(Unedited) Podcast Transcript 542: Measuring Transportation System Success

This week we’re joined once again by professor Karel Martens of Technion University. We learn about how transportation engineering is good at finding problems but not solving them, and a new tool to determine the success of transportation systems.

Fair Transport Lab Website

A scale for describing people’s mobility status – Findings Press

Listen to this episode at Streetsblog USA, or find it in our archive.

Below is an AI generated unedited full transcript of the episode.

Jeff Wood: Karel Martens, welcome to the Talking Headways podcast.

Karel Martens: Thanks for having me again.

Jeff Wood: Yeah, thanks for coming back. Yeah, yeah. Before we get started, perhaps you can share a little bit about what you’ve been doing over the last four years.

We talked to you about four years ago. What’s been going on?

Karel Martens: I’ve basically been pushing the same agenda of transport justice, but I moved a little bit away from one way of measuring justice injustice in the domain of transport was very much driven by measuring accessibility to destinations. And I still think that’s a very important way to look at transport justice.

But I’m moved more to empirical measurement of transport problems, transport challenges of people, uh, assume that they can compliment each other, and so that’s been. I’ve been working on that with PhD students, postdocs, and it’s been quite fun.

Jeff Wood: I wanna go back around that time period as well. The book was [00:03:00] published before we even chatted a while before we chatted, but mm-hmm.

What’s been the response to it so far? I mean, for me it was very revolutionary and I still consider it, uh, one of my favorite shows because of the discussion that we had and because of the eye-opening things that you brought to light and talked about. But I’m wondering what the response has been to it overall.

What have people told you? What have you heard from folks thinking about this maybe in a different way?

Karel Martens: People who don’t see it like you, uh, see it maybe will not talk so easily to me. I don’t know. But you’re not the only one who said it’s revolutionary. Uh, I know people, you know, they have the book on their table.

They have all kind of little notes in it. They use it in my, in their work. Um, so in general there’s been very positively received, but you, you always get challenging questions. Of course. One of the typical challenging questions I get is that, you know, you can’t really achieve justice in transport and then you always bring up that one remote person living somewhere in this farm far away.

How can we serve that person properly? There’s always also the issue about personal responsibility, which probably resonates [00:04:00] very strongly in the us I would expect, you know, people choose to live in certain places and so now they expect the society to actually pick up the tap and provide perfect transport when they are not very longer able to drive.

That seems to be very unfair. So there are of course challenges to my perspective. I’m working now with a, a postdoc who joined me about one and a half years ago, who’s challenging me the other way around. He says, you know, you are, you are way too modest. And your perspective will not deliver a truly fair transport system.

And by the way, there are also academics who see this, that think that I’m too moderate in a way that it will still create a class society with some people. You know, having high level of accessibility, not, or just enough. And so we should move beyond enough, or we should define enough, as much more than you have framed it.

So I guess I’m in the middle, which is a good place to be between these two voices.

Jeff Wood: Oh, your middle is my left, I guess.

Karel Martens: I guess for many people. Yeah, I agree. For a lot of people.

Jeff Wood: [00:05:00] Well, I’m super glad you’re here this time. I wanna go deeper into some of these things that, you know, have been pinging around in my mind since we last chatted, and.

Over the last few years, I’ve been thinking a lot about our discussion on the ideas in the book Transport Justice, like you know, the ideas of philosophy like Rawls Veil of Ignorance and Dorkins Desert Island Theory, and how those fit into a transportation framework, as well as your thoughts specifically on sufficiency, the transport sufficiency idea that you have.

But I wanna take a step back even further because I think that many times here in the US specifically, we talk past each other on these subjects of socially beneficial policy because we talk about transport, but actually when we’re talking about it, we’re thinking in terms that are much bigger. And over the last several years, there’s been a few animated debates on things like free transit.

And it kind of feels like a proxy for social policy overall related to the increased cost of living and public wellbeing that other countries have addressed, but we haven’t addressed as well. And so transportation gets kind of lumped in there, or at least, you know, put into that basket and people talk at each other discussing transport, but what they’re really talking about is public wellbeing and things [00:06:00] like that.

With that said, I wanna first ask you what you feel we’re trying to solve with transportation provision. I realize this is like kind of a big question. Mm-hmm. But I think it’s a good place to start. Yeah. Because it gives people an understanding of what the basic definitions you have, you know, and the basis for the discussions that we mm-hmm.

Are going to have in the future.

