The Density Needed For Walkability Myth
Continuing the walkability theme from yesterday, I thought it would be interesting to explore the assertion that walkability requires density. So I decided to look at 1st tier suburb Kirkwood MO and 2nd tier suburb Ballwin MO to see if this is the case.  If you buy into the theory that walkability requires density then you probably think  Kirkwood is more walkable because it has greater density than Ballwin.
As you will see, walkability has less to do with density and everything to do with how the land is used, a reflection of the era in which they were created.
Kirkwood, MO:
- Established: 1853
- Total Area: 9.3 square miles
- Density in 2000:Â 2,961.3/sq mi
- Source: Wikipedia
- Walk Score:Â 88/100 – Very Walkable (101-115 W Adams Ave)
Ballwin, MO:
- Established: 1950 (incorporated)
- Total Area: 9 square miles
- Density in 2009:Â 3,494.6/sq mi
- Source: Wikipedia
- Walk Score: 62/100 – Somewhat Walkable (15025 Manchester Rd)
For the Walk Score of both suburbs I just put in the city name, it determined the address it must consider the center point.
So the older, less dense, suburb is more walkable than the newer, more dense, suburb. Â How can this be? Â Ballwin was planned at a time when people thought nothing of getting in the car for every trip. Â The lady of the house had her own car now so she could drive the kids to school, do some shopping and get groceries on the way home. Kirkwood, on the other hand, was laid out long before the car. Â Being near the train station was important for reaching St. Louis.
Residential lots in Kirkwood are about the same size as those in Ballwin, the big difference is the Kirkwood lots are narrow & deep whereas the Ballwin lots are wide & shallow. Â Commercial districts are vastly different between the two. Â Kirkwood has too much newer auto-dependent retail but it also has a nice 19th century downtown.
Fortunately, Ballwin is not a lost cause.  It, and many other 2nd tier suburbs of the same era can be retrofitted to be more walkable.   The existing residential neighborhoods of single-family detached homes can remain unchanged, except for the addition of sidewalks internally and leading out to the commercial areas. Manchester Rd in Ballwin running through Kirkwood and into the City of St. Louis is an ideal corridor to be retrofitted. New structures can be built to infill the massive parking lots.  I can picture enhanced bus service or even a streetcar line the entire distance.
– Steve Patterson
You have to be careful about the whole walkscore thing–it only tells you the walkscore for a single point on the map and not the average for a city. When you look up the walk score for a city or town, walkscore gives you the score from the point that google maps tells it is the “city center”. In the case of Kirkwood, this is the walkable downtown. I'd say that most of Kirkwood and Ballwin have about the same walk score. My parents lived in an “average” Kirkwood home about a 1/2 mile from downtown and the walk score for their old house is only a 49.
…also, I like the streetcar down Manchester idea!
Agreed, but it shows that each of these two are vastly different at their centers. I think a 19th century walkable center helps for those living near it.
Walking down Manchester Road in Ballwin is miserable, regardless of what Walkscore says. I've done it, and there are so many curb cuts that you take your life into your hands every twenty feet.
24 years ago when we moved to St. Louis, our first home was in Glendale. There were no sidewalks on the street where we lived, but a decent neighborhood grocery store, churches, the post office, city hall, elementary school, and a bus to downtown were all within an easy walk. On more ambitious outings, downtown Kirkwood was not out of the question. Four years later we moved to the Central West End which is even more compact and far more amenities such as restaurants, movies, retail are available. I would say some of my neighbors in the city don't take advantage of the walkable neighborhood, still driving everywhere. And, this time of the year, far too many apparently don't notice that they should take a few minutes to shovel the snow off their walks. Far too much of the CWE is hazardous for walkers and bikers after a moderate snow.
The clearing of sidewalks after this last storm was pathetic–probably the worst job I've seen in years. It wasn't that much snow, people!
