Understanding gridlock
Want to know why there’s so much traffic congestion? I had to explain it recently and have the resulting (non-academic) essay sitting here on my computer, so online it goes. It’s fairly long, though, especially since I’m including half a chapter from a planning text as supplement, so it’s going below the fold where it won’t take up as much space.
Here’s a clue to get things rolling: Density has absolutely nothing to do with traffic congestion.
The common misperception is of course that it does. That density and “overdevelopment” are to blame for our woes, and if only greedy developers wouldn’t build so many houses and the government get its act together on building more new roads, the problem would be solved.
That is a myth. It is completely and utterly untrue. It is so overwhelmingly erroneous that most people simply refuse to believe it. People don’t like to be *that* wrong.
But they are. I-66, my local highway, is a perfect example. I live in a neighborhood filled with 20-plus story buildings with no setbacks from the street. It is tremendously dense, as are the adjacent neighborhoods east of mine. To accommodate all this density, I-66 has only two travel lanes in each direction, yet it is not particularly congested. If one travels a few miles west, however (that’s *away* from downtown, for the record), the situation is markedly different. I-66 expands to eight lanes and is surrounded by single family detached homes, strip malls and occasional office parks instead of highly urban neighborhoods, yet it is tremendously congested. Low-density, eight lane I-66 carries 182,000 cars per day at a crawling pace, while high-density, four lane, closer to downtown I-66 carries just 71,000, most of which move along briskly. Metro runs the length of the corridor.
There are many other examples. Atlanta is one of the most sparsely populated metropolitan areas in America and has more freeway miles per capita than any other comparably-sized city, yet it is one of the most horribly congested regions in the country.
Clearly, something besides capacity and density is causing all this congestion.
So OK. What is it, already?
Traffic congestion is caused primarily by two factors.
- A building arrangement wherein people must drive many miles for all their needs. If, like most Americans, you live in a suburban neighborhood, odds are none of the errands you need to run each day are located anywhere near your home. Three miles to the grocery store doesn’t sound like very far. After all, it’s just a 6 minute car ride at 30 miles per hour. But three miles is far enough that the only way to get there is to drive. The average American family generates 11 car trips per day. That’s a lot of driving. Spreading things out further just makes it worse.
The arterial-collector-local road network that forces all traffic onto a tiny handful of key roads. Take a look at the graphic to the right. First of all, the community on the bottom is much more walkable and transit-friendly than the community on top, but for the sake of argument let’s ignore that and assume that everyone in both communities will drive for all trips. In the top community, anyone traveling from their home to work, a store, or any place that isn’t in their immediate residential “pod” must first drive on to the shared arterial highway. All traffic, regardless of where it is going or coming from, uses that same single road to get there. That road inevitably becomes hopelessly congested. Conversely, the community on the bottom is laid out in a manner that provides many different routes for any given trip. Different people could travel between the same destinations any number of different ways, and there is no need for anyone who isn’t traveling outside the area altogether to get on the arterial highway even once, if they don’t want to. No single road gets too busy because every street in the neighborhood is created equal. This seems obvious, but it’s counterintuitive to how most people think. Neighborhood associations all over the country fight tooth and nail against through traffic in residential neighborhoods because they are afraid of congestion close to home. City hall usually complies, because the sort of traffic calming neighborhood associations fight for is cheap and provides a good example of “responsive government”, but in pushing all traffic outside the neighborhoods and onto one or two key arteries (with occasional, less politically connected neighborhood streets acting as mini-arteries), those handful of roads get busier and busier.
Below is an excerpt from Suburban Nation by Andres Duany, Elizabeth Plater-Zyberk and Jeff Speck that further explores the issue. Suburban Nation is arguably the second most influential book in the world of contemporary American city-planning, and well worth the read for anyone interested in these topics.
The simple truth is that building more highways and widening existing roads, almost always motivated by concern over traffic, does nothing to reduce traffic. In the long run, it actually increases traffic. The revelation is so counterintuitive that it bears repeating: adding lanes makes traffic worse. This paradox was suspected as early as 1942 by Robert Moses, who noticed that the highways he had built around New York City in 1939 were somehow generating greater traffic problems than had existed previously. Since then, the phenomenon has been well documented, most notably in 1989, when the Southern California Association of Governments concluded that traffic-assistance measures, be they adding lanes, or even double-decking the roadways, would have no more than a cosmetic effect on Los Angeles’ traffic problems. The best it could offer was to tell people to work closer to home, which is precisely what highway building mitigates against.
Across the Atlantic, the British government reached a similar conclusion. Its studies showed that increased traffic capacity causes people to drive more – a lot more – such that half of any driving timesavings generated by new roadways are lost in the short run. In the Long run, potentially all savings are expected to be lost. In the words of the Transport Minister, “The fact of the matter is that we cannot tackle our traffic problems by buildings more roads.” While the British has responded to this discovery by drastically cutting their road-building budgets, no such thing can be said about Americans.
There is no shortage of hard data. A recent University of California at Berkeley study covering thirty California counties between 1973 and 1990 found that, for every 10 percent increase in roadway capacity, traffic increased 9 percent within four year’s time. For anecdotal evidence, one need only look at commuting patterns in those cities with expansive new highway systems: USA Today published the following report on Atlanta: “For years, Atlanta tried to ward off traffic problems by building more highways per capita than any other urban area except Kansas City… As a result of the area’s sprawl, Atlantans now derive an average of 35 miles a day, more than residents of any other city.” This phenomenon, which is now well known to those members of the transportation industry who wish to acknowledge it, has come to be called induced traffic.
The mechanism at work behind induced traffic is elegantly explained by an aphorism gaining popularity among traffic engineers: “Trying to cure traffic congestion by adding more capacity is like trying to cure obesity by loosening your belt.” Increased traffic capacity makes longer commutes less burdensome, and as a result, people are willing to live farther and farther from their workplace. As increasing numbers of people make similar decisions, the long-distance commute grows as crowded as the inner city, commuters clamor for additional lanes, and the cycle repeats itself. This problem is compounded by the hierarchical organization of the new roadways, which concentrate through traffic on as few streets as possible.
The phenomenon of induced traffic works in reverse as well. When New York’s West Side Highway collapsed in 1973, and NYDOT study showed that 93 percent of the car trips lost did not reappear elsewhere; people simply stopped driving. A similar result accompanied the destruction of San Francisco’s Embarcadero Freeway in the 1989 earthquake. Citizens voted to remove the freeway entirely despite the apocalyptic warnings of traffic engineers. Surprisingly, a recent British study found that downtown road removals tend to boost local economies, while new roads lead to higher urban unemployment. So much for road-building as a way to spur the economy.
If traffic is to be discussed responsibly, it must first be made clear that the level of traffic which drivers experience daily, and which they bemoan so vehemently, is only as high as they are willing to countenance. If it were not, they would adjust their behavior and move, carpool, take transit, or just stay at home, as some choose to do. How crowded a roadway is at any given moment represents a condition of equilibrium between people’s desire to drive and their reluctance to fight traffic. Because people are willing to suffer inordinately in traffic before seeking alternatives – other than clamoring for more highways – the state of equilibrium of all busy roads is to be stop-and-go traffic. The question is not how many lanes must be built to ease congestion, but how many lanes of congestion you want. Do you favor four lanes of bumper-to-bumper traffic at rush hour, or sixteen?
This condition is best explained by what specialists call latent demand. Since the real constraint on driving is traffic, not cost, people are always ready to make more trips when the traffic goes away. The number of latent trips is huge – perhaps 30 percent of existing traffic. Because of latent demand, adding lanes is futile, since drivers are already poised to use them up.
While the befuddling fact of induced traffic is well understood by sophisticated traffic engineers, it might as well be a secret, so poorly has it been disseminated. The computer models that transportation consultants use do not even consider it, and most local public works directors have never heard of it at all. As a result, from Maine to Hawaii, city, county and even state engineering departments continue to build more roadways in anticipation of increased traffic, and, in so doing, create that traffic. The most irksome aspect of this situation is that these road-builders are never proved wrong; in fact, they are always proved right: “You see,” they say, “I told you that traffic was coming.”
The ramifications are quite unsettling. Almost all of the billions of dollars spent on road-building over the past decades have accomplished only one thing, which is to increase the amount of time that we must spend in our cars each day. Americans now drive twice as many miles per year as they did just twenty years ago. Since 1969, the number of miles cars travel has grown at four times the population rate. And we’re just getting started: federal highway officials predict that over the next twenty years congestion will quadruple. Still, every congressman, it seems, wants a new highway to his credit.
Tags: urbanism
posted by Cirrus of Malla @ 12:10 PM ⢠Permalink
19 comments
19 Comments:
This is total and utter horse manure. Despite what you say, traffic congestion is always associated with density, but not necessarily housing density although that plays a part. It is job density that drivves many people to attemt to go to the same place at the same time, and that causes congestion.
For starters 66 inside the beltway is tremendously crowded at certain times. When it is not crowded it is because it is restricted to HOV and the crowds have been shifted to Rte 50 and GW Parkway. 66 outside the beltway is jammed up because either it changes to two lanes or is diverted at the beltway or it changes to fewer lanes as you go out, down to two lanes each direction at Gainesville.
Coming from 66 West there are back-ups at Gainesville and Manassas as a result of the uncertainty of merging, but once the merges are completed it speeds up. 66 is a lousy example to make your point.
With respect to pods, what you say makes some sense, but remember, we moved to that system specifically to get away from problems asscoiated with grid layouts. While it is true there are multiple routes, there are also more intersections and intersections cause delay and accidents. It is not entirely a one way argument.
The Suburban Nation article misses the point entirely. It is not the point of building more highways to reduce traffic. The argument about induced traffic has been endlessly debated, but the current scientific position is that induced traffic does exist, but that effect is much smaller than previously supposed. Your data showing that half to 90 % of new capacity is consumed by new traffic has been superceded by later studies, although the argument still rages.
As I understand it, the current position is that indeed, new roads fill up, but it depends on circumstances. In some cases it means that other nearby and less favored roads are less congested, in others it simply means that latent, pre-existing demand was finally met with new construction. In any case, the new construction allows more traffic to flow, even if it is still under congested conditions. The question is not how many lanes of congestion do you want, it is how important is it to move X number of people to the same(more or less) place. That question is the same, whether you consider roads, subways, commuter rail, buses or any other method.
The Texas transportation studies show how difficult it is to relieve congestion through road construction, but they clearly show that less construction leads to even greater congestion. Two cities that lost significant numbers of jobs also "enjoyed" less congestion.
Why should we consider that people who live far from where they work have made a bad decision any more than we should consider that employers who locate where they will cause congestion have made a bad decision? Our transportation system subsideses employers as much or more than it does employees.
What we need to do is to ask, where are all these people going? If they are going to work and conducting business, we cannot expect to reduce road usage without sereiosly reducing commerce and our well being. Your argument about roadbuilding and the economy flies in the face of facts. People don't drive around for no reason, and the usual reason is to conduct some form of business. In London, after central driving area tolls came into play, BMW made plans to move out of town.
It is true that some people in suburban areas drive farther, but the time they spend in traveling is actually less than urban folks spend, and far less than those that use transit. Many studies have shown that rural people spend far less on travel than urban residents, even though they drive farther.
I'm sorry, the arguments you make have been argued widely and they are frequently repeated, but they are out of date, wrong, and offer no real alternative. As such the arguments are unacceptable.
Job density versus housing density: That is a good point, and I should have been more careful to make a distinction between the two. The point of my bringing up density repeatedly is that it is frequently used by NIMBYs fighting against new housing proposals near their neighborhoods. Pay attention to any public discussion in the Washington area where local officials are debating a substantial new housing development and I guarantee one of the prime issues is activists contending the residential density will cause traffic congestion.