Karel Martens: I like that question very much ’cause it allows me to make a point that I think people in transport or maybe people around the core of transport, it tends to pull transport in all kind of directions. Uh, having a different transport system is good for your health. Having a different transport system is good for social cohesion and having a a different transport system is good for wellbeing, is good for political participation, and I wouldn’t deny that these links exist.

They do exist, and of course the strongest discourse has been, you know, transport. Is part of the environmental problem, part of climate change, and if we talk transport, we should talk climate change. But in my perspective, all these perspectives, while they’re relevant, they tend to then [00:07:00] obscure the essence of transport, which is something that cannot be replaced.

Anything else except a little bit virtual, uh, interaction. But it’s basically the ability to get to places and the only way to get there is through transport. And so all these side effects are great, but you know, we wouldn’t plan a school in a five story building without an elevator saying, you know, that’s great.

The kids will move with their bodies a little bit more. And so it’s good for their health. No, we’ll design a school. So kids can have good education and whether there will be five stories or will be only one story is a different question. It would be based on consideration of education and probably, you know, the land use context, how dense the city is somewhere.

Space you have and health would be not part of the story except that you want to have a sports field next to your school maybe. And it’s quite bizarre that we, you know, even. Reflect on transport in that way. You know, let’s design it so people walk a little bit more. No, let’s design it. So it’s convenient for people to get there.

Well, preferably by walking, cycling, perfect transport, definitely. And if then we have [00:08:00] health benefits, that’s great, but it shouldn’t be a design principle in my perspective. And so I think on the one hand. Discourse are great because they push the other dimensions, which transport effects. They push the transport discourse away from its traditional perspective on moving traffic, uh, to something much more holistic and more sustainable and, and more people friendly.

But on the other hand, there’s a risk that we, again, forget what transport is all about. Enabling people, all people to get to places.

Jeff Wood: That mirrors the discussion that we had in the two thousands, 2010s here in the United States, which is, you know, all these transit projects were getting more expensive like subways and light rail and things like that.

Mm-hmm. And so we started talking about street cars, but then the streetcar was sold as an economic development program, not necessarily a transport program. And so, you know, it kind of got off track in that respect. And so you can see it in the real world how these things that are important, obviously economic development’s important.

Yeah. And getting people to, you know, exercise, public health, et cetera. And they’re all connected, obviously, but. It also obscures, like you said, [00:09:00] the need for people to access places and to access destinations.

Karel Martens: Yeah, and I think the, the ling of economy is, is traditional, of course very strong in transport and I think allowing people to get to places will generate economic benefits, but I think we should measure what it’s really about and then, you know, the economic benefits will follow and not focus purely on economic benefits.

And so also there, I think it was too narrow perspective.

Jeff Wood: That leads to kind of a discussion about transport problems, right? The problems that we are trying to solve when we’re, we’re building transport or we’re bringing people accessibility. We’re good at solving for things like congestion or delay or things like that, but not necessarily other things like whether somebody needs to ask for a ride.

So I’m curious, like where or when did transport planning start solving for the wrong equation?

Karel Martens: I mean, you said we’re good in solving congestion. I don’t think we’re good in solving congestion. We’re very good in, well,

Jeff Wood: we’re not, we’re not even good at that. Right. We’re

Karel Martens: very good in diagnosing it and thinking of solutions that don’t work in the end.

Jeff Wood: Right, and that’s what I meant, I guess by that, not necessarily solving [00:10:00] it because we’re obviously, we’re not good at solving it.

Karel Martens: No. It’s also a problem you cannot solve in, in a successful city unless you do either congestion, pricing, or duration in, which is even more complex. It’s, it’s the wrong problem in my perspective.

Basically from when the field of transport planning policy develops at the end of the fifties already congestion in the US cities so developed in the US congestion was seen as a major problem and very visible for the eye, and very experienced by the people doing the planning. You know, higher income people, well-trained, uh, professional civil engineers, most of them.

So both seeing the problems, experiencing the problems, thinking these are the problems. And so let’s set up a, uh, transport policy approach, rental planning approach that’s addressing that most important problem. I think it evolves, even going back a little bit further from civil engineering, ’cause most of these people were civil engineers, not trained as transport planners because there was no such.

Discipline in the early fifties or late fifties. And so they came with their engineering [00:11:00] perspective, used to build systems and make sure they work water systems, electricity systems, sewage systems. And, and the goal of of engineers, since every these systems started to develop in the 19th century was basically to connect each and every home to these systems and then make sure they work and.