The other thing to take into account with the walkability score is that it does not take into account what the streets are like. Just what is near by and not how easy it is to get there. I live in Kirkwood just North of downtown. I disagree that the walkable downtown of Kirkwood is great. The problem is that most people do not walk to get to downtown. They drive and then walk around because Kirkwood has some of the worst sidewalks, or lack of sidewalks anywhere in St. Louis. None of the lights in downtown has pedestrian buttons and other than right in the middle of downtown there are very few lights or crosswalks to let you get across Kirkwood Rd. Once you leave the immediate downtown area and get into the neighborhoods it gets every worse. Kirkwood does not require or even maintain sidewalks. So the street I live on does not have any sidewalks and most of the streets around us have few or even incomplete sidewalks. We are five blocks from my kids schools and two are blocks we have to walk in the street and then cross a very busy street because the sidewalk is not finished so you have to cross in the middle of the street. If you run the Walkability score for my house it is a 43 but the only reason it is that high is I live close to Manchester Road and there is quite a few things within walking distance. The issue is that it should not even be that high because most of the places are on the North Side of Manchester and crossing Manchester is even worse then Kirkwood Rd and there are no sidewalks on Manchester Rd. and you would never walk to any of them. Go into Pi Pizza and talk to them about people having to run across Manchester Road and almost getting hit or people even getting hit to get to there restaurant.
Steve I would love to see you come to Kirkwood and give it a run thru on getting around in Kirkwood.
“New structures can be built to infill the massive parking lots. I can picture enhanced bus service or even a streetcar line the entire distance.” I'd like to get ahold of whatever you've been smokin'. Sure, they “can”. The real question is if and when.
Getting down to parsing words, there's a big difference between walkable and people walking. Most any area is walkable if you really want to walk. The Appalachian Trail is walkable. The new sidewalk along Lindbergh is walkable. Eager Road between Hanley and Brentwood is walkable. The Eads Bridge is walkable. There's a sidewalk between the two terminals at Lambert Airport.
There are also two kinds of walking, receational and destination-oriented. People walk every day to walk their dog and for exrecise (and that's a good thing). However, places like Kirkwood weren't laid out for that reason. Kirkwood and the other older urban centers were laid out and built out when many people didn't have their own motor vehicles. If you wanted to get around, you did walk, take the streetcar, use a bicycle or ride a horse. That's simply not the case today. We don't walk because we don't have to. Sure, there's a minority that chooses walking over driving, but the majority of us chooses to drive, even in these areas, to the municipal parking lots behind the main street businesses!
New Town St. Charles is walkable by design. I'm betting that every family out there has at least two motor vehicles. Do they walk more than a comparble family in the proverbial Glen Arbor Oaks subdivision down the road? Yes. Do they do it every day? Probably. Do they do it all day long? No way. Once they want to leave the reservation, they're no different than any other suburbanite – they drive to the 80,000 square foot grocery store with the surface parking lot, they drive to their jobs, and they line up in their minivans at their elementary schools.
You may think that density doesn't play a role. I disagree. I don't see a lot of positive coming from laying down random strips of concrete. The goal should be getting people out of their cars, not creating unused infrastructure. The goal should be preserving the countryside, not creating cartoon villages that are segregated from everything else in the county and the region. Density combined with mixed use IS the answer. It's the ONLY way that most people will be lured out of their cars . . .
Both Kirkwood and Ballwin will never be walkable for the vast majority of their residents. Yes, Kirkwood's downtown and an improved Manchester Ave. streetscape (which the Great Streets Initiative is trying to help) can make improvements at the margins, but most of the housing stock is too distant from these nodes to make them walkable in any meaningful way.
By the way, the population estimates show that both of these cities lost population over the course of the past decade.
You're right that the overall density of a city doesn't matter, but local density does.
