Human Transit

dissent of the week: should minimum walking distance be measured round trip?

Last week I posted on the odd phenomenon of one-way splits, where the two directions of transit service are moved some distance apart.  My point was that it's a great example of symbolic transit, in that the lines appear to cover more area, in a way that looks nice to developers, but they actually cover less, if you define "cover" as "being within walking distance of both directions of service."  If we assume an acceptable walk to be, say, 400m, then as you can see in the diagram below, the area within walking distance of both directions of service (blue in these diagrams) shrinks as the separation gets larger. 

Most commenters agreed, but there was an interesting dissent.

The post is misleading. It makes more sense to calculate the round-trip walking distance, in which case the areas covered in blue would be the same.

As I understand it, this commenter thought I should draw the situation this way:

I have to admit I had never heard this definition of coverage before.  It implies that if we have to walk 800m to the bus in the morning but 0m to return in the afternoon, we're within a acceptable 400m walk.

What do you think?

wellington: responding to environmentalist critiques of a network plan

I spent much of 2011 leading a major study for Greater Wellington Regional Council, which developed a proposed new bus network for New Zealand's national capital.  The resulting proposal re-allocates existing resources to create a much more extensive network of high-frequency all-day service, so as to expand the area in which transit can be used spontaneously for all of life's needs.  This kind of extreme convenience is essential to reach a range of sustainability outcomes, notably by making it much easier for someone to chose to own fewer cars. 

A remarkable outcome of our proposal is that the percentage of the population within walking distance of frequent all-day service goes up from 58% to 75%.  The last 25% live in very hard-to-reach or low-density places where it would be cost-prohibitive to run frequent service, so we are bringing high-frequency service to almost all parts of the city where it the densities and road network make it viable.

That's important because frequency is freedom. High-frequency service (every 7-15 minutes or better) is service that's ready to go whenever you need to go, and that can even be used spontaneously to move around the city.

This kind of network design work is a holistic exercise in multi-variable problem solving.  Each idea for improvement has many knock-on effects that we have to evaluate, and it takes skill and experience to see the best network patterns that optmise across so many issues.  The outcomes don't please everyone, and especially don't please people who are opposed to all kinds of change.  But it is exactly the sort of network design that leads to stronger urban transit networks that more people find useful.  This kind of design also supports more intense urban development where that's appropriate, thus providing more alternatives to horizontal expansion of the urban region. 

However, we're getting a lot of objections from the Green end of the spectrum.  These have been summarised for me as follows.  For the benefit of Wellington readers and also for anyone interested in the issue of Green responses to network plans, I'd like to reply briefly to each.  These are my own views and do not represent the policy of Greater Wellington Regional Council or any other agency involved in the proposal. The concerns are in italic, with my responses below.

1. Some concern that technocrat/experts (even smart, nice people with great intentions) may not come up with a better solution than the whole community using the network--therefore it would be desirable to have more extended community input before implementing changes, or alternatively, to take staged approach to implementation to see if it works as intended.

I'm a big advocate of crowdsourcing network design ideas, and often run workshops that invite local stakeholders to think together about a network design problem.  That tool was not used in this study, but we did look for every possible source to inform the planning, and of course the government is consulting with the public now in anticipation that some changes will occur as a result.  Note also that we relied extensively on data about how the system is currently being used, and this too is a form of public input.  

However, if you haven't been designing transit networks for a long time, there are some facts and techniques that won't be intuitively obvious to you.  My book Human Transit is designed to address exactly this comprehension gap.  It helps people who aren't transit planners to understand the transit network design problem, and thus to better evaluate transit proposals.  And as I say many times in the book, the point is not to impose my values but to help communities express and implement theirs.  

2. Concern that current journeys are influenced by the transfer fare penalty, and until we implement integrated ticketing, we won't really know the journey desire lines and therefore that really needs to happen before a network review.

I have said consistently throughout the project that to fully succeed, the proposal will need integrated fares permitting free connections, at least within Go Wellington, the operator covering most of the city. I'm told that a separate study on fare policy, including connections, will be underway soon.

Again, charging for connections is perverse, because connections are a necessary inconvenience, not a value-added feature.  One strong reason to potentially postpone the Northern Suburbs area of the plan is that it requires free connections between different operators (two bus companies and a rail line) and this is much harder to achieve than free connections within the one operating company that covers the rest of Wellington.  It makes sense to start in the main part of Wellington covered by one operator, where eliminating connection penalties is realistic in the short term.

