"Are trolley lines more than just a fashionable bit of nostalgia? Basic answer: Yes, as long as cities use them to drive economic development and more intense land use."

Mass transit is only economical in cases where you have lots of people going between the same places at the same times. I mean it's not only intuitive but well known that near empty trains and buses cost a lot more to build and operate 'per person' when there are only a few people using them. They actually cost much more 'per person' than someone transporting themselves there in an automobile. I.e., Mass transit requires (literally) a critical mass before it can be economical.

So it follows that in places where you don't yet have that critical mass it's far far more economical from a macro perspective to let people get there via cars or bicycles or by foot or whatever ... but not by a capital outlay-intensive and operational level-inflexible means such as mass transit. It's like you wouldn't build a 6 lane separated highway out to a new development. You'd build a 2 lane road out to it and then wait till there was the demand to justify all the fixed capital costs (and varialble operating cost) inherent to such a massive project.

Except of course if you were a developer. Since the taxpayer would be paying to fund a system of proportions so large that the need demand cannot meet supply, what do you have to lose? You buy the land cheap, build something nice on it (or in the place of something not so nice) and then let the taxpayers foot the bill for making your land more accessible ... Actually not just 'more' accessible, but 'unreasonably' easily accessible. It's a win win for the developers ... but a big waste of money for the taxpayers.

The money going to fund mass transit for developers would be far far better spent going to fund mass transit in areas that are already at the critical mass point where the economies of scale make the cost of mass transit less than the costs of non-mass transit. For example, we have lots of parts of this city where adding streetcars with dedicated paths could help eleviate some of the car and bicycle traffic that is clogging our streets. It could be a real value add to allow more people to get around cheaply in these areas. There would be an immediate return to the taxpayers by funding a streetcar system that would be of an immediate benefit to them. Instead, we're doing as Mr. Lewis proposes, and spending millions (and later on maybe billions) of dollars funding a streetcar system that will be of immediate value to developers and maybe future residents moving to the as yet to be developed areas... vs. to the people actually footing the bill for these expenses via their taxes today. And all so that it can be 'cheaper' to get to these areas for a few people ... 'cheaper' than just driving their cars or riding their bikes down to an area that has more than enough parking. To put it bluntly, it's a waste of our tax dollars.

by Lance on Oct 10, 2010 10:58 am • link • report

I'm happy when developers help pay for mass transit improvements, as in the case of the Silver Line extension to Dulles. Most transit lines aren't built in the middle of nowhere because of the ridership estimates (usually based on existing use) that are required for federal funding. And if transit development happens downtown, even if less clogged areas (if that's possible), everyone benefits. Maybe I'm just having trouble thinking of a specific example that matches your argument, Lance.

Re: CaBi, I like the idea of adding free time for taking a bike to the top of a hill. If they gave me an extra 30 minutes to do it, the $1.50 subsidy would cover the cost of taking a bus to carry the bike for me!

by Omar on Oct 10, 2010 11:07 am • link • report

I'm disappointed that Lewis just lumps streetcars and light rail together as one. The example he uses of Salt Lake City isn't really a streetcar at all, the train there has dedicated right of way on SLC streets at almost all times (partially thanks to Salt Lake City's enormously wide streets)

by Alex B. on Oct 10, 2010 11:13 am • link • report

Any articles trying to determine the success/failure of CaBi based on station density and commuting patterns are way premature. Besides the fact that the complete system isn't even done yet, distribution patterns take a little bit of time to determine. When the system is in place (and hopefully more stations added beyond those already planned) usage patterns will become perfectly clear and CaBi very probably won't need the three hours to determine when and where to move the bikes.

In addition, TheCityFix article makes a nonsensical comparison between the number of people who commute into D.C. with the number of potential bicycle commuters. What does it matter how many Marylanders and Virginians commute into D.C. everyday when determining how to manage a bikeshare system?

by Adam L on Oct 10, 2010 12:29 pm • link • report

The key point is about station density. They're saying that it's better to roll out to a smaller area and serve fewer neighborhoods to prevent the kind of spotty service the D.C. system is experiencing (and will likely continue to experience once the current phase of the system is built out). To spread so many CaBi stations over so wide an area was a political decision rather than an engineering one—an understandable political decision, but political nonetheless.

The fact that so many people are clamoring for stations near where they live and work is a good sign, though, and hopefully bodes well for the system's future.

by Omar on Oct 10, 2010 12:34 pm • link • report

Fashionable nostalgia?

