(The following is an edited, abridged and updated version of the section LEVEL OF AMENITIES DICTATES A MAXIMUM FOR DENSITIES from “Populating Urbanism,” published as an essay in Living Urbanism, Issue 1, April 2008; the full text can be found here)
See introductory notes here.
Intuitively we understand people crowd onto the island of Manhattan because it provides global level amenities. Residents pay high prices for small dwelling units in exchange for easy and frequent access to an incomparable variety and quality of services and opportunities. Conversely, we would not expect individuals, by choice, to tolerate such densities where comparable amenities/services are lacking. In other words, places with a R6* level of density should have R6-level amenities and residents of R3 places should not expect to get R4, 5 or 6 level amenities.
Places should not expect to get a quality and quantity of services and amenities well beyond the population’s ability to support those services and amenities. Returning to the Manhattan example, in order to gain access to the global level of services, in addition to submitting to higher densities on the island, the city must tolerate a large numbers of interlopers, both daily commuters and out of town visitors. Many urbanists would argue that the massive allure of such central cities for visiting populations is the lack of supply of urbanism elsewhere in the area; I would agree that is part of, though not the whole, answer.
I will propose that service level for a town, village or neighborhood could be classed by the frequency with which we need the top-level services offered:
(R2: Rural; R1: Natural – Limited service)
We have started with an anecdote at the global level, now we jump to the opposite end of the spectrum. Every urban neighborhood should have services sufficient to meet daily needs. Only truly rural places should not be within the ordinary catchment** of a center with daily services. In a place with only daily amenities, residents should expect to easily fall below the low levels of auto traffic suggested by Appleby and Lintel.
What comprises daily services? Today in the United States that would likely mean the corner or convenience/drug store. A coffee shop, a gym, daycare, school, laundry or an Automated Teller Machine may also fall under this category. In Europe, daily needs could be the butcher, the baker and the pub. What we consider our daily needs is not entirely fixed; it varies between cultures and even between individuals. The key however is that such a neighborhood should not expect to draw in a great deal of traffic from the outside, because most of the services present are not unique. Every other neighborhood should have and be able to support the similar services and therefore the incentive for cross-travel is low.
This analysis of “daily services” is admittedly blind to one thing most of the population does every day, which is go to work. As our society is constructed today, whether traveling by car or transit, most people do not work in the same neighborhood where they live. This analysis does not attempt to resolve that. As one goes up the ladder from R3 to R6, just as variety and number of services improves, the likelihood of being able to live where you work increases. The acceptable balance of this mismatch is related to “visitor population” tolerances addressed in “Populating Urbanism”.
Realistically many working Americans do little on workdays other than go to work and return home so there are very few needs that are truly daily (whether this is a result or a cause of the lack of neighborhood services is open for debate). In fact some of the most common services offered in many new urbanist town centers, like boutique clothes shopping, are monthly or yearly needs. This may help explain why town centers in many new urbanists neighborhoods often struggle; the fact is that most residents don’t need the services offered all that frequently. Similarly the willingness to make trade-offs (e.g. live at higher densities) to be able to walk to those services low.
WEEKLY AND ABOVE
R6-G: Rarely – a nationally recognized Opera, The United Nations, Mann Chinese Theater
R6: Occasionally – Community theater, Courthouse, Sports Venue, Local Symphony
R5: Monthly – Movie theater, Discount Retailer, Clothes shopping
R4: Weekly – Grocery Store, Pharmacy, Doctor’s Office
R3: Daily/Limited – Cafe, Daycare, ATM, Laundromat
Weekly needs could include large grocery stores in the United States, banks and restaurants. As the level of services increases people should expect to experience increased outside traffic (this after all is where those who live in “daily services” neighborhoods come for their weekly needs) and to live at increased densities. It will also be more desirable for individuals with more disposable income, who take advantage of expensive services more frequently, to live in such locations. These neighborhoods should include all the daily needs, likely in greater quantity and variety, plus all weekly needs. Inman Square in Cambridge, MA (primarily 3 story structures), provides not just one corner store but one on virtually every block in addition to a wide range of other services.
At the “weekly” level it is still conceivable that users within the neighborhood could support most of these businesses. However, if we assume that some places do not have these levels of amenities then that implies that those who live in “daily need” places will take advantage of some of the services available in “weekly+ need” places. Therefore these outside users must be taken into account.