3. Concern that the infrastructure for the trolley buses will be immediately removed on the routes that will be superseded, before we know if the new proposed routes will be successful and permanent.

At no time during the project did anyone suggest to me that this be done.  On the contrary, the thinking throughout the project was about ensuring that the trolleys had every possible opportunity to succeed.  In fact, one proposed line, Line C, is mostly under trolley wire and designed to be easily converted to trolleys once there is a decision to expand the trolley network.

I personally am a trolleybus advocate, as you can see here.  However, our task in this study was to design the best possible network for the city, without presuming an outcome about a future debate about trolleybuses.  That debate will occur in a separate study that the council is undertaking now.

Planning is impossible without separability.  Greens in particular tend to see how everything is deeply connected to everything else, so they are understandably suspicious about studying anything in isolation. Often they're right, and two issues really have to be studied together.  Still, while everything is connected, governments simply can't do studies of every connected issue at once.

In this study, we dealt with this question around the trolley bus issue and also around the fare policy issue. Wellington needs a study and public discussion about the future of the trolleys.  Wellington also needs a review of fare policy addressing the connection issue.  But if we did a plan that tried to deal with all those issues as well, it would be so huge and multi-messaged that few people could follow the whole thing, or see how the parts are related.  The hard experience of planners and elected officials is that such overly grand proposals usually fall over from their sheer complexity of message and the diversity of controversies they raise.

So the Regional Council took the view that we should study the city's mobility and access needs first, and get some clarity about what the network should ultimately look like.  That's because they wanted their fare policy, and their trolley policy, to follow from the city's transport network rather than preceding and constraining it.  If your goal is a city where people feel freer and more empowered to access the riches of their city at all times of day, that's the logical course.  

4. That the network review will be used as an opportunity to force the wholesale removal of trolleys by those who are politically inclined to ditch them because of short term thinking.

I can't control how the proposal is used, but this was not the intent, and I don't agree that the proposal advances a case for removing trolleybuses.

The proposal does reduce the number of trolley buses needed in service, but only in the context of reducing the number of all buses needed.  Fleet requirement overall drops dramatically under the proposal, because there is tremendous inefficiency in the existing peak schedules and we wanted to reallocate that service to create more abundant mobility all day.  (However, we also checked existing peak loads, and took great care to ensure that the plan will be able to handle all peak crowding that's currently observed.)

Since the trolley fleet requirement goes down in the context of the whole fleet requirement going down, we clearly weren't discriminating against trolleybuses in any way.  

The proposed fleet reduction is a good thing for Wellington, becuase it makes service cheaper to operate and thus creates the potential for future abundance.  Don't measure the plan by its fleet requirement.  Measure it by the mobility outcomes that it delivers, including the dramatic expansion of all day frequency.

5. The debate about the trade off between maximizing patronage and maximizing coverage (for social inclusion) was not explicitly had by the community... This policy to maximize patronage is essentially a decree by Government, and GW should be consulting more explicitly about this trade off before just deciding to redesign the network to maximize patronage. 

I have facilitated exactly this debate many times, and Chapter 10 of Human Transit is all about it.  I agree that the debate should ideally be a public one.  

In this study, the decision was to avoid reducing coverage as much as possible.  We retained coverage to virtually all areas now served, though not necessarily on the same streets, and in some cases we expanded it.  If we had been charged with shifting priorities massively toward ridership, we'd have proposed deleting a lot of coverage to low density or hard-to-serve areas, including areas like Kowhai Park, Wright's Hill, Houghton Bay, Melrose, Owhiro Bay, Broadmeadows etc.  We didn't do that.

Related point: radical network redesign has a human cost in terms of elderly, infirm, other marginalized users who may be too confused, disoriented or afraid to use the new network because they will feel lost.

This is a major issue in every network design.  But this concern, stated so broadly, becomes a reason to never change anything, and to allow the network to retreat into irrelevance by never updating it to reflect the needs of a changing city with growing sustainability goals.

The answer to all these needs is a mixture of (a) careful information and travel training, to ensure that information or understanding is not a barrier to the new network, (b) appropriate on-demand services for people who simply cannot use fixed route service, in their location, for reasons of disability, and (c) urgent attention to where facilities are built for seniors, disabled persons, and others at risk of falling into this "marginalized" category.  Locate on the network of high-quality freqeuent services (which the plan expands) and you won't need to worry about your service changing often in the future, because the more frequent a service is, the more permanent it is.  