Regarding I-66: The busiest place on the highway is between the Beltway and Chain Bridge Road. Your arguments regarding lanes being dropped do not apply there. Likewise, the sections of I-66 inside the Beltway that are congested are *west* of Ballton (where density picks up), run in the off-peak direction where HOV restrictions do not apply, and are used most heavily by people commuting to Tysons Corner, which has woefully inadequate transit access and a horribly pedestrian-hostile, auto-centric layout. Those stretches of road support my position more than they do yours. Speaking of the HOV restrictions, your points regarding them are likewise immaterial. Route 50 handles more traffic outside the Beltway in Fairfax than it does in Arlington. 80k/day near Fairfax County Parkway versus 56k/day east of Washington Blvd. Numbers for GW Parkway aren't handy, but I would find it very, *very* hard to believe that they would be high enough to account for the difference, especially considering much of the traffic on GW Pkwy originates in MD and wouldn't be using 66 in any event. Speaking of Maryland, without a single real freeway running into the District and (I believe but have not checked) more commuters into the city, it manages to be less congested than VA. That Maryland relies less on roads is not a coincidence.
Grids: We moved away from grids because we *thought* an arterial system would be safer. It turns out people drive as quickly and dangerously as they are comfortable driving, and they drive much more dangerously on arterial highways than neighborhood streets. Although it is counterintuitive, mixed streets with more people walking are statistically much *safer* than arterials designed only to move cars as quickly as possible. I would also challenge your assertion that the artery system is faster for internal, day-to-day trips. The person going to the grocery story in the top community has to sit at stop lights, at least one of which is likely to be long, then crawl along in the arterial street unless it's an off hour. The same person in the bottom community may have to stop at a sign, but only for a second.
"The question is not how many lanes of congestion do you want, it is how important is it to move X number of people to the same(more or less) place. That question is the same, whether you consider roads, subways, commuter rail, buses or any other method." I would agree with that, with the stipulations that a) it be understood that adding lanes will not decrease congestion, only increase capacity to permit the same level of congestion; and b) that subways, commuter rail, buses, biking, walking and every other mode are more efficient than congested freeways, particularly since the added lanes are only going to be necessary at the busiest time of day. If the point is to move X number of people to the same place, then a congested highway is not the best way to do it.
" Why should we consider that people who live far from where they work have made a bad decision any more than we should consider that employers who locate where they will cause congestion have made a bad decision? Our transportation system subsideses employers as much or more than it does employees." Agreed. Both are responsible. That doesn't mean responsibility should be forgiven to either. The fact that employers are wrong does not excuse homeowners, or vice versa. I will add here that the list of direct or indirect subsidies at every level of government that have gone into producing our current situation is very, very long, and that "this is what the free market wants" is an even bigger myth than "residential density causes traffic congestion". That, however, is another discussion for another day.
"What we need to do is to ask, where are all these people going? If they are going to work and conducting business, we cannot expect to reduce road usage without sereiosly reducing commerce and our well being." That is utterly incorrect. By choice I do not have a car. I live in a walkable neighborhood near Metro, even though my job is in a suburban part of Fairfax. I am a productive, white collar, fully-contributing member of society. I would not suggest that my lifestyle is for everyone, but it is absolutely possible to reduce road usage by building more communities that are less dependent on automobiles for all daily needs, and more suitable for transit-based commutes and walking. Perhaps not in absolute terms as the economy grows, but certainly, unquestionably, in terms of mode-share. Traffic congestion has not increased significantly on Wilson Boulevard in 30 years, despite massive increases in both residential and job density, because that development has been well-designed. Here there is no middle ground or room for debate. You are simply wrong.
"It is true that some people in suburban areas drive farther, but the time they spend in traveling is actually less than urban folks spend, and far less than those that use transit. Many studies have shown that rural people spend far less on travel than urban residents, even though they drive farther." Those are disingenuous arguments and you know it. First of all, it depends where you are going. Many transit trips take much longer than they should because transit is inherently incapable of serving spread-out population/jobs well. When you design your cities primarily around cars, of course cars are fastest. Secondly, transit is woefully underfunded compared to roads. Give it the same money and it will compare more favorably. Furthermore, travel time is not the only issue. I can nap, read and work on my laptop while on the Metro. Drivers must drive. There are air quality issues. There are cost issues. There are safety issues. Shall I go on? I will grant that the situation is drastically different in rural areas. Clearly there is a point somewhere between the truly rural and suburban where density does make a difference, but you and I both know that is not what we’re talking about here. Unless you suggest metropolitan areas be abandoned completely and our entire population be fit (and forced, since not everyone will want to go) into rural areas, the countryside is immaterial to this argument.
Offering alternatives: Excuse me? I absolutely offered solutions. They may not have been solutions *you like* or solutions that don’t require someone to not live in Delaplane, VA and commute to the city if they don’t want to be part of the problem, but I absolutely offered solutions. Your rhetoric - throughout your entire reply - does not hold up to the facts.
Re: The question is not how many lanes of congestion do you want, it is how important is it to move X number of people to the same(more or less) place. That question is the same, whether you consider roads, subways, commuter rail, buses or any other method."
The only thing that I thought would have strengthened your post is some data on throughput.
Jeff Tunlin of Nelson-Nygaard participated in DC DDOT's "Great Streets" conference in January. His presentation stated that in one hour one lane of road (or equivalent) can move 900 automobiles, 6,250 people as passengers on regular bus service, 10,000 people via bus rapid transit, and 16,000 people via light rail.
However, I have not seen (nor have I asked for it) the data behind this slide.
This is the flip side of your argument. It is impossible, even discounting induced traffic and latent demand, to move low-density people to all the things they want to see and do and buy via individually driven cars. It just can't be done.
Also, there is a website about Virginia issues called Bacon's Rebellion. Their land use commentator is a guy named Ed Risse and he has some good writings about this issue as well--he calls it the "private-vehicle mobility myth."
E.g., he writes:
For those who just came in, here is a refresher on the Private-Vehicle Mobility Myth:
Regardless of where they live, work, seek services and participate in leisure activities, citizens believe that it is physically possible for the government to build a roadway system that allows them to drive wherever they want to, whenever they want to go there and arrive in a timely and safe manor.
The Private-Vehicle Mobility Myth helps parents convince themselves that the house with the “big yard” may be a long way from where the jobs, services, recreation and amenities are now, but that will change. Politicians reinforce the myth by continuing to promise that “soon” they will improve the roads and the big yard owners will be able to get to wherever quickly.
This is kind of the behavior explanation of latent and induced demand.
If "Suburban Nation is arguably the second most influential book in the world of contemporary American city-planning," what is the most influentilal book?
Ken
Death and Life of American Cities by Jane Jacobs, of course. Its what got me into this subject. I'm gonna have to read Suburban Nation now...it sounds really on point.
I coudn't agree with you more, Dan. NoVa is built for bad traffic. Destinations are so far apart that, you are compelled to drive everywhere you go. A friend of mine in Tysons once told me that "Everything in Northern Virginia takes 1/2 hour to get to". I have found it to be a reasonably accurate benchmark. Building extra lanes will not accomplish anything but awaken latent demand and pave a little more over the green.
I moved into the District last summer, and gave up driving completely. Just like that. Poof. Gone.
I had a brand new zippy little GTI, that I did not want to give up. But I bit the bullet and took the leap. I couldn't be happier. NOTHING in DC takes me 1/2 hour to get to; most of my destinations are only only 10-15 minutes away by foot and metro. I can get from my front door to Ballston or Bethesda in 15 minutes. Try that in traffic with a car! The only time I ever take longer, is when I WANT to take my time, which incidentally I have a lot more of since I gave up my car.
My job is 10 minutes walk from my apartment. When I tell people this, they tell me how lucky I am, but luck has nothing to with it. I moved where I moved for the same reason that I gave up my car: to eliminate completely unnessary stress from my life, save time, and do my part to make a difference.
These days, as I walk to work down Mass Ave, I watch the aggressive drivers cutting each other off, gesturing, hopelessly trying not to be late, I just laugh and shake my head. I am free.
Dan:
I agree with your first paragraph. Here is a question. There are many who claim that increasing residential density is the answer to traffic congestion, yet NIMBY's consistently take the opposite view. And it matters not whether they are in sparsely populated Fauquier, rapidly growing Loudoun, or densely populated areas of Fairfax, Arlington, or the District. Why is it that the popular perception is so different from that proposed by so called experts?
Regarding youe second paragraph, I don't understand your statements, but what I do read, I see differently. I see westbound traffic dropping off after the exit for Dulles. Is that the same as traffic to Tysons? Isn't that exit for Dulles traffic only? I'm confused.
What you say about traffic outside the beltway on 50 makes sense, and it is consistent with your argument about cul-de-sacs. I still claim that people who would otherwise drive Route 66 straight through are diverted somewhere because of the lane restrictions.
My view is that this causes massive backups on 66 eastbound as people divert either east or west to alternate routes. Those routes are not accessible from 66 west of the beltway, so there is no real alternative. You can pick up rout 50 at Nutley, and many people take that route. Maybe our views are different because our desitnation and start are different.
I used to keep my baot in a marina that had a strage tradition. Each year ever power boat operator would take a sailor out, and vice versa. Maybe we should do the same.
No, I'm pretty sure Maryland has far fewer commuters into the District than VA.
I didn't make any assertion that the artery system, per se, is faster. I have no real experience with it. I lived in a grid neighborhood in Alexandria and now live in a rural area. Before Alexandria, I lived in Cleveland Park, in the District. I can tell you unequivocally, that my trip to the grocery store now (6 miles) takes less time than it did in Alexandria (2 miles). I also have accurate data on my auto use and costs dating back many years. Cars themselves are more expensive now (and better) but my mileage and time adjusted costs are the same or lower than before.
This is born out by many studies. Suburban residents travel more, but spend less time in transit. Rural residents spend less on travel than urban residents. (They may not feel the pressure to drive a Beemer, and their theft risk is lower.) The real difference is th time spent idling, and the time spent at optimum economic speed.
However, you are correct in what you say about safety, at least in several respects. A pedestrian struck in the country is pretty much road kill. In the city, most elderly residents who are killed, are struck in a crosswalk. Lower speeds in the city lead to lower death rates, and yet insurance costs are higher. The jury is still out.
With regard to the "efficiency of Metro" etc. Not true. Completely not true. (bikes excepted, I like bikes and I agree they are the best). The cost of meeting peak load for transit systems of all types is higher than the cost of meeting peak load for roads. This is not generally accepted, but it is true.
I would refer you to a Brookings Report by Winston and Shirley, entitled "Alternate Route, Toward Efficient Urban Transportation". They offer a compelling argument against your frequently accepted assertion.
It turns out that the load capcity for autos is mostly higher than for transit systems. (The possible load factor for transit systems is higher, but they never meet it so their real efficiency is lower.)
Despite the low efficiency of individual combustion engines, the overall cost of autos is only a couple of cents per mile more than standing room only on Metro, all things considered. And that is charging autos for ALL of their external costs. Given the other attributes of autos, flexibility, freight, convenience, speed, comfort, etc. etc. cars win hands down.
Adding more Metro will also not decrease congestion, only increase the capacity to permit the same level of congestion. So the difference comes down to how much the added capacity will cost. Added capacity for Metro is also only needed at the busiest time of day. Unlike cars, Metro has a huge labor cost, even at night when the trains are not running. (That is when the track inspectors work.) And the Metro cars have to be carried back out at rush hour, mostly empty (cars never travel empty, but they chew up praking spaces. Metro uses parking spaces, too.) The starting efficiency of trains is only 50%, and goes down rapidly from there. The TOTAL picture for Metro, and other transit is not good.
"In 2002, on a typical weekday morning, more than 52,000 people in Virginia used the HOV lanes to the District and Arlington, while Metrorail carried 41,300 passengers and Virginia Railway Express carried 4,310 to the same destinations. On Shirley Highway in 2003, the two HOV lanes moved 29,400 people in 7,900 vehicles while the four regular lanes moved 21,300 people in 18,450 vehicles."
http://www.virginiadot.org/infoservice/news/newsrelease.asp?ID=NOVA-NR05-02
So, Shirley highway alone carried more people than all of Metro in 2002. And that is just the rush hour. If you look at the 24 hour picture Metro looks pretty grim, especially at 3 AM. Your arguments against highways apply doubly to rail.