When's the last time someone at The Newpaper called the new Mustang, Camaro and Challenger, or any of those cowboy-voiced truck ads, nostalgia?

by ThresherK on Oct 10, 2010 1:17 pm • link • report

The redunancy in a term like "voluntary agreement" is pretty much a dead giveaway of the coercion that is involved. Why not just call them nimby-property-owner-ultimatums? This would be far more descriptive. In this case, some guy with an MBA thought "cooperative agreement" was more palatable, when in reality it's just as silly. In five years, he'll have to think outside the box again and come up with something even cleverer...

by aaa on Oct 10, 2010 1:43 pm • link • report

@Omar Maybe I'm just having trouble thinking of a specific example that matches your argument, Lance.

The DC Streetcar lines. Starting the system by providing lines along streets such as Connecticut Avenue, Wisconsin Avenue, and 14th Street would be far more cost effective than building them out in places such as H St NE where there isn't a traffic jam or parking shortage issue. There'd also be far more support for it if 'the many' who can benefit from a streetcar system (and will ultimately pay for it) were the first to get it ... and not the last as is now being planned.

by Lance on Oct 10, 2010 3:13 pm • link • report

My first time using Capital Bikeshare was not successful. I got my $5.00 24-hour membership at Stadium-Armory, near my house. The bike I took wouldn't adjust its seat but I went along with it. When I got to my destination, Eastern Market, all the stalls were full, so I rode to the nearest location, north of the market by the pool.

After enjoying breakfast and doing my errands, I headed over to the Eastern Market station to grab a new bike to ride home. However, I punched my code into each one of the bikes, and not one of them would unlock. I took Metro home instead, which was faster and, given the $5.00 daily membership fee, cheaper.

It seems reasonable to save the money towards buying my own bike.

by Packherd on Oct 10, 2010 3:20 pm • link • report

Pakherd: Did you try calling CaBi customer service when the bikes wouldn't unlock?

by David Alpert on Oct 10, 2010 4:09 pm • link • report

@Pakherd

You need a new code to unlock the bike. I'm pretty sure I read on the kiosk that you just re-insert your card and the screen displays a new code. It would probably make sense to allow people to use the same code over again for the 24-hour period, but I assume that there are reasons that there are reasons that it just gives you a new code each time... like people who forget it.

by Adam L on Oct 10, 2010 4:14 pm • link • report

@ David - I did not. I had perishables with me and just didn't feel like dealing with customer service personnel on a Sunday morning.

@ Adam - I suspected that was the case and so checked my balance, but all that seemed to come up was a display of the charges to my card. It don't recall it asking if I wanted to view or print a code, as it had done when I initially paid the membership. Whatever the case, it wasn't intuitive to me.

by Packherd on Oct 10, 2010 4:28 pm • link • report

@Packherd: It would take me a long time to save the annual membership fee toward the price of a bike I wanted. And waiting for Metro isn't always fun, especially late nights.

@Lance: I see what you're saying, but if part of the goal of a streetcar is to encourage new development, it also makes sense to encourage development in areas that are economically disadvantaged. Directing development away from already crowded areas also benefits those areas indirectly through traffic reduction. I also suspect that starting the streetcars in the less desirable neighborhoods of DC was a politically motivated decision, one that I'm fine with to get the overall system built. You could also argue that Connecticut Avenue is already served by Metro; the idea is to build out new capacity. I'm not sure if Wisconsin Avenue residents are all in favor of streetcars either, but correct me if I'm wrong. It doesn't seem like a corridor that would lend itself to dense urban development in any case.

by Omar on Oct 10, 2010 4:37 pm • link • report

Lance, in 20 years, we will be glad we added those streetcars to growing corridors now rather than waiting until later, because later will mean never.

Of course, that argument you made was coming from a guy who couldn't possibly understand why the Hudson river might need another rail tunnel.

by Tyro on Oct 10, 2010 5:07 pm • link • report

If you feed the libertarian trolls, they'll just hijack things further.

by Rich on Oct 10, 2010 5:11 pm • link • report

@Omar it also makes sense to encourage development in areas that are economically disadvantaged/

My point was basically that such encouragement can be 'overkill' and not very 'efficient'. It's putting the cart before the horse because everyone is hoping that the demand will grow to fill the supply. As we can see with a lot of the metro stations in the area, that doesn't always happen.

And in the case of transit, it just seems to me that those places lacking development can do just fine with reliance solely on personal vehicles. Getting mass transit to a place that doesn't have the critical mass to support mass transit seems to be inefficient, especially in times when funding is limited.