6. Some concern about the principle of designing for an all day basic network because the nature of journeys in Wellington  is very different during the day for work trips and other kinds of trips off peak etc.

While travel patterns change in the course of the day, there is a very important reason to resist changing the pattern of service several times during the day.  Such changes result in maddening complexity, and dramatically undermine's people ability to use the system freely and spontaneously.  

Imagine what driving in Wellington would be like if you had to remember that the basic geography of the city changes completely several times during the day.  Suppose that during the day streets changed direction, or became longer or shorter, or closed at certain hours (different for each street), or even that streets connected to each other differently depending on the time.  You might justify this based on optmising flows for each moment of demand, but it would drive any motorist (or cyclist, or pedestrian) mad.

Buses physically can change their patterns in very microscopic ways, but good urban networks don't do that much because the overriding goal is to be simple and legible.  Sustainability outcomes in particular, notably reducing car ownership, require that the transit options be simple and easy to trust.  And that means transit lines that are doing the same thing all the time.

So instead of changing patterns by time of day, we focus on designing a basic network that makes sense at all times, and then adding supplemental peak-only services when certain flows appear.  Even then, we focus on trying to make the all-day network useful for as many trips as possible.  Because that's how you build a network that feels, and is, permanent, ready to support the further growth of the city in a more sustainable form.  For a more complete discussion of this issue, see Chapter 8 of Human Transit.

To sum up, the proposal is designed to aggressively support the growth of Wellington in more sustainable ways, and as such it deserves some Green support.  Unfortunately, mobility and access outcomes are hard to illustrate.  I wish every citizen had access to an online tool that would show how their personal transit access would change under the proposal.  It would something like mapnificent.net for both existing and proposed networks, showing how the proposal expands the area that you can get to quickly, and thus, in a very basic sense, expands your freedom.    

Only a network that does that will help people sell cars and trust public transport more for all the needs of life.  So consider judging the plan on how well it does that.

the bbc discovers me

Just a few paragraphs at the end of an excellent article by Brian Wheeler.

on infrastructure usa

Infrastructure USA, a national blog of repute, interviewed me last week, and rendered the result as a series of clips, here.

one-way splits as symbolic transit

 Now and then I see a professional study of a transit line -- often light rail or streetcar -- that suggests that the two directions of service should be a little bit apart from each other, say on different streets, so that they "cover more area."

This is the clearest and simplest example I've seen of the conflict between symbolic transit and actual transit.  If you are creating transit for symbolic purposes -- say, to give the appearance of permanent mobility so as to stimulate development, then it's certainly true that separating the two directions, so that rails and stops appear on two streets instead of one, will spread that appearance over a larger area.

However, if you care about people getting where they're going, the one-way split reduces the area served by a transit line.  That's because for a two-way line to be useful, you have to be able to walk to both directions of a service.  The further apart the two directions are, the smaller the area (light blue) that will have a reasonable walk to both of them.

 

If you're wondering whether a project is about getting people where they're going or just appearing to do so, the handling of one-way splits is often a clue. 

Obviously, one-way splits for transit are often required by a one-way street pattern, but even in these cases, when we're planning for both legibility and ease of use, planners sometimes suggest combining the two directions on one street.  This can be done by giving transit a lane that allows it to travel against the main traffic direction (called a contraflow lane), so that although traffic is split between two streets all transit is on one.  That maximizes the area actually covered by the line, but of course, it may reduce the area that has symbols of being covered! 

a great summary of my talk at apta, washington dc

In Greater Greater Washington (GGW), Jenifer Joy Madden and Malcolm Kenton have written an excellent summary of my talk at APTA in Washington DC last week, which GGW also partly sponsored. It also includes this photo, which makes me look a bit like a preacher.   (Click to enlarge, if you must.)

If you missed my talk(s), please read the article but also the comment thread, in which some people accuse me of "anti-rail bias" and others say everything I would say in response to that.  This is gratifying to say the least.  It's fun to be applauded, but it's far more fun to be understood.

a towering presence

That's what John Hendel calls me in his TBD profile of me, based on a few minutes we spent together on a Washington DC streetcorner (K and Connecticut).  Hendel elicited some comments on the challenged of multiple entangled transit systems, so it's worth a quick read. 