Buses aren't much better. The best system is one we don't have: jitneys. Jitneys are across between taxis and buses, and they are outlawed here by virtue of the taxi drivers union. They work very well in the Carribbean and in Norway.
That, however, is another discussion for another day. . I don't think so. It is the primary discussion that must be conducted in order to solve this problem. I don't like subsidies either, but we have gotten way out of hand in accusing every other party in getting one.
Consider Metro. To a large extent it has not replaced auto use so much as it has changed the location to which suburbanites drive. For what Metro costs we could build office buildings where the Metro stops are AND GIVE THE OFFICE SPACE AWAY, FREE. The net result would be the same.
I don't know if you are fortunate you do not have a car, or deprived because you do not have one. What do you do when you want to go to the boat in Annapolis, or take a sudden ski trip to Lake Placid? I appreciate your situation. When I moved to Alexandria, I deliberately moved to within a mile of my job. It was a luxury I enjoyed for nine years, before my company moved to the suburbs, and mad me a reverse commuter. Even so, I had to drive to work. I needed the car as part of my job, and the road from my house to my office was too narrow, crooked, and steep for easy bike or walking.
Even so, I built a 30 ft. sailboat with the time most people spend commuting. I doubt very seriously if i could have done that in your Arlington neighborhood. (My brother lives in Ballston, so I understand.)
Your situation fits you, but it only fits a very small percentage of others. The reduction in auto use we can reasonably achieve is also accordingly small. Your argument does not address at all the close correlation between road usage and economic growth. Even if you don't drive, you depend on roads for nearly everything that is delivered to you.
I will guarantee that you have your AC on on days when I have the windows open. Urban areas are huge energy sinks, even if everybody walks.
I agree with what you say about Wilson Blvd. But the idea that it can be universally extended is simply wrong. I own and operate a large farm, and I commute a fair distance to work. My tractors burn more fuel and travel nearly as far as my commuter car. It is possible to design some communites that are less dependent on autos, but overall that reduction is going to be vanishingly small.
The arguments about travel time and costs are absolutely not disingenuous. Those are real impacts on real people. It is hard to work on your laptop when Metro is SRO. Metro runs on electricity, the least efficient form of power. Just because Metro isn't spewing fumes locally, by no means indicates it is clean. But you are right, transit is better where it works best. The problem is, that is only a small percent of where it actually works, and so we are losing money hand over fist supporting transit where we should not. It might be true that more transit could be supported cost effectively, if only we rebuilt the entire city around transit.
How cost effective is that?
I have to tell you, I originally held your opinions, but over time and with more study, I decided they did't make sense. Winston and Shirley make the convincing argument that if we simply selected the best modes of transport, and if we charged each mode their full allocated costs, then the results would be far different from what we expect.
Auto usage would cost far more than it does now, but it would increase from around 79% mode share to around 82%. Carpool would increase from 14.3% to 15.6%. Rail would decrease from 1.1% to 0.6 % Bus would decrease from 4.7% to 0.9% and taxis, that bastion of free enterprise, would stay the same.
In order to accomplish this, auto charges would vary over time, with a peak additional rush hour cost of 6.4 cents per mile. Bus fares would increase from 13.2 cents per mile to 55.2 cents per mile, in order to cover their true costs. Rail would go from 17.4 cents per mile to 34.8 cents per mile. ( 1990 dollars)
Bus freqeuncy would decrease from .95 miles per hour to 0.26 miles per hour, and rail frequncy would decrease from 20.8 miles per hour to 12.0 miles per hour.
According to their (compelling) calculations, such changes would have resulted in the best net public benefit in 1990. Changes in density may have altered those figures somewhat by now, but not by orders of magnitude.
This isn't about what I like. I could care less, and I have lived in circumstances not too disimilar from yours. But one size does not fit all. What do you do if you live in an apartment and like to play the tuba or drums? How many paople in Arlington have motor homes, skimobiles, jetskis or sailboats? Where do they keep them and how do they get there?
I have actually studied this pretty carefully, staring from your side of the fence. What I found was that the facts are far different from what is popularly quoted.
Following your post, is an entry from Richar Layman. He quotes some figures about traffic, buses, rail etc. His figures are theoretically correst. But the actual figures are much lower. Light rail in Charlotte caries almost no one, in spite of billions spent. There isn't any light rail system anywhere that crries anywhere near 16,000 people per lane mile. Not even during rush hour, and cetainly not on a 24 hour basis.
I recently spent a week at a (very boring) meeting in Portland. Looking out the window I could see thousands of empty light rail seats passing my window, hour after hour. During rush hour they were more full, but not compltely. The streets were always full of cars, until late into the evening.
I live in a very low density area. I can always go where I want with very little delay, and the closer I go to more densely populated areas, the more trouble I have.
I still own my house in Alexandria. Recently I had to replace water heaters in both homes. The one in Alexandria cost 50% more than the one out her in remote jepip, even though the installer had to drive 30 miles. I complained bitterly to my Alexandria installer, someone I have done business with for years. His explanation? It costs more here and it costs more to get around here.
Nuff said.
Finally, about Ed Risse. This guy is the JRR Tolkien of land use planning. You will seldom see him quote any authority other than his own. His math is untraceable, and anything he disagrees with, he dismisses as a myth, irrelevant, out of context, attacks the source, or simply claims as ignorant. As far as I can figure out, he is a paid flack for a special interest group.
I have to say, I was initally attracted to Ed's writings through a search on another topic. At first, I thought he was on to something, but the more I got into his arguments, the more specious I found them. I have an enormously large "yard". There are parts of it I have not even seen in a few years. If I am not at work at my regular job, you would have to drag me out of my "yard" with a team of Clydesdales.
Ed Risse's view of where the jobs, services, and recreation are is based on his own narrow view. I have never met him, but as I understand it, he lives on a large lot, drives an SUV, and he formerly lived in Faifax, then Reston, and now Warrenton. In the latter respect, he is not so different from me. In particular, bothof us spent our younger years in a rural environment: he in Montana, I in Ohio, and later in rural New England.
Despite our similar migration patterns, our views are like night and day.
When I am not at my regular job, my recreation consists of providing my own services, and taking care of a vast swath of open space that benefits others as much (or more) than it does me.
I enjoy the bistros and ethnic food in Arlington, I'm not a Luddite. But I'll invite you to come out here and bust your hump for a weekend, actually trying to improve the real world environment, and then tell me what you think of Metro.
The problem with Metro and Autos is simply that we cannot all go to the same place at the same time. Redskins games are a zoo. You very seldom have traffic jams in a forest, you also have very little in the way of commerce. What we need is more places, not more Metro.
The farm is a big metaphor for me. If I mow and rake and bale ten acres, I will have driven the tractor a hundred miles, not to mention the trucks. It is the same work whether I have a good harvest or not. I cannot afford to buy enough fertilzer, or haul enough organic matter to guarantee a good harvest for the whole farm.
After I decide what I can afford, I have to decide if I will spend the money and effort where the crops are naturally good, or if I will expend the effort in other areas to promote new growth.
If I don't do a certain amount of work, the whole place will go to hell and it will be much more work to fix later. It is not a question of either, or. It is a question of how much do I have to spend, and where do I spend it such that it gives me the greatest (overall) return.
I gotta tell you. It doesn't make any sense to me to put my eggs in one basket.
Why is it that the popular perception is so different from that proposed by so called experts?Because it’s counterintuitive, and because one of the glorifying illusions of suburbia is that it’s an “escape from the hassles of the city”, including, presumably, congestion. I will point out that while yes, you do see NIMBY opposition in the District and Arlington, it tends to be either in the more suburban neighborhoods like Tenleytown (which share that glorifying illusion) or spurred by other issues entirely, such as economic fears over gentrification. In particularly dense areas like the Rosslyn/Ballston corridor or Columbia Heights, it’s more well understood that density makes the neighborhood function better, not worse.
I see westbound traffic dropping off after the exit for Dulles. Is that the same as traffic to Tysons? Isn't that exit for Dulles traffic only? I'm confused.It is the exit for traffic to Tysons. It doesn’t become a toll road until outside the Beltway, and there is an exit at Chain Bridge Rd.
I still claim that people who would otherwise drive Route 66 straight through are diverted somewhere because of the lane restrictions.Yeah. Like the Orange Line.
Some of them are likely diverted to other roads, but isn’t it amazing how many of them disappear off the roads entirely when options other than driving are made easier.
No, I'm pretty sure Maryland has far fewer commuters into the District than VA. I looked it up. According to the census more come from MD, although the numbers for both look suspiciously low, so I would feel more comfortable looking more deeply. In any event, I don’t see how VA could possibly have more jobs *and* more commuters into the city. It is only slightly bigger population-wise.
I would refer you to a Brookings Report by Winston and Shirley, entitled "Alternate Route, Toward Efficient Urban Transportation". They offer a compelling argument against your frequently accepted assertion. Without actually reading the repot, I will assume it says something to the effect that “rail transit is more expensive at moving the same number of people as other modes”. There are many reports out there that mostly say the same thing, many of them published by the paid-consultant Wendell Cox / Peter O’Toole camp. That may be true in some cases, but it is not simply a cost issue. When I say “efficient”, I also mean spatially. It takes far less *land* to move people by train than by car. HERE is a simple graphic to illustrate the point (it doesn’t mention trains specifically, but you get the idea). If that report is anything like others I have read, it doesn’t take into account on-the-ground realities of urban areas, as opposed to suburban ones. And if the answer to congestion is to make communities more urban in layout if not statistical density, then space is a real issue. Furthermore, transportation guides land use to an extent usually not understood by purely economic reports. If moving people to and from their existing destinations without accounting for future growth is the extent of the problem, then express buses running in HOV lanes are easily the most cost-effective solution. But if your city is growing at all then rail provides an opportunity to guide at least some of that growth into a form that will both minimize congestion and increase the general prosperity of the city in a way buses and highways cannot.
(The possible load factor for transit systems is higher, but they never meet it so their real efficiency is lower.) Ride the Orange Line at rush hour and tell me it isn’t at capacity. Statements like that illustrate a lack or suppression of on the ground knowledge.
Despite the low efficiency of individual combustion engines, the overall cost of autos is only a couple of cents per mile more than standing room only on Metro, all things considered. And that is charging autos for ALL of their external costs. Given the other attributes of autos, flexibility, freight, convenience, speed, comfort, etc. etc. cars win hands down. Source? The Brookings Report? I find that hard to believe. My real-life experience with both cars and Metro certainly tells me otherwise. I would also not concede the point that cars offer more flexibility, convenience, etc. They can when used sparingly, but when cities are built exclusively to their needs, we become slaves to the cars and those positives disappear (as the post above by Chris Loos asserts). Cars can be a wonderful tool, and I sometimes use cabs or Flexcars because they’re more convenient than transit. The point is we have too much of a good thing, and in most suburban and many urban areas (not rural), the law of diminishing returns has long-since taken over.
Adding more Metro will also not decrease congestion, only increase the capacity to permit the same level of congestion You will note that my reasons for and solutions to congestion centered around auto-dependence and bad design, not simply building more Metro under the same circumstances. Running Metro in the middle of a highway median addresses neither problem and therefore solves nothing. Metro in highway medians is functionally no different than adding highway lanes. However, building Metro in a way that influences urban form, as it did in the Rosslyn/Ballston corridor, is another story entirely.
Shirley highway alone carried more people than all of Metro in 2002. No, it did not move more people than all of Metrorail. It moves more people “to the same destinations”, whatever that means. Metrorail ridership is over 700,000 on a given weekday. Of course not all of them come from VA, but statements like that undermine your credibility. But beyond that, what would happen if those Metro riders were removed to cars? We would need twice the highway capacity, sure, but also twice the parking capacity and twice the surface road capacity near the destinations. This is where urban form comes in. The more land you dedicate to cars in the form of roads and parking, the less can be dedicated to other uses, which forces development to spread out further, which ultimately makes it even more dependent on cars. It’s a negative causality loop that can only be fixed with better urban design, and for that trains are tremendously helpful.