When everybody is connected to it and the system works means the water is flowing and there’s no blockage in the sewer system, uh, no electricity blackout, the engineer has done its work and that logic of the system and making the system work. Was transposed to transport. And so they looked at a new promising system of the car with the freeway system invented only in the thirties, uh, in Germany.

And you could say the car of course, earlier invented. But this combination was like a very powerful combination, a very appealing in those years as the future of transport. And it’s okay, this is the new technology. The only thing we need to do is connect each and every neighborhood to the road system and those secondary roads.

Main arterial roads and arterial [00:12:00] roads to the freeways, and we just have to keep it moving, just like the water and electricity has to keep moving in the system. And the whole analytical tools that were developed then were basically aimed to assess whether the system would keep moving in the future.

And then we already observed congestion today, but since it. 10, 15, 20 years, maybe in those years, still only 10 years now. It takes a piece, 20 years to build any piece of infrastructure. We have to look in the future. So we have to estimate, not only know what is the problem today, where’s congestion today, but where will it be in the future?

And if we can estimate it, well then we know how to build our way out of it. And so transport modeling was invented basically, which basically helps us to. Estimate trip making in the future. And we can then compare trip making on each link in the network, on major roads, on intersections, in the better cases, also public transport main links assess whether the demands exceed supply and whoever it does well, we have a problem.

It’s very clear. Uh, [00:13:00] red flag. Uh, flows up, uh, level of service A to F, you know, and in those days we were very optimistic. B was already a problem, C for sure, right? We adjusted it over the decades when we realized we cannot achieve a anymore, the free flowing standards. But this whole idea of focusing on the system, making sure it’s flowing, using a model to predict travel in the future.

Narrowed our perspective to this one problem of a system that is not working because it’s not flowing and blinded is for all kind of other issues, basically. I would argue, I mean, it was maybe very logical in those days. Okay. It’s very easy to judge, you know, inside. You know what, what these guys thinking, why did they imagine that every household would’ve a car, like many politicians in, in Europe say, you know, every house of the car also.

Most households have more than one people. What do you mean one car? What does the other person Exactly, but it’s easy to judge from far away. But I think it’s still very interesting that they wanted to solve a problem now rather than directly measuring [00:14:00] a range of problems that might occur. They set up a system that did not tell transfer planners what were the problems, but they just.

Matters. What were people’s behavior today? So they used travel behavior surveys. The first ones were already done in the 1930s and within the 1950s they developed more and more and they became integral part of transport planning. Uh, key input for transport models. And so those travel behavior surveys would ask people, you know, where did you travel to?

If it’s transport means, and what time for what reason? Sometimes with whom did you travel? Sometimes if you took luggage over time they improved. We put these data, this behavior we put in the model, we assume that the behavior in the future will be roughly the same, meaning people with cars will behave in 20 years time, like people today with cars.

Now, we also predict increase in car use, so we know there will be lots increase in trips, and then the model tells us where the problems are. This is very roundabout way to figure out where problems are. You figure out only one problem, but you also a roundabout [00:15:00] way, why not ask directly in your survey, was it easy for you to get to your places?

So we did all this travel behavior surveys, I think, you know, for decades. Still today, the UK does it every second year. Huge survey. Uh, many countries have a mobility panel every year. They follow the same people to see, to track changes in travel behavior, and they never ever ask, was it easy to get there?

Was it unpleasant? Were you scared me? Did you have to wait long? Were you frustrated in traffic or didn’t you mind, did you find parking? None of these questions are asked. We just use the behavior we assume is the result of preference of choice. Put it in the model. Assume that people will have the same choices and preference in the future, and then we tell them what their problems will be and we’re gonna solve those problems and not any other problem.

I think it’s a very. Amazing perspective. I mean, slack, you couldn’t imagine that the doctor would not ask you what is your problem when you comes to the doctor, right? No. He would just put [00:16:00] you in a model and the model will tell you, oh, sorry, your blood pressure is too high, but my tooth aching. You went to the wrong doctor first.

Why you go to the dentist? Uh, anyway, it’s bizarre, isn’t it?

Jeff Wood: But that’s an interesting question. Like what are the problems then like, so if you don’t have a doctor asking you what the problem is, and we’re telling you what your problem is, what are the actual problems that people are experiencing in the transport system?