(Small factual correction: I don't blame WMATA for me being 45 minutes late, only for 15 minutes.  I had advised John much earlier that I would be 30 min later than first planned.  John combined the two events for rhetorical effect.)

san francisco: cameras will enforce transit lanes

Lanes that protect transit's speed and reliability are only as good as their enforcement.  San Francisco, like many cities, has long had a few lanes whose enforcement was so spotty as to render them advisory.  Now, the city is getting serious, with camera enforcement and significant fines. 

I think another feature will eventually be necessary:  Full painting of bus lanes in the same way that we usually paint bike lanes, along with signage on approaching intersecting streets, so that there is no "I didn't know" excuse.  Sydney, for example, paints its bus lanes deep red, with gaps at the points where cars can cross for turns.

In addition, if you approach as a motorist on an intersecting street, you will see a sign with a lane diagram of the street you're approaching, so that there's no excluse for turning into a bus lane.  Fines (stiff) are often advertised prominently.

 

socrates visits the u.s. federal transit administration

 

FTA staffer:  Welcome to Washington, Socrates!  The literature and philosophy students on our staff can’t stop talking about you, and suggested you could help us think something through. They told us you ask good questions.

Socrates.  I hope I can help.

FTA staffer:  So we invited you here because we are devising a new way to decide which transit projects are worthy of funding, anything from a little streetcar to a busway to a big subway line.

Socrates:  And when you deem that a transit project is good, do you mean that it has some intrinsic goodness in character – perhaps its pleasing color or shape – or do you mean that its goodness lies in providing some benefit to others?

F.  In the benefit, certainly.  It’s how to describe the benefit that gets us into trouble.  Our policy is to focus on mobility and accessibility benefits – basically, people getting where they’re going.  But it’s hard to translate that into a measure …

S.  You would have to define those terms first.

F.  Of course.  You see, for a while now we’ve been scoring the benefits of a transit project based on the amount of travel time it saves.  Basically it’s person-hours or person-minutes, one person saving one minute of travel time. 

S. You mean one person gets to his destination a minute sooner than he would without the project.  That’s a person-minute?

F.  Pretty much.  We prefer to count hours because minutes seem – I don’t know – petty, somehow.  But still, you know, people getting there sooner, it seemed like a good idea for years.

S.  Tell me: if Jim has $1000, and Dave has $100, and each is given another $100, you would say that the two have benefited equally?  Even though Jim's wealth has gone up just 10% while Dave's has doubled? 

F.  I don’t follow …

S.: Well, but you were counting minutes, right, not percentage savings?

F.:  Of course.

S.:  So one commuter from the rural fringe whose commute is cut from 80 to 70 minutes … that’s exactly as valuable as one inner city traveller whose trip is cut from 15 to 5 minutes?

F.: Well, yes … I’m beginning to see your meaning …

S:  Whereas if you’d thought in percentages …

F.  … we’d have valued the inner city trip more … do you mean? …

S.  Just asking.  Don’t people really perceive travel time changes as percentages?  I mean, who would feel their options to be more transformed, and be more likely to change their behavior as a result: someone who’s travel time was cut from 80 to 70 minutes or someone who’s travel time was cut from 15 to 5?  Wouldn’t the latter be the greater transformation, more likely to change behavior? 

F.  I see your point.  Of course it’s usually easier for a project to cut 80 to 70 …

S.  Of course, so that’s what you end up funding.  What’s the consequence of that?

F.  Well, we tend to score a lot of commuter rail and long-haul busways highly, but it’s harder to assign value to shorter-distance inner-city services like bus lanes, light rail, streetcars.

S.  Because shorter distance services save fewer minutes, though a higher percentage.

F.  And streetcars running in mixed traffic, of course … well, the dirty secret is that they usually don’t cut travel time at all, compared to an “enhanced bus” alternative.  They can even make it longer.

S.  What’s wrong with scoring streetcars low, then?

F.  Well, people are telling us that streetcars in mixed traffic are just intrinsically wonderful, so we should judge them differently.  They seem to encourage economic development, and yet they’re not as expensive to build as faster and more reliable transit systems, so cities see them as something that’s within reach.  Anyway, we have an economic development factor that tries to keep track of that, but it’s really hard to score based on what a bunch of city boosters and developers tell us about how cool a place will be in 10 years.  I mean, we wish them the best, but city boosters and developers are always saying that …

S.  Of course.  No neutral objective measure.  Whereas travel time …

F.  You’re right, travel time, for all its faults, was pretty easy to measure, and to calculate for a new project.

S.  But you’re abandoning it.  So what’s the new scheme?

F.  Ridership!  Who can argue with that?  We care now about how many people are going to ride the thing, especially those who aren’t riding now. 