The best system is one we don't have: jitneys. Jitneys are absolutely something we should have more of, but despite the rhetoric of Wendell Cox, they are not a panacea. Indeed, we have a jitney-like service in the District already with the Georgetown blue buses. They haven’t made anyone stop wishing Metro went there, or solved the problem on M Street. Jitneys, like BRT, are absolutely *part* of the solution. They are not, however, *the* solution.
For what Metro costs we could build office buildings where the Metro stops are AND GIVE THE OFFICE SPACE AWAY, FREE. The net result would be the same. Oh please. The net result would not be same because people would have to get there somehow. You would need a tremendous amount of land to accommodate all the roads and parking necessary, which would dilute the effectiveness of the land subsidies because it would spread out the development.
What do you do when you want to go to the boat in Annapolis, or take a sudden ski trip to Lake Placid? I rent a car if I need one and other options (like trains) are not easier. It doesn’t happen often, and there are facilities such as flexcar in place to make it easy (and affordable). If I did things like that frequently enough I would own a car and simply not use it every day. I am not against car use in general. Only *too much* car use.
Your situation fits you, but it only fits a very small percentage of others. The reduction in auto use we can reasonably achieve is also accordingly small. Your argument does not address at all the close correlation between road usage and economic growth. Are you paying attention at all or just attacking randomly? As I have already said, my lifestyle is not for everyone, nor should it be. That’s fine. There is nothing wrong with that. Anecdotes about how it’s not for you don’t change anything. The point is there are easy ways to build neighborhoods full of even single family, detached houses that allow you to build your boat and do not require you to change your lifestyle significantly, with only relatively minor changes to urban and transportation form. Refer back to the two community graphic above.
Even if you don't drive, you depend on roads for nearly everything that is delivered to you. Again, you are mischaracterizing my position as overly anti-road. Roads are not the problem, and nighttime truck deliveries are certainly not the problem. Accept the nuances of the discussion or it will end, because I am not interested in pointless debates.
I will guarantee that you have your AC on on days when I have the windows open. Urban areas are huge energy sinks, even if everybody walks. Wrong. I use my AC far less frequently now than I did when I lived in the suburbs. I use far less energy in general now than when I lived in the suburbs. “Urban” areas are far more energy efficient than suburban areas.
I agree with what you say about Wilson Blvd. But the idea that it can be universally extended is simply wrong. I own and operate a large farm, and I commute a fair distance to work. My tractors burn more fuel and travel nearly as far as my commuter car. It is possible to design some communites that are less dependent on autos, but overall that reduction is going to be vanishingly small. Your situation is extraordinary. Most people do not live on farms. Furthermore, I already conceded that the situation is different in truly rural areas. Really. Pay attention or the debate will end. I will not waste my time on another reply if you continue in this pattern.
But you are right, transit is better where it works best. The problem is, that is only a small percent of where it actually works, and so we are losing money hand over fist supporting transit where we should not. It might be true that more transit could be supported cost effectively, if only we rebuilt the entire city around transit. Again, design is the issue. Your last sentence implies you think it would be crazy to build the entire city around transit. Why is that any more crazy than building so much of the city around cars? As I have illustrated, transit can be better accommodated without requiring people to change their lives significantly. Furthermore, you are once again ignoring the nuances. It is not a zero-sum game. *More* can be done without totally revolutionizing the *whole* metropolis. Of course we should not tear down and rebuild everything, but should we not keep design in mind with new development and naturally occurring redevelopment? Should we not support efforts like the MetroWest redevelopment when they come along, instead of oppose them because of knee-jerk NIMBYism?
Winston and Shirley make the convincing argument that if we simply selected the best modes of transport, and if we charged each mode their full allocated costs, then the results would be far different from what we expect. I am all for charging more realistic user fees, but once again, it is just not that simple. As costs and urban form change, which modes are more or less efficient will change over time. How will we accommodate the poor? How will we deal with declining oil availability? If subsidies are removed and all transportation becomes more expensive, than transportation modes that are cheap, namely walking and biking, will become more important. As they become more important, walkable cities will become more important, which will make it less and less desirable to dedicate large amounts of land to parking and roadways, and make transit, which takes up less land, more and more desirable.
This study is sounding more and more like the kind that assumes transportation has no effect on land use and that current land use patterns will (and perhaps should) remain the same forever. Studies based on those assumptions are meaningless.
Light rail in Charlotte caries almost no one, in spite of billions spent. Excuse me? Charlotte has one tiny tourist-oriented streetcar. They are building a light rail line, but it is not operational yet. The streetcar is not meant for moving large numbers of people and did not cost one billion, let alone multiple billions.
You are rapidly losing whatever credibility you had.
I live in a very low density area. I can always go where I want with very little delay, and the closer I go to more densely populated areas, the more trouble I have. You live on a *farm*. Repeat after me: Rural is not the same.
But I'll invite you to come out here and bust your hump for a weekend, actually trying to improve the real world environment, and then tell me what you think of Metro. This tells us much about your views towards urban life in general, and your assumptions about what experiences I have had in my life. Again I ask: Do you suggest everyone be moved to the country? If not, then farms are not a solution to urban (and suburban) congestion and completely immaterial to this debate.
OK I oversold my case. I apologize. I'm willing to take on the nuances. We just can't do it all in one post. I'm no more anti-transit than you are anti-car. It is not about lifestyle.
The Charlotte comment was an error. Charlotte has not spent billions but we have spent billions over all and rail ridership is still down as a percentage of all travel.
I concede the point on Maryland Commuters, I don't understand the issue. Maybe where we went wrong is that I said the district and I meant the core area (what would have been the district had it not been ceded back, maybe). I mis spoke or wasn't clear.
I agree with your comment about putting trains in the median, but wasn't that done to keep the costs somewhere on this planet?
True enough Orange line is jammed to standing room only during rush hour. But load factor is figured over all the trains that travel, so since the Orange line trains outbound are mostly empty, the staring point is near 50%, and goes down from there depending on how many empty seats are being moved the rest of the day. Orange line does move lots of people during rush hour. I'm not convinced that is enough of a success to offset the other problems.
The Shirley highway quote was only for morning rush hours, your response about total Metro ridership misses the point, I think. I took the quote to mean that they compared likely Shirley Highway destinations to similar Metro destinations: Crystal City, Pentagon, Rosslyn, but it is open to interpetation, as you note. Lies, Damn lies, and statistics.
You are right about Winston and Shirley and urban form, they make no recommendations about how it should be changed to make rail more cost effective, but their arguments will hold no matter how the urban form changes: rail will only be cost effective for the very best routes. It is going to take enormous changes to move auto use very far from its present level, and those changes are going to be costly. If we were designing the city from scratch it might be different.
It seems to me there is a circular argument. We need to build rail to guide how the city grows, and we have to make the city grow a certain way so that rail will work. You are proposing to put rail in ahead of the need as a planning tool to channel growth in a certain way. Walking and biking are cheap, but they don't get you very much. Transit extends that some, and cars extend that even more. So I don't see any future where autos become less important (unless there is simply no energy, in which case all bets are off).
You can get out of your car and walk as easy as you can get out of a Metro car and walk, but you can do it in far more places with a car. Spatial efficiency comes at a a price, I think.
I simply cannot buy the idea that density and traffic are not related: it flies in the face of every observation. It might be and probably is true that the traffic in dense areas comes from someplace else, so that density itself cannot be blamed. That argument is a stretch for me. What if you move the reasons people go there to someplace closer to where they live? Not enough density to support the business, right?
What I meant about the office space was that people drive to Metro and park anyway, why not put the offices there and eliminate the need for Metro to carry them to some other office? I didn't understand your response. Giving the space away was hyperbole, but I think the point is the same.
I still think cities are energy sinks: they don't have heat islands for nothing.
Rail has a place, but cars are better overall, so far. Times may change, and they may be changing now. We need to build rail to guide how the city grows, and we have to make the city grow a certain way so that rail will work. Where I have a problem is when we demonize autos and deliberately and needlessly force new costs onto cars as a means of eliminating competition to transit. You have stated that is not your position.
As far as I am able to see, there is no total solution to urban congestion, a position frequently put forward by Anthony Downs. However, you are right again, more can be done without re-doing the entire metropolis.
It is no more crazy to build a city around transit than it is to build it around cars, and no less crazy either. Surely there is some optimum mix of residential density, job density, street density, and transit density, but I've never seen that attacked as a total solution.
We can do more with transit and more with cars both, where they make sense.
Well I looked up some numbers on energy usage for city and rural dwellers. We are both wrong.
Urban residences use less power on average, but rural reidences contain larger families, on average.
When you take it out to power consumed per person, it turns out to be exactly equal 12.27 million BTU's.
This is household use heat, air elctriciy, wood, fuel oil, propane etc. Automotive use is not included.
but we have spent billions over all and rail ridership is still down as a percentage of all travel.
Disingenuous argument. Of course it’s down as a percentage of ALL travel. We’ve spent many more billions on roads and on communities where people have no choice but to drive, especially for things like trips to the grocery store and dropping kids off at day care. Because of how we have chosen to build the vast majority of our communities over the past half century, rail and transit ridership as a percentage of all travel could only go down.
Indeed, that proves my point. We’ve built communities focused around cars, so now people drive more. If we build communities less focused around cars, people will drive less.
I agree with your comment about putting trains in the median, but wasn't that done to keep the costs somewhere on this planet? Yes, but ultimately it makes the line far less effective. If I were the god of transportation planning in the area, I’d spend the money to tunnel and build a first-rate line through Tysons Corner, then end it and run BRT in the highway median to Dulles and Loudoun County instead of Metro.
True enough Orange line is jammed to standing room only during rush hour. But load factor is figured over all the trains that travel, so since the Orange line trains outbound are mostly empty, the staring point is near 50%, and goes down from there depending on how many empty seats are being moved the rest of the day. So the point is when you take into account the cost of operating transit in both directions all the time, it becomes only slightly more economically efficient than the cost of using your car only when you use it. OK. Fair enough. But that still ignores the land use question. It is tremendously inefficient and tremendously costly in terms of lost tax revenue and diminishing returns on the quality of the urban environment to have cars sitting in garages (or worse, surface lots) all day. Indeed, you need multiple parking spaces for every car. One at home, one at work, one at stores and other destinations. There are at least three times as many parking spaces out there as there are cars, all taking up space and hurting the city. That’s the sort of thing economic reports always miss, but is among the chief benefits of transit over driving.
It seems to me there is a circular argument. We need to build rail to guide how the city grows, and we have to make the city grow a certain way so that rail will work. It isn’t as circular as it might seem. Firstly, in sprawly Virginia the idea that the majority of growth can be funneled into infill is hard to accept, but there is no reason why it cannot. Indeed, in Montgomery County, which is not far behind Fairfax as largest and most prosperous of localities, Gaithersburg was the last suburb for nearly 100 years, and Germantown the last for at least 30 after that. Now Germantown is becoming full enough that things are moving another step to Clarksburg, but the point is uncontrolled leap-frog sprawl is not a unavoidable circumstance. Secondly, Clarksburg also offers a good example of how most new growth on the fringe ought to take place. It is a “new urbanist” community designed like the bottom one in that graphic. It has a mix of uses laid out in a walkable manner. There are plans to eventually bring a light rail line there, but even in the mean time Clarksburg will be far better off than a comparably-sized community built exclusively around cars, and once we’re ready to put in the transit line, Clarksburg will be far more capable of supporting it.
It just isn’t the zero-sum game your replies suggest it is.
(unless there is simply no energy, in which case all bets are off)
A good point to bring up. The most abundant energy sources in the future will be nuclear and coal. That means there is going to be a lot more centrally-produced electricity. In any event, energy will certainly become more expensive over time rather than less, boosting the effectiveness of transit relative to cars. Gas *will* be $5 / gallon one day, and it won’t stop there.