Karel Martens: So I, I think also in the seventies already, early seventies, but basically in the nineties, academic literature developed measuring much broader travel problems. I would also say that women and trans, with the literature also started addressing already in the eighties, broader problems, women juggling household tasks with.

The jobs and the challenges they had to quickly travel from one place to another. But in the nineties it became more broader, looking at all kind of vulnerable groups or I would say not well served group in the transport system. Analyzing, you know, how they got around, whether they managed to get to places.

The frustration of waiting [00:17:00] on for the bus, the reliability of it, high travel cost among people that own cars. So forced car ship became a topic in the late nineties. So, okay, we have a car that assumes freedom of. Travel, but actually proves that people have to spend lots of money on their car and because of it, they use it very selectively.

Often it falls out of service because it breaks down. People won’t use it for two months because they don’t have the money to repair it. And so all kind of more qualitative insights start to develop about. How difficult it’s for some people to get around how children are dependent on adults. So the travel, behavior service, the more advanced ones would tell, you know, how did you travel?

Well, I traveled with my kids to school and then I went on to do jobs. Was never perceived as a problem, as an issue. That’s okay. You know, children walking less to school was seen as a problem. Again, health is maybe more the important reason, but the lack of dependence that it reflects. Possibly it could also be a matter of choice.

You like to drive. If your kid to school and chat a little bit, that’s fine, but if [00:18:00] it’s the only option you have, it’s problematic. I would argue. So all these from people with cars and the challenges of keeping it running and paying for it to people without cars, the challenges they have to find jobs to go to the doctor.

To go to the supermarket stories like in the UK where supermarkets in the nineties moved out of cities, you know, big supermarkets on the edge of cities and little shops remained in the city. Very expensive that people would walk 20 minutes, 30 minutes to the supermarket to buy cheap stuff there, and there would be rows of taxis waiting there.

For them to go back to their homes after the public transport was completely privatized in the uk. And basically, you know, not much was left of it after, uh, the TE regime. So all this qualitative insights started to emerge, uh, much more attention for social safety issues when traveling at night. So elderly people feeling, uh, concerned about travel, not leaving the house in the evening.

And by the way, also my car. I mean, you know, traveling the evening, parking in, in dark parking lot. [00:19:00] Also, it’s not only a public transport or cycling or walking issue, it, it affects all people, people with disabilities who were always to some extent in the literature, even marginally, but also their experiences became much and more, uh, covered extensively.

I now have a PhD student covering something that wasn’t studied so far. People with anxiety disorders, people that are afraid. Some people are afraid to leave their house at all, but people are afraid to drive over bridges, to drive internals, to drive in busy traffic. People are afraid to get on buses because they’re afraid of other people, or afraid that they cannot get out.

Get out when something happens. So. You know, we are realizing gradually over the past decades that people are very diverse. Not everybody’s a wise, middle aged man, healthy with a reasonable income. Uh, you and me might be like this Jeff, but most people are not. Right. And with this understanding, more and more, mostly qualitative research develops, mapping the challenges of a, a [00:20:00] range of people.

Jeff Wood: Well, that’s the thing is there’s so much data that tells us about time delay and congestion like you were talking about earlier. But it doesn’t tell us about like affordability or, or trip suppression or people who are afraid to ask their friends for a ride. Right, exactly. And so, you know, where do we start to get the data to like start measuring these problems that people have with transport that we haven’t been solving.

Karel Martens: So several teams in in, in the world have been working on this in parallel. Several, I think three in total. Maybe there are a few more I don’t know about. So Graham Curry started to develop a, in Australia, a a kind of a survey as a follow up to a travel, uh, behavior surveys, traditional one asking people about their challenges of in travel.

And Alexandra Murphy also was in your show, developed a transport insecurity index. And we started developing also a survey tool. So these are all survey tools, uh, in a way, very traditional methods. And we were inspired by healthcare research and in health domain there are very, very simple tools that try to capture.

People’s [00:21:00] health status, not what disease they have, but in general, whether they’re very healthy or their health is very poor. And one of the most powerful tools is is called the five D. Five L. It stands for five dimensions and five levels, and basically five dimensions, only. Five questions about people’s health.

So it’s whether you have pain or whether you can walk. If you can take care of yourself. So that’s, uh, they’re asked through a question about dressing yourself, where you can do usual activities like, you know, making a sandwich, uh, cooking for yourself and where you are. Uh, experience anxiety of depression.

