.

    Lastly, I will note that the perceived need for parks is directly related to the unpleasantness of the rest of the city. The 19th Century Hypertrophic City (19th Century New York or Chicago) was very unpleasant, especially for families with children. A beautiful Traditional City, where the area outside the house (streets) is also intended to be a space for people, is much more comfortable, and thus the perceived need for parks or other "Green Space" is very much reduced.

    Today we see that the classic Italian cities have nearly no park space at all, but they are still prime tourist destinations. Although some park space would improve them, in my opinion, you hardly hear anyone complain that Venice doesn't have any parks.

  8. I think small parks definitely have their place. It makes sense to have little playgrounds scattered around neighbourhoods. A nice thing about neighbourhood I lived in as a child had several parkettes connecting crescents and cul-de-sacs, and they often had playgrounds in them. You can see one off Nottawasaga Crescent, Brampton, Ontario.

    My house actually backed onto a large park that included a large part of a river valley and was several hundred feet across and a few miles long. It was not heavily used but we went there a lot to use playgrounds, soccer fields, build pretend homes out of branches, climb trees, etc. I'm pretty sure it increased the value of our home. There weren't any dodgy characters hanging out there.

    Many new suburban townhouse complexes will have small centrally located playgrounds too, like Deanlee Court, Mississauga, Ontario.

    Otherwise though, parks in suburban Toronto are multiple acres in size, mostly made up of naturalized areas and playing fields.

    The recent new urbanist fashion seems to involve have large grass squares. The parks are very large and don't have much in them. Ex: 47 Cornwall Drive, Markham, Ontario. I have a feeling that they built the park just for the sake of having lots of greenspace. I think some people believe that neighbourhoods need a certain amount of park space. I think parks should be designed to meet a need or serve a purpose, and that overall a quality over quantity approach is best.


  9. @Nathan: although Venice doesn't have much in the way of parks, it does have a fairly even scattering of small piazzas, such that very few homes are more than 600 feet away from one. One way to look at it is that the Italian piazza is a pro-urban vision of public space, as opposed to the American town green or park which celebrates the rural or is intended as an escape from urbanism (usually in the form of a 19th century downtown as you say).

    @Nicolas: Thanks for sharing those examples. I do think it's possible that the large grassy expanses such as in the Cornwall Drive development result from pre-established formulas: e.g., 20% park space as certain planners of the 1920s recommended, in spite of the fact that the presence of backyards reduces the need for parks. The Deanlee Court example appears much better tailored to its purpose, and as a result the space feels much more inviting than Cornwall Drive.

  10. I would definitely agree with the observation that having a really good, intimate urban fabric reduces the need for parks, even for pocket parks (though they're generally a nice thing to have). When I stayed in several European "old towns" I was so engrossed with what was going on in them (commerce, entertainment, recreation, architecture) I never felt the desire to "escape" to a park.

    IMO the public plaza (defined by and enclosed with buildings) is definitely a more desirable alternative to something like a "village green." You don't see that much "green space" in the very old parts of European cities and towns because these old towns are usually so inviting and engrossing that the thought of "escaping" them never even comes to mind. As there are few lawns, children happily play in the intimate, car-free streets. (This is an important thing that American park builders don't realize: kids aren't interested in grass and lawns. They like *paved surfaces*, as these are more conducive to hosting spontaneous childrens' games:

    Having a few ballparks is useful, but other than that there's not that much a kid ever really does on a lawn. After seven or eight years of age, you even see suburban kids leaving their front/back yards and migrating to the streets, wheeling their bikes along the way. Kids like to explore, wander, and build their own little worlds, and an infantilizing lawn or "green space" isn't conducive to a child's imagination.)

    So why do US cities have so many "green spaces" and so few plazas? We have an anti-urban bias that goes all the way back to the Jeffersonian rural ideal, and as we were pretty much founded as a resource-extraction colony, from the very beginning our cities were built as fairly unpleasant resource-extraction and industrial production centers consisting mainly of warehouses, piers, import-export businesses, commercial blocks, factories, and speculative tenements and rowhouses. Except for a very few formal recreational spaces (Philly's geometric public squares/parks, Baltimore's Mt. Vernon Place, a few formal squares in smaller cities in the south, like Savannah), we just didn't put that much thought into public beautification until the brief City Beautiful movement of the late 19th/early 20th centuries. No wonder there was an urge to "escape" our unpleasant cities fairly early on and retreat to "little cabins in the woods." This desire manifested itself in railroad and streetcar suburbs (and auto suburbs later on) as well as in giant, ruralized "escape" parks in the big cities (Olmsted's romanticized "escape from the city" parks are all we have to refer to when it comes to park design, pretty much). Today this "escape from the unpleasant city" meme continues with our constant yammering for inserting "green spaces" and "open spaces" into every built environment. We can't yet do the alternative - build intimate, narrow streets enclosed with small buildings built on a fine grain - because a whole backlog of zoning/planning/code/legal/public consensus issues make that impossible.

  11. I agree Marc. Thinking back at my childhood, a lot of the games I played required only a small patch of pavement like Four-square, drawing with chalk or skipping. Others required a bit more space, like street hockey, basketball or skateboarding, but still not that much. Playgrounds, a space to build snow forts or play with toy cars or action figures or whatever the latest fad happened to be also required little space. Then there were games for which large open spaces were a downside like hide and seek, tag and to a lesser degree snowball/water gun fights. Riding bikes or rollerblading required more space, but more like streets than big grass squares. Naturalized areas could be fun too for feeding ducks, catching frogs, climbing trees, and also tag and hide and seek, but we didn't go there as often when we were young since we couldn't go alone (so we played on driveways/sidewalks instead). Large grass spaces were only used for playing soccer, football or picnics.

    Adults don't use grassy areas as much either. Adults like to go for walks (could be done on streets), or just a place to hang out and do some people watching, which could be done indoors (ex cafes), or small squares or flower gardens.

    For the picnic/BBQ/soccer/football/naturalized area needs, I guess you could have a larger park next to some schools around a ravine further from the centre of the neighbourhood like Riverdale Park in Toronto (except with a school), and then small squares, flower gardens and playgrounds scattered around the rest, like the one around 131 Cumberland St, Toronto... which probably adds as much to the neighbourhood as the much larger Riverdale Park.


  12. Charlie
    Are you still conducting this conversation? I'm an advocate for Dix Park in Raleigh North Carolina. Is anyone talkin about parks as infrastructure for research and learning. A lab or office purposed use? A direct use benefit?