You can get out of your car and walk as easy as you can get out of a Metro car and walk No you can’t, because as has already been discussed, the more space given over to cars the fewer places there are to walk *to*. That having been said:
but you can do it in far more places with a car. Spatial efficiency comes at a a price, I think. Certainly there is a balance. I have not and would never go so far as to suggest that cars should not be accommodated at all. Cars are a wonderful tool. We’ve simply gone too far with them. All I am suggesting is that relatively minor changes to urban form and street layout be made in future development so that a balance can be achieved. Remember the argument here is about traffic congestion. Traffic is bad because we’ve made it so easy to drive and so hard to do anything else, that everyone always drives everywhere they go. I am merely suggesting that we level the playing field so other modes such as walking and transit are at least as easy as driving. Surely that is not an unreasonable suggestion.
I simply cannot buy the idea that density and traffic are not related: it flies in the face of every observation No it doesn’t. It flies in the face of observations taken exclusively in suburban environments, with no attention paid to the regional scale. Add places like Rosslyn/Ballston to your observations and begin thinking about where traffic and density will go if not here (wherever “here” is) and it certainly does NOT fly in anyone’s face. Here’s the thing: When the design is bad, and those two problems I cited in the original post exist, then more density makes the problem seem worse on the local scale, since pumping more users into a system that doesn’t function in the first place will make its dysfunction more apparent. That is why small towns and smaller metropoli have less congestion than bigger metropolitan areas. But fighting to maintain the status quo hurts you as your city grows. That growth has to go somewhere (and indeed, as you pointed out, it is desirable for it to go somewhere, since the alternative is economic stagnation), so the question becomes how to accommodate it, and that is where these points come in:
1) Fighting against density will have the effect of pushing more of that growth to the exurban fringe, where it is likely to take the form of spread-out single use pods that force residents to drive a great distance for all their daily needs, further adding to the regional dysfunction.
2) High density with bad design is still better than low density with bad design (again, excluding the truly rural), because even if people are driving everywhere they go and only on a few streets, with high density at least they don’t have to drive as far for the same trips. Half a mile instead of 3 miles to the grocery store. 10 miles to work instead of 35. Etcetera.
3) Those people who now live on the exurban fringe will still have to get to the same destinations for work and many other activities, so they’ll be driving *through* the original locality that fought against their growth. Traffic becomes worse because people are driving more. This is easier to explain with a graphic, so I whipped one up quickly. CLICK HERE. It’s over simplified, but you get the idea. With the double density option, there are indeed more cars on the highway than before the growth, but with the double distance option the effect on the highway between the original homes and the ultimate destination is the same, but in addition to that people are driving much further just to get to the same starting point. Vehicle Miles Traveled go way up, and congestion inevitably follows. This is a well understood phenomenon, and one of the reasons many jobs began locating in the suburbs. That would work if people lived close to their jobs, but too many don’t (often for perfectly legitimate reasons such as mom works in Bethesda and dad in Reston), so by moving the jobs to the suburbs all that has been accomplished is people feel like they can live even *further* away, and transit has become less effective because transit relies on compact development patterns to work. Ultimately moving jobs to the suburbs made the dysfunction greater by encouraging people to drive more and making transit less effective. Keeping things compact combats those problems, so density helps rather than hurts.
Essentially, if it is a given that your urban form is bad, then adding density makes congestion worse on a local scale (surface streets) but has no effect on a regional scale (highways), while spreading out further makes congestion worse everywhere, *especially* on the regional scale, but in both cases the root of the problem is urban form, not density. On the other hand, if you have good urban form, then more density equates to more places to walk to and greater transit efficiency, which translates into *less* congestion as density increases.
Long story short:
Bad urban form = functions badly regardless of density.
Good urban form = functions better as density is added.
What I meant about the office space was that people drive to Metro and park anyway, why not put the offices there and eliminate the need for Metro to carry them to some other office?
Before I say anything, see the point above about spreading jobs out into the suburbs. Beyond that, though, if you wanted to build a series of little downtowns, that would be great and a vast improvement over the current suburban situation with or without transit, but:
1) Without transit it’s really hard to build a little downtown. Maybe impossible. You’d still have to have the same amount of parking on-site, which would in turn dilute the walkability of the mini-downtown and encourage it to spread out further. The end result would be an office park oriented towards cars rather than a little downtown. It would look more like Fair Lakes than Ballston. There’s a reason why all the best mini-downtowns in the region surround Metro stations.
2) If you did manage to do the near-impossible, hold the line on pressure to build lots of surface parking and managed to get a little downtown instead of an office park, that would be super, but the more of those you have, the more effective transit starts to become relative to roads, and in a few years it’s going to make sense to build transit anyway.
This is where the circular argument you mentioned earlier starts to look valid. If you build compactly, then it’s easier to run transit. If you build transit, it’s easier to build compactly. Of course, by the same token, it you spread out development, then it’s easier to build nothing but roads, and if you build nothing but roads, then it’s easier to spread out development.
But it’s not really a circular argument so much as it is just recognizing that transportation and land use are inexorably tied together.
I’ll add here, for the record, that most of Metro’s riders do not, in fact, use the park and rides. Most arrive by foot from stations in urban areas, where Metro is most effective. I’ve got numbers here somewhere, but would have to go find them.
Also, to clear things up: I am no great fan of park and ride stations. Like Metro in highway medians, they are functionally no different than adding lanes to the highway, except to the extent that they provide satellite parking that doesn’t hurt the core city by removing otherwise useful land for parking facilities… and in *that* case, it would be far more efficient to build simple commuter rail or even BRT instead of expensive Metro. This is where the purely economic analyses break down. Metro absolutely *is* a waste of money if you don’t use it as a tool to affect urban form, so any analysis that does not take into account changes to urban form as a result of Metro is completely and utterly useless.
Where I have a problem is when we demonize autos and deliberately and needlessly force new costs onto cars as a means of eliminating competition to transit. So you believe the playing field should be level, then? I agree. The point, however, is that it is NOT. Not by a long shot. Cars and suburban development patterns receive a lot of subsidies, both hidden and not hidden, that make them easier than transit and urban development. From tax breaks on home mortgages to roads getting more federal money than rails (and no, it doesn’t all come from our ridiculously small gas tax), to things like local area zoning that literally make it *illegal* to build anything other than spread out, pod-like development.
Our metropolitan areas don’t look like they do by accident. They’ve been carefully engineered over decades to be exactly what they are today. It’s not that it’s a conspiracy or anything so ominous. It’s just that you have to have a system, and the system we’ve set up heavily favors cars and spread out, pod development. Occasional taxes on cars that get used to pay for transit do not even come close to leveling the playing field.
I still think cities are energy sinks: they don't have heat islands for nothing. *Large numbers of people* are energy sinks. Compact cities are more efficient at handling them than spread out suburbs. Again, make the differentiation between metropolitan areas and rural areas. The people have to live somewhere, so how will we accommodate them?
Automotive use is not included.
Pretty much makes it a useless number then, eh?
Density has absolutely nothing to do with traffic congestion.
The common misperception is of course that it does. That density and “overdevelopment” are to blame for our woes, and if only greedy developers wouldn’t build so many houses and the government get its act together on building more new roads, the problem would be solved.
That is a myth. It is completely and utterly untrue. It is so overwhelmingly erroneous that most people simply refuse to believe it. People don’t like to be *that* wrong.
I-66, my local highway, is a perfect example. I live in a neighborhood filled with 20-plus story buildings with no setbacks from the street. It is tremendously dense, as are the adjacent neighborhoods east of mine. To accommodate all this density, I-66 has only two travel lanes in each direction, yet it is not particularly congested. If one travels a few miles west, however (that’s *away* from downtown, for the record), the situation is markedly different. I-66 expands to eight lanes and is surrounded by single family detached homes, strip malls and occasional office parks instead of highly urban neighborhoods, yet it is tremendously congested. Low-density, eight lane I-66 carries 182,000 cars per day at a crawling pace, while high-density, four lane, closer to downtown I-66 carries just 71,000, most of which move along briskly.
VDOT 1997 traffic volume data follows. Figures are published rounded to the nearest 100. The 13 miles from I-81 at Strasburg to VA-79 at Linden carries from 18,100 to 19,600 annual average daily traffic (AADTThe 15 miles from VA-79 at Linden to US-17 east of Marshall carries 26,000 to 34,000 AADT. The 15 miles from US-17 east of Marshall to US-29 at Gainesville carries 28,000 AADT . Traffic from US-29 at Gainesville to VA-234 is 56, and increases to 70,000 just east of Bus. VA-234; 89,000 just east of US-29 at Centreville; 131,000 just east of VA-28; 141,000 just east of VA-7100; 162,000 just east of US-50; and 202,000 from VA-243 Nutley Street to I-495. AADT just east of VA-267 Dulles Airport Access Road is 98,000; and is 102,000 at VA-120 Glebe Road, The Theodore Roosevelt Bridge carries 98,000 AADT with 9% large trucks.
Those are 1997 numbers, Lord knows what it is today. 182,000 vehicles divided by eight lanes is 22,750 vehicles per lane, per day. Using the higher figure it is 25,350 per lane per day. 71000/4 is 17750 vehicles per lane, per day, and using the higher number of 98,000 it is 24,500.
I-66 inside the beltway is restricted to HOV for several hours a day, so its congestion is artificially reduced, otherwise it would be at a standstill. The argument that it is free flowing because of the higher housing density in the area is false. It is (relatively) free flowing by law, as a means to encourage people to use the Metro. Absent the law, traffic there would come to a near stop, and that condition would encourage them to use the Metro. As a story in today’s paper put it, “Traffic forces these commuters to rise early and use transit”. Except, nobody is forcing them to live the way they do, they just prefer to create traffic than ride transit.
Using either set of numbers those figures are not appreciably different on a daily basis. But on an average daily basis neither stretch of road is particularly crowded. It is the peak load conditions that count most to the most number of people. What is perfectly clear from the numbers is that the farther you get from the highest density, the less traffic there is. It is the same with Metro. And even though Metro is somewhat better at peak load handling, it comes at a cost of money, level of service, speed, and convenience.
That does not imply that density actually causes the traffic, or the congestion, only that they are correlated. It is still entirely possible (but unlikely) that density has nothing to do with traffic congestion. But why is it whenever new housing or more dense housing is proposed, that people turn out against it at the public hearings and argue against it based on increased traffic congestion? It doesn’t matter whether it will increase the traffic from 1 to 2 or from 200,000 to 200,001. They are against it. An additional reason they are opposed is that residential density usually is thought not to pay its own way, and people feel erroneously or not, that it will result in higher taxes. If you plot the tax burden along I-66 the same way as the traffic burden, you will find a similar relationship.
Your contention that I-66 carries 182,000 cars per day at a crawling pace, while high-density, four lane, closer to downtown I-66 carries just 71,000, most of which move along briskly, is unsupported. If you have numbers which show the average speed of traffic at various times of day and locations, I would like to see them. I have spent many hours on inner 66, stuck in congestion, even when carpooling.
Your contention is at odds with my experience. When I drive into the city, I start out at high speeds, which slow appreciably in the vicinity of on ramps. The amount that traffic slows at the on ramps increases as the density increases, and as more vehicles suffer the uncertainty of merging. Even so, once the merge is complete traffic moves more quickly, until the next merge. The closer you get to more density, the closer together the on ramps there are, and the less time the traffic has to get sorted out between.
Finally, I-66 runs slam up against the restrictions and all but the car-poolers and hybrids must exit. That means they now have to merge either south or north on I-495 and that situation backs far up on I-66. One can avoid this situation, by exiting at Nutley street to take the Metro, or you can exit at 234 in Manassas and take VRE. Many people attempt to take the Vienna Metro option. Here the situation is similar to trying to get on to I-495: there is insufficient space to get everyone sorted out efficiently and traffic backs up onto I-66.
In other words, traffic is backed up on I-66 from the beltway out to 123 and beyond, partially because of Metro and restrictions designed to assist it (and keep at least some traffic moving on the roadway).