S.  Is that a new idea?

F.  Well, it’s always mattered somewhat.  In my dad’s day we used to score mostly on “cost per new rider,” so then it was the overwhelming factor.  Then we were accused of not valuing the time of people who were already riding transit – you know – their travel time savings due to the project.  It didn’t count.

S.  So you abandoned that, but now you’re going back to it?

F.  Not exactly, but …

S.  How is the new measure different?

F.  We have some other factors, like service to transit dependents …

S.  But basically, the new measure is ridership?

F.  Right.

S.  And apart from your transit dependent clause, all riders are equally valuable?  Regardless of how far they ride?

F.  Basically. 

S.  So you’re now biased the other way?  Toward the inner city service, which many people ride, and away from the long-distance commute, which serves few people but many passenger-miles, and which will score highest on travel time savings (in minutes, not percentage) because the travel times are so long anyway? 

F.  Yes, but there are lots of arguments that this is the right bias now.  The whole point of sustainable urbanism is to limit sprawl and encourage more compact cities.  When we were mostly building commuter rail all the way to the rural fringe, we were encouraging the opposite.  In fact, I’ve met people who moved from an inner city condo to a two-acre horse farm solely because a new commuter rail line made it possible.

S.  Sounds like the right bias for you, then.  But tell me, isn’t the world changing pretty fast right now?  I caught up on some of your media in the time machine.  It sounds like costs of transportation are shifting rapidly and people in the know expect options to be much different in just a few decades.  In fact, fear about the rising cost and impact of transportation is part of why you want people to live closer together, right?

F.  Absolutely.

S.  Now, when you build something big and expensive like a rail line, you’re not doing it for the benefits tomorrow, right?  You’re doing it for benefits further into the future.

F.  Forty years at least.

S.  Forty years.  So if you’re judging the merit of a project based on ridership, that must mean you know what its ridership will be 40 years from now.  Do you have many studies from 40 years ago that correctly describe ridership today? 

F.  Well, so much can happen in 40 years, you really can’t predict …

S.  But if you expect forty years of value, shouldn’t you at least be looking at the middle of that window, say 20 years out? 

F.  Well, I suppose, but that’s really the outer edge of what anyone can predict.

S.  In any case, you don’t know about your project’s ridership the way you know about its travel time.  You can figure the travel time of a new service pretty exactly, but the ridership … that’s a prediction, right?

F.  Of course.

S.  So your new policy shifts your focus from a fact to a prediction.  Even as you admit that ridership prediction is often wrong on opening day, let alone 20 or 40 years out.

F.  But they always get the order of magnitude right!  And of course things happen that they couldn't have foreseen.  And you know, ridership prediction is always getting better.  Experts are always re-calibrating their models, bringing in new factors. 

S.  What are the calibrations based on?

F.  Well, it’s complicated, and kind of mysterious even to me.  But the basic idea is that they look at the predictive factors, like travel time and land use and user experience so forth, and find examples where similar factors have led to certain ridership outcomes.

S.  In the past.

F.  Well, of course in the past.  What else do we have?

S.  But you just agreed that your world is changing more and more rapidly, which means that a given year is less and less like a year a decade earlier.  Doesn’t that mean, logically, that the past is becoming less relevant?

F.  Well, we try to use the reasonably recent past.

S.  But you need a lot of data points, surely, to calibrate?  And if the world is changing faster, doesn’t that mean that the “reasonably recent past” is shrinking?  I mean, faster change means that conditions ten years in the future are much more different from the present than conditions ten years ago are.  So logically, you can’t look as far into the past as you used to, to calibrate your models.

F.  Well of course it fluctuates.  But over the long run, I see your point. 

S.  So aren’t you approaching a condition where you run out of past?  Reach a point where the only relevant examples are so recent that they’re only just past opening day, and there simply aren't enough data points in so brief a period?

F.  You’re right.  Logically it makes no sense at all.  But what else would we do?

(Pause.)