Using your methodology (type in only the city name), density IS critical to achieving a top score. It's also a classic example of using statistics creatively and in a very flawed manner – the next-most walkable city in our area is probably Crestwood – yeah, right!:
New York City – 100/100
Pittsburgh, PA – 100/100
Boston, MA – 97/100
Miami Beach, FL – 97/100
San Francisco, CA – 97/100
Oakland, CA – 97/100
Denver, CO – 95/100
Los Angeles, CA – 94/100
Honolulu, HI – 89/100
Boulder, CO – 89/100
Crestwood – 85/100
Fergusson – 80/100
Maplewood – 75/100
Webster Groves – 71/100
Brentwood – 69/100
St. Louis – 63/100
Richmond Heights – 63/100
Fairview Heights, IL – 62/100
Rock Hill – 62/100
Des Peres – 62/100
Eureka – 58/100
Maryland Heights – 57/100
O'Fallon, IL – 57/100
St. Charles – 55/100
Wentzville – 55/100
Clayton – 54/100
O'Fallon, MO – 52/100
Chesterfield – 51/100
Arnold – 49/100
University City – 48/100
Frontenac – 46/100
East St. Louis, IL – 45/100
New Melle – 43/100
Shrewsbury – 42/100
Cottleville – 35/100
Sunset Hills – 34/100
Wellston – 32/100
Pagedale – 23/100
Creve Coeur – 22/100
Ladue – 15/100
Town & Country – 12/100
I don't believe the question purposed is whether density OR land use drive greater walkability. Obviously, they both play a role. The question, as I see it, is whether higher densities are required and result in walkable areas. And, as is pointed out in the post, higher density doesn't necessarily equal greater walkability, but walkable land use planning does. And, as you point out, high densities together with appropriate land use make for the most walkable of communities.
I also keep in mind (and everyone should) that the walkability score is a tool, but a dull one. Until a complete, updated online database exists of what the actual conditions on the ground are, I believe the walkability score should be read as red and blue on a car thermostat and not as a digital meat thermometer (one of the good ones). The score gives a good idea of where an address stands, but is not entirely accurate…yet.
proposed…good grief! =p
We basically agree. My whole point is is that making a place walkable is only one part of the equation, IF your goal is to encourage or to force people out of their SOV's. Building sidewalks alone does not make a place walkable, the distances need to be reasonable, the walk relatively pleasant and safe, and there needs to be a “there” there. It may also be a case of semantics – if I can walk, period, it's walkable; for others, walkability may entail the other components.
While some (very few) people will choose another form of mobility over the single-occupant vehicle, the vast majority of us choose to use our SOV's for the vast majority of our trips, even very short ones. It's quicker, drier, easier and tempered. Most of us are inherently lazy, even if we go to the gym every day. And how many teenagers do you know who are working their first job to pay for a bus pass? To pay for their first car, insurance, clothes, tattoos, junk food or music, yes. To pay for a bus pass, rarely. For the vast majority of them (and us), walking is way down the priority list . . . . .
No one is going to be forced out of their car to walk. (Though some are do feel forced into their cars to drive.)
I agree that a convergence of factors are required to make an area truly walkable. In fact, I believe the walkability score attempts to include all the factors you mention. North St. Louis has sidewalks and a city grid conducive to walkability, but has some of the lowest walkability scores in the entire metro area, because there is nowhere to walk to! And, absolutely, I agree that walkability means different things to different people.
But I keep getting the feeling that you're approaching the discussion as 'car vs. walk'. In my view, that's the wrong paradigm.
A few want to be 'car free.' I was car free in Hoboken, NJ for a year and a half and loved it. But that's because the subway trains went literally everywhere I needed to go. PATH and the NYC Subway was (still is) fast, frequent and fun (literally interacted with ppl I'd never meet otherwise). Car free is possible here in the Lou, but is obviously not nearly as convenient.
But there are many (and I believe the numbers are growing) who want to be in 'walkable' communities where there is a choice about whether or not to the car given the day, weather, their mood, etc. The goal should not be to make the St. Louis MSA or even St. Louis proper 98% walkable and allow (or force??) everyone to ditch their cars (thats beyond our lifetimes). But rather to make it possible, for those who choose on a given day, to walk.
I live on 17th and Pine and usually drive to the Culinaria (5 blocks?). As you rightly point out, its faster and more convenient. But just about every day I walk somewhere downtown. During the summer I'll walk to 40+ Cardinals games (an AWESOME perk for so many reasons), during hockey season I'll walk to 20 Blues games, on the weekends I'll walk to the Central Library (or will), two or three times a week I'll walk to Union Station for the food court, three days a week I'll run down to the Arch grounds/Eads Bridge for exercise, three or four times a week I'll walk to a Downtown bar for friends and drinks, at least once a week I'll walk somewhere for dinner, once a week I'll walk to my evening MBA classes………you get the point. I still drive to my store in Rock Hill, drive to my parents house in O'fallon, IL, drive to hang with my friends in Fenton or Soular or West County. But for so many things I have a choice. And that's the reason I purchased a home in a walkable neighborhood. (Though I still (unfortunately) put miles on my coupe.)