There is no easy answer. Suppose that the restrictions did not exist, and I-66 continued as 8 lanes instead of turning into an artificilly restricted inverse funnel at I-495, what would be the result?
Traffic would continue to travel “briskly”, as you say, at an average daily rate of somewhere between 17,000 and 25,000 vehicles per day per lane, depending on whose numbers you believe. Ridership on Metro would fall accordingly.
At least initially. But all those vehicles would have to go somewhere, as you point out, and that would mean a lot of parking spaces, as you point out. Parking spaces would be at a premium, and very profitable to the owners. By switching from Metro to the new, additional four lanes, drivers would save time over switching to Metro. Initially.
But because you could save time, some drivers will leave later, meaning more people will attempt to arrive at (more or less) the same place at the same time. Now the situation will have moved. Instead of having backups to get off of I-66 onto I-495 and Nutley Street, those vehicles will back up when they hit the traffic lights getting off on Glebe Rd. and the other close in exits.
So far, you have not changed either the number of cars that are moving, or the problem: they pile up when they try to exit the highway. There is not enough space to sort themselves out, too much uncertainty caused by traffic lights, other traffic, or pedestrians, and not enough parking, or parking space is expensive. The reason you have not solved the problem is that you have not changed the destinations. You need more places and more space to put them.
Running Metro to Vienna has done three things: it has moved the problem from Ballston to Vienna, offered somewhat more space to solve the congestion problem, and it has made Ballston more livable. If the Metro stopped at West Falls Church, East Falls Church, Ballston, or Clarendon respectively, then there would need to be more and more and more parking made available and more space made available for all that to sort itself out, and the remaining space would necessarily be less dense.
It is not so much that density works, as it is that density has pushed the problem someplace else: it works by using land elsewhere. You are correct in saying that Metro works best where people can walk to it: where the stations are close together. Anything after Ballston was probably a mistake. The other four stations would have been better off, say, along Columbia Pike, or in South Arlington where there isn’t service, or Georgetown. But wherever the terminal stations are, they will attract autos, just as the far flung airports do, and they will have to be accomodated.
Essentially, the closer and more compact Metro is, the more useful it is to the city. But the closer and more compact it is, the more space will have to be given over to accomodating transfers from autos, and the less the city benefits from density. The alternative is to make the Metro line less efficient and use the space out in Vienna. ( Soon to be Dulles and probably BWI)
I agree: If I were the god of transportation planning in the area, I’d have spent the money to tunnel and build a first-rate line through Tysons Corner, then end it and run BRT in the highway median to Dulles. Forget about Loudoun County: we wouldn’t need it if we followed your plan.
But Tyson’s/Vienna by itself is too big. It needs four Metro stations by alone. The way things are going, the average Wal-mart could use a couple of stations before long. The reason we didn’t do it, and the reason we won’t do it now, is that we didn’t (don’t) have the money. The only reason we have the money for Metro now now is that the States and the Feds are chipping in. That means that people who will never use Metro are paying for it.
Metro and density depend not only on other people’s land, but other peoples money. Yet you claim that roads and sprawl are over subsidized.
Certainly there is a balance. I have not and would never go so far as to suggest that cars (Metro) should not be accommodated at all. Cars (Metro) are a wonderful tool. We’ve simply gone too far with them. All I am suggesting is that we need more polycentricity in future development so that a balance can be achieved.
Your description of Gaithersburg, Germantown, Clarksburg might as well be Vienna, Centreville, Manassas. Lets run the rail right down the median on I-270 and save a lot of money. Oops, we decided that doesn’t work. What are the alternatives? Go as far as Ballston or Tyson’s and then put in a star. Maybe build the purple line: the Metro Beltway. No matter what choice you make, you follow the same mistakes we made with roads. Right now we have an axial Metro system, and we are about to expand it. The Orange line is crowded now, and the closer you get to density, the more crowded it is, same as with roads. We traded auto congestion for congestion while standing up nose to armpit: hardly a big advance in civilization, I’d say. If you add stars, branch lines or circumferential lines, then trains have trouble merging, just as VRE does now. Or you have transfers. The more metro tries to solve the autos "problems" the more it looks like a very expensive, slow, and uncomfortable auto.
So, you pointed out the problem with polycentricity. It is really hard to build a little downtown where transit works, maybe impossible. The obvious solution is to give up on transit, except where you can arrive on foot, as most Metro riders do. (I’d like to see the numbers on that.) Now you have to have a lot of stations, cross branches, etc. and you have the same problems that highways have, only slower and more expensive. It is not clear to me that if transit had the same money as roads, that it wouldn’t develop the same problems that roads have, and yet still have the problems that transit has.
This is where the circular argument comes in. If you build compactly, then it’s easier to run transit. If you build transit, it’s easier to build compactly. Of course, by the same token, it you spread out development, then it’s easier to build nothing but roads, and if you build nothing but roads, then it’s easier to spread out development. Since Metro is spreading out development anyway, what’s the difference?
Well, it is the land use. But as you point out, we are going to have cars anyway, so it is easier (and probably less expensive) to spread out development and have the polycentricity that works. (Better than major cities according to some economists) Even if Metro could work, and you had the money to spend on it, by the time you really got it working, it would use just as much land.
And why do we go to all the expense to put it underground? To keep it out of the streets! Out of the way of the cars. What if we ran cars in individual tubes and eliminated all intersections? That is a dumb idea, and just because the cars carry 88 passengers instead of 2 doesn’t make it any smarter.
I don’t know about Arlington, but citizens in Fairfax and Even Prince William are clamoring for more park space. Prince William is staring a major new initiative to increase and link their park spaces. It seems to me, that is going to necessarily increase polycentricity. As you noted, if you have enough polycentricity, it merges and you might as well have density to begin with. But if the point of density is to preserve open space, what is the point of putting it where people can’t enjoy and use it (without getting in their car)? What we see is a continuing demand for more open space (near us) and less (overall) density.
At the same time, cities and counties are finding that open space and parks are expensive. More and more they are turning to corporate sponsorships to raise the money to operate and maintain them. Once again, the cost savings from density are elusive. Instead of having private parks (yards) that individuals pay for, care for, use frequently, and pay taxes on; we have communal yards that we have to pay the government to support.
But all of this turns on the idea that people will have to travel through someplace else to get to their jobs in the city. Moving the jobs out of the city makes it hard for transit to work, and so moving jobs is bad. More circular reasoning. A bunch of little downtowns would be a vast improvement over the current sprawl, and it would be a vast improvement over high density cities, that’s why we are building have places like Clarksville (which was farmland the last time I was there). But those little downtowns can’t work without jobs.
Finally, there is the issue of evacuation. After the Houston and New Orleans debacles the argument was mad that you can’t evacuate a city with cars. You can’t do it with Metro either. All you would have is a couple of million people milling around the terminal stations, looking for their cars. If you didn’t have all those people in one place, you wouldn’t have to move them all.
Then there is the problem of where to move them to. Outside of Houston and New Orleans, there is pretty much nothing. If the goal of compact development is to save open space, then that is what you’d have. If instead, you had a higher level of polycentricity, then there would be less people to move, more places to go to, and less incentive to have an attack on the Metro, as in Spain. We need more places. We need more places that actually work, and where work exists.
We have been trying to make rail work for a hundred years, and it hasn’t yet. We had streetcar suburbs, and they failed. We have experimented with Metro for 30 years, and it has failed on a regional basis, even if it has partially succeeded in Ballston and Federal Triangle. New York, Paris, and London have experimented even longer. Financial enters and Venture capital in new York is relocating to Connecticut, Paris is suburbanizing, and London is paying for transit by charging congestion fees to cars, with the result that major businesses, like BMW, are moving out of the city.
One high government official said “Anyone who thinks Metro is about transportation, doesn’t understand what is going on.” Metro is a great thing. It is one of our National Monuments, if you will. But trying to force it to do what it can’t do is no way to achieve Balance.
Land use doesn’t have to be sprawl, and it doesn’t have to be cubicle warrens either. Cities have higher rents, higher taxes, more congestion, more pollution, and more problems. They are sucking up money from far and wide, just to keep them going, and they want more. Commuter taxes, congestion charges, dedicated funding for Metro and VRE.
For what? So the jobs can stay downtown? So we can continue to spend billions to move millions twice a day? I’m sorry. I can support higher local density, mixed use, and low average density; but this smart growth, new urbanist, transit oriented, pedestrian friendly, bicycle powered, polyethnic, low fat, smoke free, concensus driven, vibrant center city – is a myth.
Apologies for my delay. I've been busy lately.
I-66 inside the beltway is restricted to HOV for several hours a day, so its congestion is artificially reduced, otherwise it would be at a standstill.
Perhaps so, but where are the cars going? To some extent 50 and GW Pkwy, as you suggest,but many simply disappear. If anything, the HOV restrictions prove the point of induced traffic and further support my arguments.
The argument that it is free flowing because of the higher housing density in the area is false. It is (relatively) free flowing by law, as a means to encourage people to use the Metro. Absent the law, traffic there would come to a near stop... What is perfectly clear from the numbers is that the farther you get from the highest density, the less traffic there is. Now you're talking about trips that are generated outside the dense neighborhoods in question and simply passing through. Refer back to that quick graphic with the circles representing jobs and housing for an explanation. I think we can agree pass-through traffic makes little difference in a discussion about how traffic is generated. If it's easier for you, look at the surface streets. Fairfax Drive and Wilson Boulevard. Less congested than arterials in low density, suburban neighborhoods. And that's WITHOUT dedicated left turn lanes.
But why is it whenever new housing or more dense housing is proposed, that people turn out against it at the public hearings and argue against it based on increased traffic congestion? I've already answered that. Because it’s counterintuitive and because one of the glorifying illusions of suburbia is that it’s an “escape from the hassles of the city”, including, presumably, congestion. The problem is not that there are too many people in too small an area or on too few roads. Low congestion on highly dense streets proves this. If not on interstates, certainly on surface streets. The problem is that there are too many people WHOSE ONLY OPTION IS TO DRIVE GREAT DISTANCES FOR ALL THIEIR NEEDS AND WHO MUST USE THE SAME ARTERIAL ROADS FOR EVERY TRIP. Yes, dumping more cars into a system that already doesn't work exaggerates the problem, but the system doesn't work because of its inherent design. Better designed systems are perfectly capable of absorbing density, therefore density does not cause congestion. Furthermore, lowering the density by spreading development out doesn't work because the same problems exist but people are forced to drive further to meet their needs. Spread-out development in a suburban context only makes people drive more, which further breaks the system. The only solution is to change the system.
An additional reason they are opposed is that residential density usually is thought not to pay its own way, and people feel erroneously or not, that it will result in higher taxes. If you plot the tax burden along I-66 the same way as the traffic burden, you will find a similar relationship. That is absolutely false. All other things being equal, the more spread-out development is, the more it costs per unit to supply services such as police, fire, water and electricity. You will not find a credible economist, planner or police chief in the country who says otherwise. If single family homes required less subsidy, then all subsidized housing would be single family homes since that would be cheaper. How much people actually pay in taxes is inconsequential, since some parts of the state inevitably subsidize other parts of the state that are incapable or unwilling to pay their fair share. Northern Virginia offers an excellent example of that phenomenon, as Northern Virginians send more money down to Richmond than they get back.
Your contention that I-66 carries 182,000 cars per day at a crawling pace, while high-density, four lane, closer to downtown I-66 carries just 71,000, most of which move along briskly, is unsupported. The numbers come from VDOT. I'll post the link later when it's more handy.
I have spent many hours on inner 66, stuck in congestion, even when carpooling. Yes there is congestion inside the Beltway, but WEST of Ballston, where density picks up and the design problems begin to be solved. We've been over this already. Are you purposefully repeating already-addressed points or did you just forget?