You mention federal funding requiring ridership estimates. It sounds then like the feds may be thinking along the same lines.

If you compare it to softward development, you wouldn't build an expensive 'all bells and whistles' accounting or inventory (or whatever) system for a start up firm. It would be overkill. You'd give it a system to the scale and complexity appropriate for the business. And if you wanted to really 'help' the startup, you'd help it in ways that really made a difference ... and not by providing a system that by the time the startup might really grow to 'need' it, will probably be obsolete anyways.

I dunno ... maybe expensive mass transit systems are fundamentally different from expensive IT systems? And maybe not ...

by Lance on Oct 10, 2010 5:50 pm • link • report

Well, those economically disadvantaged areas aren't ghost towns, either. People do ride on those buses, and I bet that capacity alone justifies a streetcar. It's the halo effect of having a smoother, more modern form of transportation that is expected to generate the economic development (and has done so in the past for other projects, so it's not a pipe dream). There is also already economic development there that would benefit from having easier access to the center city.

And I remember a study recently that showed Washington was doing a better job than most of putting places of employment near Metro stations, especially in suburban areas because they didn't build down the middle of interstates and use park and ride lots.

Whenever you make a significant transit investment, you have to put resources where it will encourage future development. That's one of the basic ideas of smart growth. I don't think reliance on cars will help anyone living in a downtown area. The difficulty of parking and navigating traffic is a constant complaint for the suburban-minded. Light rail cuts through that.

by Omar on Oct 10, 2010 6:10 pm • link • report

I would think that it's best to add streetcars/trams/metro no necessarily where development is, but where development is desired.

by Amber on Oct 10, 2010 8:31 pm • link • report

The key point is about station density. They're saying that it's better to roll out to a smaller area and serve fewer neighborhoods to prevent the kind of spotty service the D.C. system is experiencing (and will likely continue to experience once the current phase of the system is built out). They may be saying that, but they don't prove it. Or at least I'll agree to this, there is some level of minimum system density, but no one knows what that is. We know that Montreal and Paris are above that minimum, and that DC is less dense than those two, but that does not mean that DC is below the threshold. The Paris and Montreal systems aren't even appropriate anymore. There is some indication that smartphones and their apps will make the minimum station density less important.

And what "spotty service" are you talking about that would be fixed by system density?

To spread so many CaBi stations over so wide an area was a political decision rather than an engineering one—an understandable political decision, but political nonetheless. I've been going to the bicycle advisory committee meetings for years, and this does not mesh with my experience from those meetings. There were many debates about system density. DDOT felt that it would be great to have a very dense system, but that would make it cover too small an area (SmartBike). They felt that with only 100 stations, a less dense system would replace more trips. The Paris system is designed so that most people are within walking distance of two or three stations, which maximizes convenience. DC's is designed so that most people are within walking distance of one station. In other words, the system was designed to serve the most DC riders and maximize usership. Their math may be wrong (though they did discuss that a lot) but politics were not discussed.

by David C on Oct 10, 2010 10:16 pm • link • report

I'm glad to hear the decision was less political than it looks. The idea behind station density is to keep bikes cycling through the system and to have redundancy if a station is full or empty. You don't end up replacing more trips of the station you need is empty, and 3 hours for replacement doesn't cut it during rush hour. The idea is to have more casual, dispersed use to supplement the system during times of peak one-way strain. The Bixi bikes work all right for long-distance commutes, but they are best for quick hops (a neighborhood station could replace a bus ride to the Metro, for example, and a more long-distance commuter could take the bike from there).

Of course, with limited funding, compromises must be made. I'm glad the wide deployment area is making more people interested in expanding the system. I hope to see more smaller stations added in neighborhoods over time so that short hops can help shore up supply for the longer die-hard riders. Also, the more you keep bikes moving organically in the system, the less fuel the vans have to burn doing the riders' work for them.

by Omar on Oct 10, 2010 10:58 pm • link • report

Omar, I agree with the theory. I think we'd agree that a 100 station system, where every station is located on one football-field sized area is too dense, and that a 100 station system spread evenly across North America is too disperse (Aw, the closest station is in Buffalo?). But I'm not sure that there is enough info out there to know the exact, ideal density. And I think that ideal density would probably change with the number of stations and population. It's actually a fascinating math problem, but it's one I don't think we have the answer to yet. So while people are free to speculate that we have the wrong placement for our system, I want to push back against the belief that it is a foregone conclusion that DDOT has done it wrong. (Obviously they don't have the one, ideal map as that would be almost impossible, but they may have one that meets their goals)

by David C on Oct 11, 2010 12:01 am • link • report

@Lance: "places such as H St NE where there isn't a traffic jam or parking shortage"

Check you facts: H St NE and Benning Rd are major thoroughfares to points east and do indeed have serious traffic jams every rush. Based on how hard it is to find parking around 5th & H NE in the morning, my guess is that many of the commuters work around Capitol Hill.

by goldfish on Oct 11, 2010 2:31 am • link • report

Lewis is one of those architectural critics that think everything done (architecturally) before Mies Van DeRohe is historicist nostangia, so I think one can take his use of the word nostalgia with a grain of salt.