These are the five, uh, dimensions, very simple questions. And on each of them, you can have five levels from everything is perfectly fine till you know it’s unbearable. I cannot walk at all. I cannot dress myself, uh, I’m depressed to the ex, uh, extent that I consider, uh, uh, killing myself. Maybe. I don’t think that’s the literal formulation.

I have pain that is unbearable to live with. And so people basically have very kind of a code, right? There’s five, and [00:22:00] you can score 1, 1, 1 and all of them, and then you are the best. And you can score 5, 5, 5, 5 times, and you’re the worst possible state. It is not a tool to figure out how to deal with your health problem, but it’s a very simple questionnaire in which you can track how well the population is doing.

And for instance, Corona Time, it was a very useful tool to figure out how Corona. And the measures taken to, to prevent people from spreading it, how it affected people’s wellbeing. And you could easily track an increase in anxiety and depression because of, it was a very useful tool to track, uh, how youth is doing.

And we see over time that youth is having more and more, uh, struggles with their mental wellbeing. And so there’s very simple tools and, and our vision was like, wow, if we would only have such a simple tool. And we can then map the, the mobility status as we call it, rather than the health status of the population.

It would be great, you know, you can repeat it every year. You can repeat it, uh, after you made a major investment in your transport system, a new red light rail, or a new road, or a closure of a parking lot. You know, even [00:23:00] on very small scale, you could do it. And you could track whether it would affect people negatively.

And you know, ideally you could then ultimately identify neighborhoods where you see concentrations of people, uh, reporting very low levels of, uh, mobility status and say, you know, there’s a problem. It wouldn’t tell you what problem, where this would direct you to particular parts of your city or your region.

Probably where particular population groups who would live my ex patient would be typically. With low incomes, low car ownership, far away from public transport without active transport infrastructure and so forth. When we started to develop the tool, it became. Very quickly, much more complex because where compared to health, where basically, you know, you, you can talk about yourself as the person, and it doesn’t really matter where you’re in San Francisco, you’re sitting in Tel Aviv, you know, you can report about can you dress yourself, we have pain.

Are you depressed? Uh, we both have reasons to be depressed at the moment,

Jeff Wood: unfortunately.

Karel Martens: Yeah, unfortunately, definitely transport is very context dependent. [00:24:00] It very much depends on where you live, what. Mobility tools you have, it depends on your stage in life, uh, what expectations you have. So it became very quickly, very complex to measure it.

In our first survey tool, which we did in 2017, our first survey, we managed to limit it to 12 questions. Uh, in total, uh, not five or 12. Not too bad. Where we said, okay, you can have three types of different transport related problems. Very general. We didn’t want to ask you, are you waiting long for the bus?

Are you frustrated in traffic? General questions. One were related to the trips people did make. So you did make a trip. So in a sense, transport system is successful. That allows you to get to a place, which is what transport is supposed to do, but maybe it wasn’t that pleasant. So we asked questions about how pleasant, difficult it was to get to your place.

We asked you to take too long. Was it too costly? Did it take too much effort? And there’s some kind of obscure question. I, I now think, uh, was it comfortable or uncomfortable kind of trying to capture [00:25:00] all possible other things that people want to report on in the end? It truth is we have a very strong correlation with time as the most important factor for many people.

And the other reason, okay, you did manage to get to your destination, but you couldn’t get there independently. You actually were dependent on somebody else. Ask somebody else to bring you there because there was no other solution. And so we, we tried to map, how often did people ask other people for a ride and who did they ask for a ride?

And the assumption was, the farther away the person is from you personally. The outer ego is farther away, the more severe a problem because asking your partner for a ride. That’s an easy thing to do, asking family members not living. If you is already more difficult. Asking a friend. Living even farther away is really a major barrier.

This was a thought there, and the third was well. Like you said, people actually forgo quite a number of trips and let’s try to measure those. We know, by the way, from travel Behavior survey that [00:26:00] has used this differences in trip frequencies between people and we know there’s of course a random component.

You know, some people with five cars at home and a good income stay at home the entire day. They’re neighbor with also five cars in good income leave the house all the time. There’s this random factor, but it’s very systematic components we see very clearly that travel frequency goes together with income, it goes together with car ownership.

And so actually we could know already that there is maybe suppressed travel, especially among people that have difficulty traveling. But we never analyze the data that way, or we ignored it. We took it as given. We just put them in the model and said, oh, people don’t travel. Wonderful. They won’t have congestion.