When I drive into the city, I start out at high speeds, which slow appreciably in the vicinity of on ramps. The amount that traffic slows at the on ramps increases as the density increases, and as more vehicles suffer the uncertainty of merging. Even so, once the merge is complete traffic moves more quickly, until the next merge. The closer you get to more density, the closer together the on ramps there are, and the less time the traffic has to get sorted out between. Finally, I-66 runs slam up against the restrictions and all but the car-poolers and hybrids must exit. That means they now have to merge either south or north on I-495 and that situation backs far up on I-66. One can avoid this situation, by exiting at Nutley street to take the Metro, or you can exit at 234 in Manassas and take VRE. Many people attempt to take the Vienna Metro option. Here the situation is similar to trying to get on to I-495: there is insufficient space to get everyone sorted out efficiently and traffic backs up onto I-66. These points have already been addressed. Design is the problem, not too many people in too small an area. Furthermore, you are now failing to make a distinction between residential and job density, which you chided me for earlier.
The reason you have not solved the problem is that you have not changed the destinations. You need more places and more space to put them. Which you cannot create without inducing still more congestion. When you try to solve the problem by making it still easier to drive, you make it still harder for transit or walking to be competitive, and the problem ultimately becomes worse. The solution is to design communities where people need to drive less, rather than more.
Running Metro to Vienna has made Ballston more livable. If the Metro stopped at Ballston there would need to be more and more and more parking made available and more space made available for all that to sort itself out, and the remaining space would necessarily be less dense. Right. And as more and more space was given over to cars, density decreased accordingly, Ballston would become less and less livable.
It is not so much that density works, as it is that density has pushed the problem someplace else: it works by using land elsewhere. Incorrect. That would be correct if we were only talking about through traffic generated in the suburbs and headed to the city, but this is a discussion about the causes of congestion, so how the traffic is generated is of utmost importance. You are completely ignoring the traffic generated in Ballston. It's easy to ignore, because there isn't much of it. As you decrease the relative urbanity of Ballston, making it less urban and more suburban, you would be increasing the amount of driving people in Ballston do, because as a place becomes more auto-oriented other transportation options become less attractive. For the sake of argument, let's take a couple of hypothetical situations that would be impossible (and undesirable) in real life, but could help to clarify this issue:
Hypothetical 1: The entire metro area is built like Ballston. Congestion would be low, since people would rarely need to use their cars for any trip, and only need to use arterials or highways for long distance trips, many of which could be handled by transit anyway since transit would be much more efficient than it is now and therefore likely have much greater coverage. Rather than simply “push the problem to some other location”, the problem would not exist at all, at least not in the way it does today.
Hypothetical 2: The entire metro area is built like Fairfax. Congestion would be everywhere, and indeed much worse. It would be impossible for any one place to push the problem elsewhere, because the problem would be generated by all places equally.
If density were the root cause of congestion it would be impossible for dense neighborhoods to have the effect you describe. They could not push congestion away because congestion would be generated by them. The fact that it is possible for dense neighborhoods to suffer less congestion proves my point.
You are correct in saying that Metro works best where people can walk to it: where the stations are close together. Anything after Ballston was probably a mistake... I can agree w/ all that, except that East Falls would be a better terminal than Ballston for reasons you described.
The reason we didn’t do it, and the reason we won’t do it now, is that we didn’t (don’t) have the money. The only reason we have the money for Metro now now is that the States and the Feds are chipping in. That means that people who will never use Metro are paying for it. Highways are heavily subsidized as well. As is the air travel industry. I don't have any problems with increasing user fees across the board so that all modes pay for themselves if we decide subsidies are bad, but if that happens be prepared to pay a $20 toll whenever you want to cross a bridge.
However, I will agree that in many cases Metro is not the most effective means. Light rail has much the same effect on land use and is far less expensive. It's not a panacea, since it (like BRT, which also has a place) has far less capacity than Metro, but it should absolutely be one of the tools we use. DC will be able to absolutely blanket the city in streetcars for less money than they would have had to spend on the M Street subway.
Yet you claim that roads and sprawl are over subsidized. My problem is not that they require subsidies. As I said above, every mode requires subsidies or the user costs would be tremendous. My point was simply that roads and sprawl are not purely a market phenomenon as some would have you believe, but are a result of decisions no more inherently valid than would be a decision to spend money on transit and urban neighborhoods instead.
Subsidize both transit and roads or subsidize neither. I have no problem with either solution. I only have a problem with the mindset that views road expenditures as “investment” and transit expenditures as “subsidy”.
All I am suggesting is that we need more polycentricity in future development so that a balance can be achieved. The problem with that suggestion is that within the confines of a metropolitan area (not rural), polycentricity induces more driving and forces people to drive further for the same trip. Unless you solve the design problems, spreading development out ultimately makes traffic congestion worse. It's not a coincidence that America's most decentralized city (Los Angeles) is also its most congested.
We've been trying it that way for 50 years and the problem has only gotten worse everywhere. It's madness to think decentralization is the answer, unless (again) you decentralize to the point where we no longer have cities at all, which is impossible.
The Orange line is crowded now, and the closer you get to density, the more crowded it is, same as with roads. We traded auto congestion for congestion while standing up nose to armpit: The more metro tries to solve the autos "problems" the more it looks like a very expensive, slow, and uncomfortable auto. Not true, because transit and cars are not the only options. Far more people in Ballston walk than in Vienna, because there are far more places to walk *to*. As you get more urban and the walkability increases, you take more and more people out of the transit/road equation entirely since more and more are capable of walking (or biking) for a greater percentage of their trips. If you removed Metro thousands of people who now walk for most of their daily needs, including in many cases their commute, would be added to the flow on the roads *in addition* to the people who simply switched over from Metro to car.
That's the point of Metro. When you use it right it changes the system in a fundamental way.
So, you pointed out the problem with polycentricity. It is really hard to build a little downtown where transit works, maybe impossible. How so? I've explained why it doesn't work with cars (the more land given over to cars the less walkable it is), but you offer no explanations why it can't work with transit. Indeed, it has and does. What are the likes of Ballston, Bethesda and Silver Spring if not little downtowns?
you can arrive on foot, as most Metro riders do. (I’d like to see the numbers on that.) See here. But it's easy to confirm. Metro doesn't have enough parking spaces for the majority of riders to get there via car.
Now you have to have a lot of stations, cross branches, etc. and you have the same problems that highways have Except that it takes up less land, so even if transit costs the same as roads on a per mile basis, it becomes more efficient because you don't have to build as many miles. Furthermore, if (and only if) you managed to limit horizontal sprawl, polycentricity may actually be beneficial with small downtowns rather than detrimental, since adding more transit and pedestrian oriented nodes would not induce people to drive further.
For the sake of intellectual honesty I will admit that if you somehow managed to limit horizontal sprawl (it can be done, for the record) then polycentricity might help even a suburban, auto-oriented metropolis, except for the fact that it's impossible to limit horizontal growth without increasing density, and it would be stupid to increase density without trying to make it more walkable.
Since Metro is spreading out development anyway, what’s the difference? Walkability.
we are going to have cars anyway, so it is easier to spread out development and have the polycentricity that works. (Better than major cities according to some economists)
1) We can have FEWER cars driving FEWER miles. Remember the solutions I presented in the original post do not require transit to be in place. If we're going to build roads anyway, why not build them more smartly?
2) Polycentricity doesn't works
3) Getting rid of major cities is irrelevant because it's impossible. Even if you could do it, not everyone would want to leave. Again, unless you propose communist-style mass forced relocations, that is no solution.
And why do we go to all the expense to put it underground? To keep it out of the streets! Out of the way of the cars. Incorrect. It's underground so it takes up less space for pedestrians and allows more street level urbanity. You could achieve separation from cars without going underground by simply blocking off intersections and limiting access in the same way a surface-level freeway limits access.
What if we ran cars in individual tubes and eliminated all intersections? They're called interstate highways. Some of them are even underground.
What we see is a continuing demand for more open space (near us) and less (overall) density. Open space as a matter of quantity (instead of quality) is frequently a guise for simply not having more development nearby, which people fear for reasons already explained. Particularly in places like Fairfax and Prince William, people usually don't so much want open space as they simply *don't* want land to be developed because they erroneously think residential density causes traffic congestion. Actively used open space is another matter, but passive open space – open space just for having open space – is more often than not just a ruse... Especially in the suburbs where everyone has a lawn (or a lawn nearby) and tree save buffer areas surround just about every housing subdivision, and there is no great need for facilities like Rock Creek park to serve the general population.
Instead of having private parks (yards) that individuals pay for, care for, use frequently, and pay taxes on; we have communal yards that we have to pay the government to support. Disingenuous argument. It's not *instead of* a yard, it's *in addition* to a yard, assuming you're in a single-family home subdivision. People are willing to pay twice instead of once because as I said, it's not really about providing *parks* as it is about NIMBYism.
But all of this turns on the idea that people will have to travel through someplace else to get to their jobs in the city. Moving the jobs out of the city makes it hard for transit to work, and so moving jobs is bad. More circular reasoning. A bunch of little downtowns would be a vast improvement over the current sprawl, and it would be a vast improvement over high density cities, that’s why we are building have places like Clarksville (which was farmland the last time I was there). But those little downtowns can’t work without jobs. Again, the little downtowns would work if not for the fact that when someone works in downtown Rockville instead of downtown DC, they frequently live in Frederick instead of Rockville. If you found (and agreed) to a way to take Frederick out of the equation, then and only then would the polycentric model work.
Finally, there is the issue of evacuation. After the Houston and New Orleans debacles the argument was mad that you can’t evacuate a city with cars. You can’t do it with Metro either. All you would have is a couple of million people milling around the terminal stations, looking for their cars. If you didn’t have all those people in one place, you wouldn’t have to move them all. That's the most ridiculous thing you've said yet. That's like saying the Beltway is useless in emergencies because it only goes around the city and never leaves it. The argument also applies to wide suburban highways that narrow to 4 lanes outside the metropolitan area and couldn't handle the same capacity. Unless you propose widening every interstate highway in the entire country to 10 lanes you must recognize the difference between local/regional transportation needs and wider-area transportation needs.
And if you'd paid attention, you'd know the argument regarding Katrina evacuees wasn't over an urban subway system or even a commuter rail system, but for intercity rail like Amtrak, which would not end at a park and ride 10 miles outside town.
If the goal of compact development is to save open space Since when has that been the primary goal of compact development? Certainly nowhere in this thread about traffic congestion. You're doing it again. Please try to stay focused. It takes a long time to compose a response here (as I'm sure it does for you as well). I'd like the time not to be wasted.
and less incentive to have an attack on the Metro, as in Spain. Whoa whoa whoa. Let's not pretend highways, bridges and shopping malls are immune to terrorist attacks. Again your solution here only works if you decrease density so much that what you're left with cannot even be called a suburb, but rural, and it is impossible to do that everywhere.
We have been trying to make rail work for a hundred years, and it hasn’t yet. We had streetcar suburbs, and they failed. Actually they worked wonderfully until we started changing the rules and creating subsidies that had the effect of making that kind of development at first more expensive and later downright illegal. Read The Geography of Nowhere for a good discussion on that topic.
We have experimented with Metro for 30 years, and it has failed on a regional basis, even if it has partially succeeded in Ballston and Federal Triangle. New York, Paris, and London have experimented even longer. Financial enters and Venture capital in new York is relocating to Connecticut, Paris is suburbanizing, and London is paying for transit by charging congestion fees to cars, with the result that major businesses, like BMW, are moving out of the city. Again, all because we've created a set of rules that make suburban development artificially easier and cheaper. The playing field is not level, not by a long shot. In Europe (and Canada and just about every other country) where it is closer to level people are naturally much less attached to their cars.
And please. One *car* corporation relocating because a city decided to make life better for transit instead of cars instead of the other way around, as is usually the case, does not prove that London's economy is suffering. AOL's corporate headquarters are now in New York instead of Loudoun County. That doesn't mean Loudoun County is dying.
One high government official said “Anyone who thinks Metro is about transportation, doesn’t understand what is going on.” Right. It's about accommodating land use. Of course, for the record, as has already been mentioned it could just as easily be said that anyone who thinks adding highways lanes is about relieving congestion doesn't understand what is going on.