S.  Well, what’s the purpose of public transit?

F.  Oh that’s easy.  Public transit delivers a range of benefits that all go toward building a stronger, healthier, and more just America.  It is the lifeblood and foundation of cities, which are the engines of the innovation that will keep our country strong and competitive.  Public transit serves the cause of environmental and social justice, helping low-income and minority participate in the life of the city, so that they can climb the ladder of success by their own hard work.  And of course, it’s all about jobs–-

S.  Wait.  That’s a lot of purposes!  How on earth would you measure all of those things?

F.  Well, public transit has lots of benefits!  That’s what makes it so essential to a strong, healthy, and just Amer--

S.  But I asked about purpose, not benefits.  My business, philosophy, has zillions of benefits.  You wouldn’t be here without it, and you certainly wouldn’t be thinking this clearly.  But philosophy’s purpose is not too hard to capture.  Maybe something like “understanding the fundamental nature of existence, and what this may imply for how people should live.”  We philosophers argue about the details, but we’re positively unanimous compared to all the ways you describe transit’s purpose.

F.  Well, we don’t really use the word purpose much.

S.  Tell me, what’s the purpose of the police?

F.  Well, law enforcement of course.

S.  But policing has lots of benefits!  Controlling crime is important for investment, and thus for prosperity.  It contributes directly to quality of life, maybe even to happiness.  And besides, police do good works for all kinds of community causes.  And if you didn’t have police, you wouldn’t have plots for many of the stories that your people find entertaining, from detective novels to forensic dramas!  And admit it, don’t ten year old boys find sirens exciting?

F.  Yes, policing does all those things.  But law enforcement, you know, that’s their real job, isn’t it?  They generate all those benefits simply by doing their job, which is law enforcement.

S.  Exactly.  So it’s not enough to talk about transit’s benefits.  You have to think about its purpose, or as you put it, it’s real job.

F.  Well, moving people …

S.  Anywhere?  Around in circles?  Is a Ferris wheel public transit?

F.  No, I mean to their destinations.  Except for tourists and recreational riders maybe.  They like to go in circles sometimes.

S.  So apart from tourists, transit is about people getting to where they’re going?

F.  Sure, that’s the thing transit does I guess.  And it does it in shared, scheduled vehicles instead of each one driving alone.

S.  Well, we could spend another hour getting down to a definition, but the first thing that comes to your mind is often, in the end, the most useful one.  “Moving people,” you said, “to their destinations.” 

F.  That sounds like a good start.

S.  The destination, of course, isn’t really just a place but an intention, right?  We want to get to work, to home, to school, to a recreation opportunity.

F.  Right.  That’s why cool people are talking about access now, not just mobility.  Mobility is how far you can move, but access is how much useful stuff you can get to quickly.  So transit also has this role of helping things to get built closer together, so that things you need aren’t as far away.  That’s called density, but it doesn’t work without transit, so transit helps to stimulate it.  So I guess that’s a purpose too.

S.  Is that separate purpose of transit?  Or just another benefit?  In other words, can you serve that purpose best just by making it really easy and fast for people to get where they’re going?

F.  Well, the developers and city boosters don’t think so.  They think we need a separate measure to capture the way transit might stimulate development, quite apart from its usefulness in getting you places.

S.  But developers are merchants, right?  They need people to buy their product.

F.  Of course.

S.  So let’s think about their customer.   If you’re deciding whether to live in a transit-oriented place, you’re going to care about the transit, right?  It has to be there.  It has to be good, right?

F.  Right.  That’s why transit effectively stimulates development.

S.  But what does that customer care about, really?  The ability to get where they’re going, right, since that’s transit’s purpose?

F.  Of course.

S.  So even the development output of transit, as you’re describing it, is ultimately about travel time.  How soon you get where you’re going – that’s travel time, right?   That’s the thing about transit that would attract people.

F.  Well yes, but there are so many other emotional factors that affect people’s choices, right?   People just like certain transit technologies, so they use them more.

S.  What, for example?

F.  Well, streetcars, you know, in mixed traffic.  Such a huge political movement.  No travel time benefits at all, really, but this huge emotional response.  Developers just love them, because their customers do.  We figure, by counting ridership, we properly include those factors.

S.  Suppose your Parks agency does some improvements to a park, builds some new attractions there, and as a result more people come.  Does that mean it’s something you should have funded?

F.  Well, no, I mean, we’re a transportation agency.

S.  That’s right.  In fact, I was reading your “Notice of Proposed Rule Making” in the time machine, and noticed it explicitly says that “mobility and accessibility are the primary benefits of transportation investments. 