I would be careful assuming the walking priorities for the majority of us who are heavy auto users. It seems to me people increasingly want options. It is not a walk OR drive community we're after here, it's a choice.
When it comes to urban design, policy and priorities, I actually see three interrelated issues, not just two – walking, driving (usually a SOV) and public transit. And, as I said earlier, I see a big difference between the dynamics of recreational walking and those of destination-driven walking, and alternative transportation (cycling, scooters, skateboards, etc.) may be(come?) a fourth issue, as well.
You obviously “get it” and choose to both live and work in a highly-walkable area, yet you'd be the first to admit that you're a true minority, in both the city and the region. I also don't think that I'm making a car versus walking argument, but I am trying to clarify just what “walkable” really means. At it's most basic, walking is just about putting feet on the ground, and a rudimentary trail will suffice, assuming no major physical obstacles. At the other end of the spectrum, a conscious, comprehensive built environment is actually able to attract pedestrians who have other options.
You also state that “No one is going to be forced out of their car to walk.” I disagree. It's that kind of thinking that will just see a continued shift to the autocentric, suburban building model. There probably won't be any out and out prohibitions, but I can see things like less available parking, higher costs to park, higher costs to drive, and less new construction/added capacity and increasing congestion all making both walking and transit more attractive by encouraging higher densities.
Is there one single answer? No. It's a complex topic with multiple options and multiple outcomes. But it won't change until the questions are asked and alternatives proposed and implemented . . .
“You also state that 'No one is going to be forced out of their car to walk.' I disagree. It's that kind of thinking that will just see a continued shift to the autocentric, suburban building model.”
I literally have no idea what youre trying to say here. =D Please clarify…
“You obviously 'get it' and choose to both live and work in a highly-walkable area, yet you'd be the first to admit that you're a true minority, in both the city and the region.”
?? I drive to work. And yes, most people in the region do not live in a neighborhood with a walkability score of 92. This point has been well established already. Is your point that most people, therefore, don't want to live in a more walkable community? I'm not sure that's true.
'Walkability' in an urban areas would likely mean, as you say, “a conscious, comprehensive built environment [that] is actually able to attract pedestrians who have other options.” And don't see how this necessarily results in “less available parking, higher costs to park, higher costs to drive, etc.” As far as I can see, the walkable renaissance of Downtown, Midtown Alley and Grand Center has, so far, resulted in even more parking options not less (unfortunately).
Again, I'm not sure I see the goal of walkable communities being a house without two cars out front. But rather cars that are used less often.
Simply saying that a walkable community is not worth the effort because everyone already drives everywhere is silly. Thats like saying its not worth building a stadium for a new football team because everyone goes to watch hockey. No one goes because there's no stadium! It's clear people are prefering more and more walkable communities. Given more sustainable, walkable environments, many will walk. If my parents in the Metro East are begining to bike/walk to their grocery store and church (not routinely, but often enough), then there is a demand that has yet to meet a solution in our region.
One big challenge is defining “walkable”. At its most basic, most parts of the world are walkable if you really want to walk, the exceptions being things like busy freeways, long railroad bridges, strip mines, land fills, etc. – things with sharp drop-offs and/or that will get you killed quickly. In pretty much every other environment, if you're careful, you can walk. It may not be pleasant, and it may not be sompleace you'd let a kid walk alone, but it's “walkable”. This differs from my perception of “walkable” in the urban design context and is what I'm trying to articulate (apparently not very successfully) in response to this post's headline, content and comments.
Many of the precepts of Transit Oriented Development (TOD) define the ideal walkable urban environment – the 1/4 and 1/2 mile concentric circles and the high-density and a mixed-use core served by a viable public transit system. This is very different from a suburban monoculture of single-family lots with 200' frontages and no services within a mile – adding sidewalks does little to make the area more walkable, as in people doing more than walking their dog. 99% of the people will get in their cars to go anywhere, and the sidewalk is just decoration, much like the brick-wrapped mailbox.