If you wait till you have a critical mass to build light rail, it might be economically prohibitive. Also, who is subsidizing aoutomibile construction when everyones tax dollars go to highway construction? It's a red herring argument. Do smart planning, build a light rail infrastructure that will support it, and let the market place do it's think, assuming you are lucky enough to live in a part of the country that still has a market.

by Thayer-D on Oct 11, 2010 8:17 am • link • report

I guess Committee of 100 people do not leave the cushy confines of NW DC too often. Or, for that matter, ride the bus too often. The X2 bus that serves the H Street-Benning Road corridor is one of the most heavily traveled bus lines in the city. Even with short headways and articulated buses, the X2 is almost always crammed to capacity. The corridor was also historically served by a streetcar line. So, based on Lance's criteria, the H Street-Benning Road corridor is the perfect place for a streetcar. Keep serving up the softballs!

by rg on Oct 11, 2010 10:58 am • link • report

@Lance: Do you have any data to prove that H Street and points east intended to be served by streetcars are not experiencing congestion? Again the X2, or any of the buses that traffic the area, are often packed to the gills. I know anecdotal evidence means little, but that experience tells me more than suppositions without data.

by copperred on Oct 11, 2010 2:04 pm • link • report

@David: The math seems pretty simple to me: more bikes, more possible combinations of trips, fewer service outages. You also have the experience of the cities that started the concept of bikesharing to begin with to build upon. City Fix's mention of Maryland and Virginia commuters was a mitigating point to soften the argument that denser stations were needed since the city's off-peak population is much smaller than those of Paris or Montreal. But even then it's amazing how many stations go mostly empty while a few sit nearly constantly full. Some of this could be due to the hill problem, but this is another reason denser station placement helps: stations at the top of a hill could be served by other stations at the top of a hill.

I'm also confused by the logic that you replace more trips by placing stations farther apart Â… this reduces the number of available bikes to make trips and forces people to walk farther to get to a bike, making the system less convenient overall. It would have been more frustrating to have a smaller system; it would have seemed less useful. But it also would have been more reliable.

It is "only a theory," but it has been borne out in practice twice, and the only argument against the theory is an appeal to complexity. If the question is exactly how far apart the stations should be placed, then I would say however far apart would be convenient to walk. Anything more than that means someone is overly inconvenienced when a station is full, diminishing the value of their membership. And, lo and behold, this is a common complaint on CaBi's Facebook page.

Besides the functional argument, one of the most common complaints I hear about the pricing scheme from those new to the idea is the steep increase in fees once you get past 1 hour. Having more stations closer together would also be a psychological aid to help people understand that the system is meant for short trips (and possibly make them more likely to try it out). It's hard to make a second first impression. But it's too late to change that.

Finally, I would like to point out that CaBi could run more like a traditional transit system (hub-and-spoke, no pun intended), if it had major stations supplied by several small feeder stations around it. It might be difficult to find space for these in neighborhoods where people wouldn't object to having a parking space taken out, but I hope there will be political will to get it done.

Best of luck to CaBi and DDOT going forward. Hopefully it all works out in the long run. And I definitely appreciate the efforts made so far. Just looking forward to more, and hoping that the planning process follows established best practices in transit and other bike-sharing systems around the world. :)

by Omar on Oct 11, 2010 4:57 pm • link • report

Omar, if you're arguing that we need more bikes, then we're really arguing past each other. I'm arguing that the Paris arrangement of one station every 200 m would be a bad option for DC since we only have 100 stations. If we had 1000 that might be different.

The math part comes from asking "Where would you put stations if you only had 2? 3? 20? 200? etc..." It may be that going from 2 to 3 does not mean just adding one, but moving one or both of the others. It's, in part, a geometry problem and as I see it, is not simple.

The experience of Montreal and Paris is only applicable to a point. They had more stations, and larger systems. They have different population densities etc...It would be foolish to take Paris' system for deciding where to place stations and plop it in to DC.