F.  That’s right.

S.  So if a project is not delivering those benefits, that doesn’t mean it doesn't provide any benefits, right?  It just means it doesn't provide the benefits that your agency is responsible for delivering, so it's not your job to fund it.  It could still be funded by others, even other government agencies, the way a new statue in a park might be.

F.  Yes, this is the argument that we should value mixed-traffic streetcars exactly the way we value brick paving and planter boxes, as amenities whose purpose is to attract investment.  It makes sense, but somehow, because streetcars move, and people can ride them, people insist that we fund them as transit services, even though there's no mobility or access benefit compared to an "enhanced bus" option.

S.  Hmm.  But again, we’re talking about long-term investments, right?

F.  Certainly.

S.  So with your ridership metric, you must show that lots of people will be attracted to a streetcar when you open it, even in the absence of travel time savings, and you do that by effectively citing recent examples where streetcars replaced buses and ridership went up, even though the service wasn’t any faster than before.

F. Right.  That’s a nice example of the problem with judging projects on travel time.

S.  But in addition, because this is a long term investment, you must show that the emotional reaction that is causing this extra ridership is durable over the long term, don’t you?  That people will continue to have that preference for streetcars even when streetcars are no longer a novelty, and even as other technologies improve their ability to do the same things?

F.  Well, of course, nobody can know that.

S.  No, that would certainly be a prediction.  But are some predictions maybe more confident than others, purely on philosophical grounds?

F.  Well, that’s your department, Socrates.

S.  It’s not hard.  Your new evaluation system is based on ridership, and we’ve talked now about two causes of ridership.  One is various emotional attractions of a vehicle, like the streetcars you mentioned, but the other is travel time -- ridership that is attracted because transit gets people where they’re going quickly.  Your models already weigh that, don’t they?  They already assume that travel time is a major indicator of ridership?      

F.  Absolutely, and on very solid grounds.  That’s always been true.

S.  Truer than you think maybe.  If I hire a – well, you might call it a pedicab – to get me across Athens, perhaps because I am late to meeting some friends there, I do it because I’m in a hurry, or more exactly, I want to be at my destination now, because my life is on hold until I do.  The young men who run with those carts go much faster than I can walk.  I get on with my life sooner, and so they get my ridership.

F.  So …

S.  So I can assure you that in my home era, 2500 years ago, people already care about travel time.  Certainly, a time that we consider fast would strike you as slow.  But we want to get to work on our tasks, which require being in certain places.  We want to get home to our families.  We want to see our friends and get a good seat at the theatre.  Our armies want to get to battlefields before their enemies do.  So usually, when we set out on those trips, it’s with a desire to be at the destination, to already be doing whatever we were going to do there.  Of course, sometimes we pause to smell the flowers, and enjoy the trip, and sometimes we walk around just for pleasure.  But most of the time, we need to get there.

F. … and because people have always cared about that, for many centuries, it would seem to have more predictive value!  If we have to predict, we should give more weight to factors that have governed ridership more consistently over longer spans of history … Is that what you’re saying?   

S.  So suppose the project you approve runs for 100 years, as much of your old transit infrastructure has already done …

F.  100 years … Well, I can’t begin to imagine what my great great great grandchildren are going to value when it comes to technology, or even what their choices will be.  But you’re right …  I’m on firmer ground guessing that they’ll want to get where they’re going, and soon.

S.  … which means …

F.  Travel time!  Damn you, Socrates!

S.  So why are you abandoning travel time again?

Pause.

F.  Look, I think there’s a deeper problem with travel time.  It connects with people when they’re thinking about the trips they make, but it doesn’t connect to – well, city builders, you know?  Architects, developers, urban visionaries, and a lot of ordinary citizens who are excited by their ideas.  You even have academics and urban designers saying transit should be slower, to encourage people to not travel as far, as though we could ever do that kind of social engineering.  How can we keep talking about travel time in the face of all that?

S.  Well, then, what’s another way to describe it?

F.  Hmm. 

S.  What do people in your country value?   What motivates them?

F.  Too many things.  You have fresh eyes on it, Socrates, what do you think?

S.  We’re in Washington DC.  Look around, on the monuments.  Or turn on the radio, anywhere in this country it seems.

F.  Liberty, you mean.   Freedom. 

S.  People in most countries value freedom, but nobody talks about it as obsessively as Americans do.

F.  Well, of course.  It was a rallying cry of our revolution, and then of the fight against slavery, and certainly World War II.  Longing for freedom, and then more recently a desire to liberate others, drives so much of our history …

S.  Well, then, why don’t you base your evaluation method on freedom?

F.  You don't mean that freedom boils down to travel time, do you?  That would be a hard line to sell.

S.  But if people can get places faster …

F.  They can get to more places in a given amount of time, so they have more (snaps fingers) … choices! 

(Pause.  S and F look at each other.)