We don’t have to improve the situation for them. Uh, anyway, in our survey we try to systematically ask people to report on, on foregone trips. That’s a complex question. Um, I think to report, you know, did you forgo a trip? Which links into another component that makes it difficult to analyze transport.

It’s like, what is the timeframe you ask about travel behavior Survey is typically for [00:27:00] one day, very rarely is for a whole week. Places do this every now and then. Typically for one day only. We assure that transport problems are less frequent than trip making itself, and so we decided to go for three days.

We have now a newer version of our survey. We decided to go for seven days. Alexandra Murphy decided for 30 days. I don’t think there’s a perfect answer yet, but especially trips for Gone, I think you need to have a longer period than just certainly one day and three days was our first proxy and for trips for Gone, we basically asked the same reasons as for making trips.

So was Didn’t you make the trip because it would take you too long? It would be to expand it, it would require too much. You know, walking too far, cycling too far, waiting too long for the bus, which goes together with time of course, and comfort. And we asked whether you could return on the same day, assume that be very relevant for public transport users that maybe would want to go to a party or to an event in the evening and wouldn’t know how to come back.

And so we [00:28:00] decided to skip it. So these were the questions we asked 12 questions. Also, we opened our survey with a general question, how convenient in general is transport for you from one to four scores? And uh, it was very interesting if you ask that question, you basically don’t capture, you don’t capture problems.

The differences between different populations, which are very, very small. Most people report, it’s generally okay. Some people say it’s great, which relates to another challenge of asking about promise. Expectation bias. We try to avoid that by, you know, already typically asking for trips for gone.

It triggers people to think about something they might not think about if you just ask, uh, whether you have a problem. So by probing them about trips for gone, hoping that people would report on these, but the bias in our survey was quite clear, the time question. So people who made the trips, did you make the trip? Were you satisfied with the trip? Uh, we did ask, were you satisfied? Did it take too long? Well, those who have the fastest means of transport reported more that it was too long than [00:29:00] people without cars, which is because we didn’t ask it in the right way, I think.

Anyway, these were the 12 questions.

If I summarize the results, we came to a very interesting conclusion, I think, which was kind of opposite of what we wanted to do, but I think it’s very, uh, useful. ’cause we came to the conclusion that if you cluster all the answers together over the 12 questions and basically have a group that says, well, you know, no, my travel is typically not too long.

Or maybe once in a while I forgot a trip. But most of them, they, no, no. Oh, well, nothing happens. I never rely on others. I don’t forgo trips. And trip is maybe every now and then too long and the rest is all fine. Two third. Two third of the population status there is basically no problem. I am doing fine.

And this was in Tel Aviv in 2017 when congestion in the Tel Aviv metropolitan area was already very severe and so slightly more than quarter said It’s okay, except I’m really suffering from too long travel times. And that’s also affects my comfort in travel. And every time that sometimes it leads to forgoing a trip [00:30:00] and then a third very small cluster, only 6% reported actually.

Consistent problems. So being dependent on others and forgoing trips where I expected maybe, you know, one would solve the other. So I, I ask for rides, so I don’t forgo trips. No. They go together and I ask for rides and sometimes I don’t dare to ask. I don’t have the right to, to get, so I forgo trips. So 6% were of people were really, really suffering.

So this was, of course, first of all, very surprising, the distribution. I, I think partly because Tel Aviv has a reasonably good public transport system, uh, quite dense cities. Local services very close by walking distance. Doctors shops, hairdressers, and you know, many, many services you can really find in most neighborhoods in walking distance.

And so that’s part of the reason that there’s a relatively low group of people with severe problems. But what we, we develop the tool to measure problems and what we developed as a tool to measure success. We didn’t till that [00:31:00] moment have a measure of success in transport. We only have a model that tells us where the problems are.

And since we never ever solved congestion, the problems always were cured. And you could never say, you know, we did something and the situation improved. No, about two years later it was the same. Now my assumption is that if you would systematically apply our tool, you could actually identify if you build a better transport system, it really serves more people better, that a percentage of two thirds will go up and up and up.

Now, I don’t think we will easily get to 90%. You know, we build a system that is very problematic for anybody without a car. In most places in the world. Paris, London, Berlin, Vienna, uh, put aside. But most city regions. Large share of people are struggling and more than 6% I would argue, but we have a tool to measure success, which I think is very attractive for policy makers, which is hopefully what will trigger them to use this tool.