At a macro-level one could calculate how many “weekly need” places must exist to serve a certain number of “daily need” places while ensuring that the residents and streets of the “weekly” needs place are not overwhelmed by visitors. Auto and pedestrian users both have relevant considerations. Total visitors is the number of people walking down the sidewalk (after they get out of their cars, off their bikes, off the train or even after walking from another center) who do not consider this place their home territory. Too large a visitor to resident ratio will start to break down the character and identity of a place and make it dysfunctional from a retail standpoint. That ratio should be different for each place type (in a weekly neighborhood it may be 0.5 : 1 at any given moment and in midtown Manhattan it may be 10:1–perhaps this is why many Manhattan residents avoid midtown).
Since each visitor only patronizes the neighborhood weekly, there can be many more “dependents” than actual residents but they are not present all at the same time or if they are it is in concentrated bursts, such as weekdays or events. For example, if the “daily need” places have ½ the population of the “weekly need” place, then a “weekly need” neighborhood could support 7 “daily need” neighborhoods in addition to it’s own population. These numbers are not prescriptive but are intended to give a rough idea of the potential magnitude of these relationships.
These calculations of visitor population must be paired with calculations for supporting business. The combination of visitor and resident population must be able to support the level of amenities without the ratio of visitors to residents being too unbalanced. Whereas for weekly need places it is conceivable that the local population could support most of the businesses based on higher population density, in higher categories–such as “occasional needs” places–visitors will be required to maintain business viability. As amenity level increases, residents’ tolerance for living at higher densities and for motor traffic, pedestrian traffic and visitor population also increases. These factors are related to one another: as density and pedestrian traffic increases, strangers become less distinguishable and motor traffic becomes less imposing. At higher amenity levels we can also assume that visitors’ willingness to walk greater distances from adjacent neighborhoods for weekly and occasional needs increases.
Managing the cars
Visitors who arrive by car will count against the maximum number of car-dependent users that a place can support. This is the total number of visitors arriving by car AND those who consider this neighborhood their primary commercial center / home territory but travel to it primarily by car. The number of people that can arrive by other means of transportation, such as bicycle or public transit, is more flexible though the maximum comfortable visitor population still applies.
At this level, it is logical for transit to enter into the conversation. We have discussed limits on visitor population without fully distinguish among visitors traveling by different modes. If most visitors come by car then tolerable visitor population may be effectively lowered unless transit is provided; if those additional visitor are required to sustain a certain level of business/civic activity (e.g. R6) then transit becomes necessary. Even with increasing local density and traffic, eventually the level of amenities will require more users than can comfortably either live within walking distance or arrive by car. This will very likely be the situation for any place with monthly, occasional or rare services. Thus transit can increase the pedestrian density and visitor population without increasing the motor traffic levels. This is not the only time that transit ought to be available but it is the level where it must be available.
One interesting idea for accommodating a wider range of services for a small population is the idea of intermittent services. This could include a weekly farmers or trade market. A colleague travelling in southern France remarked that the weekly or biweekly market held in the towns was the equivalent of supermarket and Walmart in one. It could also include more rare travelling events such as a circus or fair or local events such as town festivals. Especially as the economics of building place become more restrictive we should consider programming flexible spaces for such activities, sometimes in lieu of permanent buildings.
This article has suggested a hierarchy of places based on intensity as well as some considerations that are necessary to maintain these places in appropriate balance. Much of the information presented here may ring familiar. The goal is not to present and entirely new concepts but to provide an alternate way of thinking about ideas we already understand intuitively.
Additionally the earlier paper from which this is excerpted provided a logic for minimum densities. The concept of minimum densities and minimum population remains foreign to many planners, decision-makers, and citizens outside of New Urbanism and a firm logic may help them to understand and accept it.
FURTHER RESEARCH: Empirical tests or economic analysis could suggest the appropriate relationship between density and amenity levels and research could classify different services by frequency of use in order to classify specific services as part of daily, weekly or rare usage categories.
*As noted by Bruce Donnelly and others “transect zones” are intended to describe parts of a neighborhoods whereas people commonly misconstrue them to mean parts of a region. It would seem that in my original paper on this topic I made that very error or glossed over it. As such I have replaced, where appropriate, the mentions of T1 through T6 with R1 through R6. R1 – R6 does not represent an agreed upon taxonomy.
**The ordinary catchment here means the type of village or area described in “Populating Urbanism”, consisting of the population within walking distance of the center and a limited number of auto-dependent users.
NOTE: These suggestions are beholden to Christopher Waller’s Central Place Theory and Christopher Alexander’s “Web of Shopping” (Pattern 19). However, they do not necessarily incorporate all the related assumptions of those two authors.