S.  During that infernal time machine ride, I saw some footage of your southwestern cities, which seem to be fleeing from themselves across the desert.  And I noticed the same shop everywhere … a “convenience store” you call it.  They were advertising that customers had a choice of several flavors of something.  But their slogan was, “Americans love the freedom.”

F.  Yes, freedom of choice. 

S.  So faster travel means …

F.  Literally more stuff within reach.  So more choices.  And hence more freedom.  Not just choices of flavors or gas stations or convenience stores.  It means you have more choices of schools for your children, paths for your career.

S.  Those sound like important freedoms, freedoms that people fight for, as we did.

F.  Yes!!  (Pacing.)  You’d have to refine it.  But surely, if you can get where you’re going sooner, that means you can get to more places in a fixed amount of time.  More of the city is available to you – more jobs, friends, places to shop, unusual things that you value.  You can do more of whatever you want to do, which is part of being whoever you want to be.  Sheesh!  Now I sound like the Education Department!  But …  but this is transportation’s place in the same crusade, isn’t it?

S.  Even in my day, people leave small towns for the city, because there are more options there.  Freedom of choice, you’d call it.

F.  So … it’s not travel time, exactly.  It’s more like ...  Yes!  I remember this funny little tool that Walkscore.com created.  (Sitting at desk, typing urgently).  Here it is!  Look here … (rotates the monitor, triumphantly)

 

S.  A map of San Francisco.  And you have a Greek word for those blobs …

F.  “Isochrones,” yes!   We’d never say that word in public, of course, but those blobs show how much of your city you can get to on transit in a given amount of time, depending on where you are.  The idea was to help people see the transit mobility consequences of their choices about where to locate.  You' move the red pointer, and the blobs would show where you can get to quickly if you locate there.  But really … it’s a map of … freedom!

S.  So …

F.   So, what if our metric was:  How much does a project grow these blobs?  Reduce travel time, but specifically with the effect of bringing more choices into range for each person, so they have more freedom!  Not just the freedom to ride your horse in any direction on a ranch, but the freedom to make real choices, about friends, work, values that arise from the options presented by a city!

S.  Grow the blobs in any direction?

F.  Of course not, that would be the old model of mobility.   It would be about access.  Not just square miles of area you can get to, but the amount of stuff in them.  Something like “how many new choices – jobs, shopping, schools, houses of worship or philosophy, sports facilities, and so on, are brought within a given travel time of how many people, just because of this proposed project?” 

S.  One given travel time?  What will it be, 17 minutes? 

F.  (Laughs.)  Imagine getting consensus on that!  Several travel time thresholds of course.  As you pointed out, we care about cutting travel times from 20 to 10 minutes, at least as much as we care about cutting them from 80 to 70 … Or, wait, maybe we care more!  Is there a way to do this with percentages, as you suggested …?

S.  Lots of details to work out, but philosophically …

F.  This isn’t just philosophy, Socrates!  Even better, it’s rhetoric!  “FTA to score transit projects on liberation value!”   “President Obama puts freedom at the center of transportation policy!” 

S.  So why is your agency abandoning travel time as a criterion for selecting projects?

F.  (Sighs.  Collapses in his chair.)  I don’t know, Socrates.  It seemed like the thing to do.  I have to admit I was never comfortable, and I’d love to chase this idea of freedom as the ultimate measure.  But in the end, you know … people really, really love streetcars, even the really slow ones in mixed traffic, and this measure won’t score those very highly!  I mean ... Would people really sacrifice streetcars for freedom?  In America?

 

 

 

 

 

 

 

 

 

 

 

my new consulting practice

I'm now set up in North America as a consulting practice, Jarrett Walker and Associates.  (I'm still an Principal Consultant of MRCagney for work in Australia and New Zealand.)

My consulting website, jarrettwalker.com, is not as flashy as most, but I hope it gives a sense of the ways I can help governments, advocates, and other city-builders to make good decisions about transit, and to bring it into harmony with everything they do.  Have a look!