I mean, you, you identify the problems, but you also identify success and you can show your public, you know, yeah, that’s congestion, but actually no two third of you is still [00:32:00] satisfied. And it makes sense, and I tell you why it makes sense because most of your trips are not during congestion hours. Many people do not travel into the city centers.

You know, they travel from suburb to suburb and there’s a little bit congestion, but it’s not too bad. And they have a free parking at their destination next to the workplace and it’s always available and there’s no worries. You know, and then you travel in the evening, you travel in the weekend where there’s little congestion.

So most of people’s trips, if you seriously back three days, are not in congestion. Certainly not in serious congestion as you’re frustrated, and so in a way, the result was not surprising, but it was completely opposite of the discourse in the Tel Aviv metropolitan area. We did repeat the survey again, 2020, end of 2020, beginning of 2021 when I was still Corona.

Pandemic going on, but there was no lockdown in Israel at that time. We actually stopped the survey for a while, when it was shortly a lockdown, and then we reopened the survey when the lockdown was over very short. I think this was our third [00:33:00] lockdown only on the fourth, and we didn’t have so much lockdowns in Israel in general, and we could perfectly.

See that our tool works. So the moment, as you know, you know, traffic is down, congestion goes, people don’t report so many problems. And they did report slightly more dependence on others, which is also, I think, explainable because of Corona. So people work more from home. So the car is at home. So partners chairman could ask more for rides because the car was at home.

It was more easy to take people to places ’cause there was no congestion. People were worried about public transport. So it’s more logical to slightly more rely on rides rather than use public transport. So our tool. Proved to be very robust. It actually managed to capture the improvements that were generated by, uh, the Corona pandemic, at least for car drives.

And there was no change at all for public transport, for people without cars, which I think is a combination of slightly less frequent public transport. Not much in Israel, but higher speeds, more reliability [00:34:00] because of less congestion.

Jeff Wood: So what’s interesting to me about this is if you, for example, let’s use two transport projects that people have built around the world.

So if you think about like the Elizabeth Line in London, right? Yeah. That seems to be a big success. People are riding it. There’s a lot of transit ridership that’s coming from it. Yeah. People feel like there’s more accessibility, et cetera. So if you did the tool, you know, if you ask people the questions before they built that line, you probably get a certain percentage of people that are saying that the system’s successful.

Yeah. And then after that. You would measure, you know, after you built that one big project, then you’d probably see an an uptick in the people that you asked the question. Now, contrast that with something like the 4 0 5 freeway extension in Los Angeles, which once they opened it, it was more congested than it was before.

And so the success of that project was probably less. So the number that you would’ve found if you did that survey before and after would’ve gone down. And so that’s an interesting kind of look at like what. A successful project could do if you looked at, you know, the value of what people are saying before versus the value of what people are saying after.

And has nothing to do [00:35:00] with like the engineering necessarily of that project, the level of service, the things like that. It’s actually how people feel about the end result.

Karel Martens: I think also like with public transport, typically what is being asked is your satisfaction with the new service, which I think is fine to ask.

I don’t think it’s a bad thing to ask. Definitely, uh, useful input, you know, to improve your service in all kind of ways. Better information, for instance, but satisfaction questions. Do not really capture, uh, problems so well. And I think they also suffer more from expectation bias than when you ask questions in a more challenging way to people, to really that force ’em to reflect more seriously on how easy it for them is to get around.

The challenge, of course, is that people. Living around the Elizabeth line, not necessarily use it all the time. So you also have lots of noise of people, you know, there’s no change in their travel situation and their reported travel problems because you know the line is not relevant for them. So you might have want to, uh, combine it with a travel behavior server, you [00:36:00] know, where people are travel, are traveling to have even better results.

I want to challenge you a little bit ’cause people always say, you know, there’s no point to broadening highways. You get the same congestion back. I don’t think we should broaden highways. Definitely not. ’cause car drives have been served well enough over the past 70 years, but. Saying it has no impact and it doesn’t benefit people is a mistake.

I mean, you widen highways and the congestion returns because people change their behavior and they actually move to an option that’s better for them. I don’t say it’s better for the world, but you know, people either change travel time, they left very early, and now they move back to a normal time to leave and, and they recreate congestion with over less hours.

These people are more satisfied because now they can eat actually on the. They prefer to leave or people that were using public transport because it was so frustrating on the roads, they moved back to the car. ’cause now it’s slightly less frustrating and faster and so more convenient. And so you could capture that also with the tool.