I guess I'm both more of an idealist and more of a pragmatist. What you did in NJ is hardcore walkable and what I'd like to see a lot more of happening around here – if you have a car, it only gets used on the weekend. But what I'm seeing happening around here, if walkable is attempted at all, is much more likely to be walkable light and/or walkable cute – New Town St. Charles in the middle of a cornfield, downtown Kirkwood with free municipal parking lots behind the old storefronts, The Boulevard in Brentwood, etc, etc. – you get a few blocks of nice, walkable space, but you gotta drive there to walk! (If Kirkwood were truly walkable, many employees would walk to work – I'd bet that 99% of everyone working in Kirkwood drives to their jobs every day, so the city ain't “walkable”/actually walking!) And this is all happening because it's what's “normal”, it's what sells, and because the vast majority of our lives revolve around using our SOV's as the primary means of transportation!
In my perfect world, there would be a lot more transit use, especially for daily commutes, and a lot less reliance on the SOV, again, for those repetetive, daily trips. We have 8-lane freeways not because we need them 24/7, we have them because we “need” 4 lanes, one way, an hour a day, 5 days a week! That's the “forcing” issue – yeah, we have congestion and people whine about it, but it's not painful enough to force them to look at transit as a viable alternative. If we didn't keep building new highways and bridges, and invested, instead, in building a comprehensive transit system with frequent service, then people might be more inclined to pull off I-64 at Ballas and switch to a bus or a train to complete their trip to downtown or Clayton. And with the exception of a few employers like BJC and Wash U., who get it, all subsidies now seem to go into providing parking for employees, like you, instead of providing the free or low-cost transit passes that actually change behavior and increase demand. And once somene is trapped at work over their lunch hour without their car, then they're forced to walk around, instead of driving, if they want to get out of the office, again increasing demand for pedestrian-oriented services that define and inprove the ideal urban environment.
So, back at you – “It's clear people are prefering more and more walkable communities.” I see little real evidence of that around here. Our surrounding counties are filled with the crappy subdivisions we love to deride, yet they continue to sell, both as new construction and as resales, while denser, older, urban, walkable neighborhoods in the city continue to languish or decline. I'm not seeing a huge groudswell in demand, marked by increasing prices, scrape-offs replaced by larger units, multiple new in-fill developments selling out before the first model is built or transit-oriented development around many, if any, of our transit centers. Sure, there are a few/some people making the switch, but I'm seeing a lot more supporting the non-walkable suburban model by buying into it, both literally and figuratively!
Upon further review, it's still a matter of semantics, but the short answer is that sidewalks do make for more-walkable streets, but that they do not necessarily make for truly walkable neighborhoods. That requires mixed uses in a compact area, aka “density”, and most likely, a viable transit system.
I'm not going to get hung up on the subtle nuances of varying levels of walkability. Clearly, there are different levels of walkability and varying conditions for greater and lesser walkability. Its been established that several factors go into defining what walkability is. A home near the entrance of a subdivision with sidewalks that connect to retail around the corner is more walkable than a home situated a half-mile into the development.
“Walkable” will always be variable. The criteria used for measuring “walkability” are fairly easy to figure out IMO.
Here is Walk Score's method:
The Walk Score algorithm awards points based on the distance to amenities in each category. If an amenity is within .25 miles (or .4 km), we assign the maximum number of points. The number of points declines as the distance approaches 1 mile (or 1.6 km)—no points are awarded for amenities further than 1 mile. The points are summed and normalized to yield a score from 0—100. The number of nearby amenities is the leading predictor of whether people walk.
Walk Score self-admits problems with its method:
A number of factors that contribute to walkability that are not part of our algorithm:
•Street width and block length: Narrow streets slow down traffic. Short blocks provide more routes to the same destination and make it easier to take a direct route.
•Street design: Sidewalks and safe crossings are essential to walkability. Appropriate automobile speeds, trees, and other features also help.
•Safety from crime and crashes: How much crime is in the neighborhood? How many traffic accidents are there? Are streets well-lit?
•Pedestrian-friendly community design: Are buildings close to the sidewalk with parking in back? Are destinations clustered together?
•Topography: Hills can make walking difficult, especially if you're carrying groceries.