Cities have higher rents, higher taxes, more congestion, more pollution, and more problems. The rural issue again. Offer a solution that doesn't involve wholesale abandonment of all communities where people live on anything other than farms and I will start to listen.
They are sucking up money from far and wide, just to keep them going, and they want more. Commuter taxes, congestion charges, dedicated funding for Metro and VRE. That is simply incorrect. Cities *make* money from far and wide. Again I will mention Northern Virginia and Norfolk's subsidization of the rest of Virginia, and at the national level the money from the urban states being funneled to rural states. If cities didn't make money we wouldn't have them, period.
So we can continue to spend billions to move millions twice a day? The only way to get out of spending billions to move millions twice a day is if people live close enough to walk, which means redesigning according to my suggestions and using more transit and less cars. If the concept of a commute if your issue, then your position regarding metropolitan areas must agree with mine, or you are a hypocrite.
Because it’s counterintuitive and because one of the glorifying illusions of suburbia is that it’s an “escape from the hassles of the city”, including, presumably, congestion.
I'd buy that if people fought densification only in the suburbs. But there are enormous fights over densification even in urban areas.
Perhaps so, but where are the cars going? To some extent 50 and GW Pkwy, as you suggest,but many simply disappear.
This is just wrong. The cars that are stacked up on eight lanes of 66 prior to the beltway obviously don't "disappear". They go somewhere and ultimately reach their destinations. This is the kind of reasoning that makes your arguments suspect: they depend on magic.
The ones that don't get on 66 inside the beltway must go either north or south on the beltway. After that your conjecture is as good as mine as long as it does not include vanishment. I suggest that 50 and GW parkway are part of the answer only because I have followed vehicles on both those routes. If their ultimate destination is not Ballston or thereabouts, then your argument that the Arlington Metro development has reduced congestion does not apply. If that is their destination then your argument that density reduces congestion is wrong.
If there are cars that "disappeared" from route 66, then they are the ones that got off at Nutley Street and are now parked in the Metro lots there. Even *if* it is true that I-66 inside the beltway is less congested, it is not because of the dense development at Ballston, but because the congestion was moved to the Metro Parking lots in Vienna. For those vehicles we have not eliminated a trip, only changed the destination, and that is the crux of my argument: we need more places, not more places at the same place.
You concede that cars are part of the picture, and I concede that transit is part of the picture. Walking is a necessary portion of every trip, whether it includes auto or transit or not.
We disagree on the proper mix, the costs, the benefits, and the proposed policies.
I assume we agree that cars don't disappear.
> I'd buy that if people fought densification only in the suburbs. But there are enormous fights over densification even in urban areas.
I already addressed that.
> The cars that are stacked up on eight lanes of 66 prior to the beltway obviously don't "disappear". They go somewhere and ultimately reach their destinations
1) Cars absolutely do disappear off the roads as neighborhoods become more walkable.
2) The cars parked at Vienna Metro have disappeared off I-66 inside the Beltway. Yes it’s true they have gone somewhere, and it’s true that some of them would still choose to drive if not for the HOV restrictions, but the point is by making another option besides driving easier, traffic congestion on I-66 inside the Beltway in the peak direction was largely eliminated.
This is the incontrovertible lesson of induced traffic. The easier you make it to drive, the more people do. Of course it works the same for any mode. The easier you make it to walk or take transit, the more people do. By investing in more and bigger highways we make the decision that people should be driving more than they are now, and that is why congestion continues to get worse year after year despite the fact that we keep building new and ever-wider roads.
> If their ultimate destination is not Ballston or thereabouts, then your argument that the Arlington Metro development has reduced congestion does not apply. If that is their destination then your argument that density reduces congestion is wrong.
Again, already addressed. Here:
> It is not so much that density works, as it is that density has pushed the problem someplace else: it works by using land elsewhere.
Incorrect. That would be correct if we were only talking about through traffic generated in the suburbs and headed to the city, but this is a discussion about the causes of congestion, so how the traffic is generated is of utmost importance. You are completely ignoring the traffic generated in Ballston. It's easy to ignore, because there isn't much of it. As you decrease the relative urbanity of Ballston, making it less urban and more suburban, you would be increasing the amount of driving people in Ballston do, because as a place becomes more auto-oriented other transportation options become less attractive.
Trips generated in the suburbs and going THROUGH Ballston are inconsequential to the question of how Ballston’s density affects congestion. Density doesn’t magically cure congestion generated elsewhere, it simply doesn’t generate congestion on its own. You admit this yourself. It would be impossible for dense areas to push congestion away if they were generating it in the first place.
You are obviously getting desperate. You avoided totally the issue of what happens to the cars that are stacked up on 66 prior to the beltway. I wasn't talking about walkable cities or those that went to Vienna Metro.
Your arguments are getting weaker and less convincing. Simply saying you have already addressed an issue isn't adequate when the issue is clearly not the same. Being obtuse is not the same as being convincing.
You have over reacted to the rural componenent of my argument. I maintain that there is an optimum mix of jobs, streets, homes, and transit, and it does not require density. rural areas have no traffic, but no business either.
The fact is, whether you choose to believe it or not, that we now have evidence that those who live in the least dense (but still urban) areas drive only a little more than those in the most dense. The difference amounts to 44 gallons a year, the cost savings of which is offset by the costs of transit, somewhat.
Furthermore even though those drivers travel more, they mostly spend less time doing it: trvel times are highest in the most densely populated AND CONGESTED areas.
These issues have been documented and measured and with competing issues factored out.
Here is what Richard L. Morrill, professor emeritus of Geography at the University of Washington has to say
Density? Minor, but opposite of what many believe. The higher the density, the greater the congestion. Although higher density reduces auto dependence, and increases transit usage, congestion also rises because the demand for trips-per-unit-area increases faster than the shift to transit...." "The larger the city, the greater the congestion, simply because of specialization, larger concentrations of jobs and of people and the greater separation of people and activities."
"Any realistic and fair assessment of mobility requirements for a large metropolis reveals that significant investments are needed in both road and in rapid-transit systems. Anyone who imagines that we can get by with just roads or just transit is not facing the real world. One is not better than the other; they serve different needs." "[We have] Under-investment in transport capacity, for public transit, for quality roads and for off-street parking."
Now, those quotes and those figures pretty much match my own actual experience, financial and otherwise, in living a number of places, and my own actual observations.
Your claims don't match what I see when I look around, and they don't match what I can figure from competing studies and measures by professionals without an obvious axe to grind.
I'll give you credit, you are passionate and clever with your arguments, but I'm not convinced. We agre there is a place for cars and transit, but beyond that I find your arguments thin and not credible.
Nice talking with you.
>You are obviously getting desperate. You avoided totally the issue of what happens to the cars that are stacked up on 66 prior to the beltway. I wasn't talking about walkable cities or those that went to Vienna Metro.
Excuse me? We’ve already been through those numbers. I will not endlessly repeat arguments that have been made and not refuted. You have ignored large sections of my arguments when it suited you. You have no high ground on which to stand in that matter, particularly when you choose to ignore me when I originally discussed that point. Your ploy to make me appear desperate simply by calling me desperate, like most of your arguments, is in conflict with what has actually been said.
> I maintain that there is an optimum mix of jobs, streets, homes, and transit, and it does not require density.
Indeed. That’s why my solutions center around design, not density.
> we now have evidence that those who live in the least dense (but still urban) areas drive only a little more than those in the most dense.
I have not seen, nor have you posted those numbers. I seriously doubt them. In any event, if they do not take into account the design of the neighborhood then they’re worthless, because it is possible to achieve density with a pod-like, auto-centric design. People in the Landmark area of Alexandria may only drive a little less than people in Chantilly, but people in Ballston drive far less.
> trvel times are highest in the most densely populated AND CONGESTED areas. Again you are failing to take into account the difference between through traffic and locally generated traffic.
> the demand for trips-per-unit-area increases faster than the shift to transit
1. Same issue here, unless Mr Morrill believes people who live in apartments run more errands than those who live in single family homes, the density is not causing the congestion.
2. And what about walkers? The majority of the trips most people in Ballston make are not in the car or on the train or bus. They are on foot. If Morrill studies only the relationship of transit to density then he is completely missing the boat. Of course I can’t tell, because you haven’t given a source document, only appealed to the authority of someone with an impressive job title.
> Anyone who imagines that we can get by with just roads or just transit is not facing the real world. You are painting me as something I am not and you know it.
We already agreed that you understand cars have a place in the world. I am not painting you as something you are not. What I am saying is that (some of) your ideas are nonsense and unsupportable by fact.
The demand for trips per unit area increses faster than the shift to transit. That is a fact. Measured and proven, as far as I am concerned, and by several different authors. If you choose not to believe, that is OK with me.
It matters not in the least if the traffic is local, through, or destination driven. the fact remains there is usually more wher the density is greatest, whether it is residential density or job density.
I have lived in areas like Ballston, and my brother lives there now. I know better than to believe that Ballston residents necessarily drive less. He frequently, but not always takes the Metro to his job. He is within easy walking distance of the station, being the first residential house off the strip, and quite near the University station not far from the (Safeway?). He logs around 15k miles a year on his car. The plain fact is that people seldom walk more than 1000 ft. and even less so if they have to carry anything. Ken does not walk to the grocery, even though he could: it is more convenient to drive. Increased density brings a slight decline in auto use, for the most part.
The exact numbers are that people in the lowest quarter of density in urban areas use a grand total of 44 gallons a year more than those that live in the highest density quarter. We are talking here of a difference in density of 400%, for 44 gallons. If you haven't seen the numbers it is because your mind is made up and have not bothered to look.
Truly rural places have lower auto use than the less densely populated exurban areas. In fact, in truly rural areas auto use approaches the level seen in the most dense areas. We agree that rural areas are not part of this discussion, but I bring it up for comparison. Travel costs, as opposed to auto use, are much lower in rural areas.
It may be true that people in Ballston drive less, despite what my brother's experience is, or mine. You have not shown the data, either. If you can have density with pod-like design, then density alone is not the issue. If it is a matter of design and mix of use, then it should not matter where that design takes place.
The data I quoted is aggregate. However, they were built up from the block density level not from large areas like Landmark or Chantilly. The authors went to great pains to take out confounding influences.
You are still avoiding the point that started this. The data clearly show that traffic on 66 increases as you get closer to the city. It inceases more even after you pass Ballston. Even in the areas you claim are less traveled, if you consider lane miles. The success you claim for 66 as a result of Ballston is simply not related. My question was pretty simple. Of the cars that are stcked up between Nutley street and the beltway, what happens to them? Your answer seems to be that they disappear. This cannot be true. Instead, they are diverted from your neighborhood as a matter of law. If that is their destination, they must travel a cicuitous route to get there. I know because I have done it.
Maybe we are reading the same numbers differently, but that only shows both answers are open to question,not that either is correct. I stand by my numbers (VDOT), my experience, and my observations.
You cannot very well claim that this reduces auto use. It does however reduce congestion because it spreads out the traffic in both space and time.
Finally, although we agree that residential energy use is the same regardless of density, and I dcincede that suburban auto use is slightly higher, there is another issue. Once you get above four stories, construction costs per square foot are more than double. Capital costs and complexity are the major drivers of expense. Even if Ballston residents drive less, their choice probably costs them more.
Here is an example of the political reality.
We are one of only five [Neighborhoods] coded in yellow, as having a "net residential population density" of less than 15 persons per acre (the other four are Rosedale, Crestview, Windsor Road, and Barton Hills).
One member of the Planning Commission is quoted in this article as saying "'It can give us an idea of which neighborhoods have absorbed more (population) . . . and which ones we need to scrutinize more.'"
I don't know about you, but one reason the Keohanes purchased in Allandale is that it is an area primarily of single family homes with reasonably spacious grounds, and that is part of the quality of life we were buying into when we bought. It may take many of us, working together, to protect this, and to make the point, throughout Austin, that neighborhoods such as ours close-in, add not only trees, space, and ambience to the residents, but also real value to the quality of life for the city as a whole.
Post a Comment
<< Home