I have not noticed a huge problem with stations going empty or full. Note: there is no problem when a station is "mostly" empty or mostly full. The only problem is when they are all empty or all full.

I'm also confused by the logic that you replace more trips by placing stations farther apart Again, I'm working from the assumption that we only have 100 stations. With limited stations, moving them farther apart means that more people/places are within walking distance of at least one station, thereby increasing the number of potential customers. It's true that we make the system less redundant, but it is possible that doesn't matter.

It is "only a theory," but it has been borne out in practice twice, and the only argument against the theory is an appeal to complexity. What we know is that it "worked" in Montreal and Paris. We do not know that less density wouldn't work in those cities or that it wouldn't work in DC. So that's two variables. At what point does increased dispersion not work? (unknown) In what ways is DC different than the existing models?

And, lo and behold, this is a common complaint on CaBi's Facebook page. I have not seen those complaints, but even if true that does not mean the current system fails to maximize the 100 stations. If stations were set up on a Paris Model they were serve and area about 4km square (a little more than downtown) so that would remove many possible trips and users. The current complaints about full/empties (of which there would still be some) might be dwarfed by complaints about "why isn't there one near me?"

Besides the functional argument, one of the most common complaints I hear about the pricing scheme from those new to the idea is the steep increase in fees once you get past 1 hour. Most trips are probably under an hour. This is much like Montreal's [So here we seem to switch places, with me arguing that DC's pricing system will work because it worked in Montreal, and you arguing that DC is somehow different so that it won't work. Though I will point out that this isn't exactly a role reversal. You're arguing that DC's placement system won't work because it isn't what worked elsewhere and, logically, it is a different argument].

by David C on Oct 11, 2010 10:02 pm • link • report

Montreal's price structure isn't a problem; I'm saying the distance between stations serves as an impediment for adoption by new or casual users. You don't know that the ride between distant stations is consistently less than 30 minutes until after you've given the system a fair shot. I try to get that point in now when people ask me about the pricing. The impression is that "it's expensive." People misunderstand what the purpose of the bikes are. If it had rolled out to a smaller area first as it did in Montreal and Paris, I'm thinking that the short trip aspect of the system would have been more abundantly clear.

Answering the complaints would have been simple. "Why isn't CaBi in my neighborhood yet?" "Because to work best, bikesharing needs a sufficient density of stations to run smoothly. Rather than split the baby and serve as many neighborhoods as possible, we want to serve a few neighborhoods well and expand the system as more funding becomes available." Would this have been a popular answer? No. But it would have been best for the system's optimal operation. This is why I assumed earlier that the decision to spread out the stations was political.

With a smaller initial system, the difficulties being experienced with the new system could have been avoided, and CaBi could have benefitted from better word-of-mouth reputation for future expansion efforts. The stations were also designed to go in parking spaces, unobtrusively Â… I wonder how successful future efforts to expand into smaller neighborhoods will be if people think that stations have to be at busy nodes. The opposition to the Lincoln Park location is a prime example.

But we'll see. None of this may matter in the long run. I'm good at worrying about these things. :) And while I'm laying out my position as clearly and distinctly as possible, I absolutely see merit in your points as well.

by Omar on Oct 11, 2010 11:24 pm • link • report

Because to work best, bikesharing needs a sufficient density of stations to run smoothly. I see what you're talking about, but it depends on what you mean as "best". If by best you mean it is easiest to use then that may be true because there would be more redundancy and require less planning. But if by "Best" you mean "I can go to so many places with it" then that isn't true. So there are two distinct and opposing goals. (1) Provide coverage to every area that makes sense - which pushes the system out (2) providing redundancy and ease in finding stations - which pushes the system in. It may be that different cities will choose different paths because they value different outcomes.

In the end, I think the system on hand will meet it's goals - as I recall them - of covering operating costs and having 4 uses per bike per day by year 5. If it does, it will be a success even if it is not ideal.

by David C on Oct 11, 2010 11:34 pm • link • report

Omar, Toronto, it turns out, is doing the opposite of DC. 1000 bikes but in a very small area. It will be informative to see who has more success.

by David C on Oct 11, 2010 11:55 pm • link • report

@David C It will be informative to see who has more success.

And how do we measure success in this circumstance?

by Lance on Oct 12, 2010 12:04 am • link • report

Lance, Ridership. Revenue. Membership. Rides per member. But it still may not compare well because of different geography, weather and population density.

by David C on Oct 12, 2010 12:36 am • link • report