•Freeways and bodies of water: Freeways can divide neighborhoods. Swimming is harder than walking.
•Weather: In some places it's just too hot or cold to walk regularly.
These factors are currently not easily distilled into meaningful data sets that can be plugged into algorithms. Walk Score defends their results by pointing out that the concentration of nearby amenities is the predictor of whether people walk.
As far as a movement towards the desirability of walkable communities, I think you need to look no further than the New Town development you mentioned. Why would developers spend a dime on creating at least the perception of a more walkable community (however flawed) if no one really cares to walk anyway? Or why would the NorthSide project even see the light of tax credits and private financing if demand so clearly wants new construction even further east and west from the region's center? These examples obviously don't prove a thing, but are two examples that beg the question.
Bottom line, decades long trends have halted in their tracks. And not only stopped in their tracks, but have reversed. People stopped leaving the City (in remarkable DROVES) and actually moved back in (obviously, not in the same numbers as they left). Given the dramatic residential shifts in the St Louis MSA during the 60's through the 90's, something clearly changed during the 00's. I'm willing to bet that part of teh explaination for the most recent decade are the benefits of a denser urban area which include great walkability.
*Walk Score defends their results by pointing out that the concentration of nearby amenities is the LEADING predictor of whether people walk.
I also don't understand how one can be more pragmatic and more ideallistic at the same time. =p More of one usually leads to less of the other, IMO.
My hot button was pushed by the title, “The Density Needed For Walkability Myth”, and pushed again by the assertion that “Ballwin is not a lost cause. . . . The existing residential neighborhoods of single-family detached homes can remain unchanged, except for the addition of sidewalks internally and leading out to the commercial areas.”
Building sidewaks in low density areas does little, if anything, to get people out of their cars. Neighbors already walk across their yards and down the sides of their low-traffic suburban streets – it works, for them (and I grew up in several of them – I know)! Sidewalks are mostly eyewash because all the other components / “amenities” are absent! “The number of nearby amenities is the leading predictor of whether people walk.” Putting lipstick on a pig doesn't make it much prettier.
Steve is currently on a Quixotic campaign to convince suburbanites that they need sidewalks. It's his thing, not mine – I'd rather see more effort and money invested in improving the pedestrian experience in denser urban areas – more bang for the buck and much more likelihood of real success.
Interesting comments, I went from a somewhat walkable shrewsbury to a not so walkable Lafayette, CA (Bay area). A big difference is sidewalks and controlled crosswalks on busy streets/interchanges in my opinion. In Shrewsbury you had sidewalks on the main streets and some decent crosswalks. While I didn't have a sidewalk on my side street it was a matter of a short walk before getting some refuge and some assurance that my daughter had a fair shake at getting to the pool without getting run over or you could get to parks, Shrewsbury center, the local gas station, and even Old Webster.
In lafayette, sidewalks let alone street lighting doesn't happen til you get downtown for the most part. Lafayette like Kirkwood has an older well established downtown with a mix of old mom & pop and new mingled. However, like Kirkwood, I believe most people by and far drive and then walk.
I think the way you calculated density is throwing things off. Kirkwood is definitely more dense than Ballwin if you only use inhabitable land as the denominator. Kirkwood has hundreds of acres of industrial and park land near I-270 and the Meramec river that inflate the total square mileage but have no people living on them.
Municipal boundaries are administrative and really don't do a good job of describing most ways people move and interact within space. Like St. Louis City's inflated crime ranking, this is an example of what geographer's would call the “boundary problem.”
Ballwin may also be more dense, statistically, because it has more condos and apartments, per capita, while Kirkwood may also be less dense because it has more parkland than Ballwin. But, again, it's all about picking and massaging the numbers to prove a predetermined point. To say that Crestwood, with a walk score of 85, is truly more walkable than Clayton, with a 54, or U. City, with a 48, sums it up in a nutshell . . .
This really doesn't discount the overall point, however. Even if Ballwin were less dense than Kirkwood, these two commuities are similar enough in size for the purposes of the discussion about the differences in walkability. (And given JZ's point about the abundance of large rental complexes in Ballwin, Ballwin may likely be more dense.)
related: http://www.post-gazette.com/pg…