Now these are not a argument in favor of widening roads because my [00:37:00] transport justice book, I think very strongly argues why we should improve the situation of people who are currently served most poorly. But it does mean that the tool is very open-minded. It can capture benefits for multiple groups, but for multiple investments.

Jeff Wood: Well, that has a question for me about big projects versus small projects. Right? Like what little tweaks in a neighborhood are different than like a massive project like Elizabeth Line on 4 0 5 Freeway. So it’s a good point. You know, how do you measure that?

Karel Martens: Yeah. Yeah. That will be interesting though. We were talking about Rotterdam before the interview started, and Rotterdam basically started, I think around 2010, with a walking program is, you know, for, for Dutch cities it’s, it’s the most American, Dutch city we have.

Because it was rule in the second World War, nothing left with Old City Center. And we had high rises in the city center, uh, for Dutch standards. And so very unworkable for Dutch standards and, and they actually started investing in war. And I don’t think you will see in two years time in impact, but you could.

Potentially if you ask good questions and gradually over time, [00:38:00] I would expect to see a reduction in reports about travel problems and increase in not in satisfaction, but I don’t measure satisfaction in my tool, and a reduction in reporting of travel problems, a reduction in dependence on others, for instance, children.

Elderly, more independently traveling around and possibly less forgo fit because people feel more safe. It’s more walkable, more people in the street. And so women, elderly feel less threatened and theoretically tool could capture that. This bring me to another point, which I don’t claim that my tool is perfect than we, we developed a new tool with 35 questions now because we felt we’re not capturing everything.

Um, you just got the results in now. What I think is needed is the same effort as we did with travel behavior surveys. The first one in the 1930s. It started to become a standard tool in the late 1950s, and since that moment, we have had a huge community of both practitioners and academics continuously.[00:39:00]

Conducting travel behavior surveys, refining them, improving them, testing them, comparing them across currencies, code across code, assessing new tools to do travel behavior surveys, GPS based, uh, online and so forth. The tools are now more robust than they were. But never perfect. And I think we need the same effort to develop tools to measure travel problems.

Again, want to draw the parallel with health, uh, status research. The tool I was talking about, 5G five L has been developed by eight. Huge team internationally refined. It used to be only three questions and three levels. It moved to five questions and three levels, and then to five questions and five levels, which they felt was more robust, better capture.

The diversity in the population is now used in hundred 20 countries, so we can compare the health situation of people across countries if we can get there as a community. We are going to have a really robust tool. And if you can design [00:40:00] it, that bias, expectation bias is as minimal as possible, it actually becomes very reliable over time.

And we can compare. Are we really building with our 15 minute city with our increasing bicycle powers, with investments in public transport, a better transport system, as many people assume? Is it really true?

Jeff Wood: Hmm. So how does this match with like, the discussions about accessibility and, you know, sufficiency.

That you had in, in your book about transport justice? Like what does this, you know, the, the survey tool and, and the answers that you get out of it, what does that tell you about those measures of success?

Karel Martens: I think they are complementary measures of success. I think, uh, accessibility tools are gradually being introduced in practice, but there’s still a lot of reluctance of doing it, and sometimes they’re framed in such a way that they do not really capture the proper meaning of accessibility.

Sometimes it’s just they use the term accessibility. But I just measuring travel time and travel time reductions, which is a component of accessibility, but not the only one. [00:41:00] Accessibility in its essence is can I get to a range of places Also, if I never go there. What is my freedom to go somewhere given a certain time budget, certain money, budget, maybe certain effort I’m willing to invest, and the more places I can reach within this same budget.

Higher my accessibility. And so the higher my freedom to decide what to do tomorrow, to take another job, to choose a better doctor, to go to a nicer shop, to meet another friend. And that concept, I think is very clear and very powerful. I think people understand the moment you say like accessibility is fundamentally about freedom.

Higher accessibility means that you’re more free to do what you want to do. I think people capture that very easily. It’s not complex, but when you start defining it and how to measure it, operationalizing it, it becomes very complex because. Accessibility to what for whom? Clearly for children and you have accessibility to jobs is not relevant for elderly people if they stop working.

Also not relevant. I take jobs not relevant. If you’re trained as a plumber, plumbing [00:42:00] jobs not relevant if you change as a transport academic. So how do you measure accessibility in a general way? The debates in academic literature are endless. People are trying to outsmart each other, uh, rather than to come up to kind of an agreement of, you know, a workable approach, which I think the engineers were much better in the sixties, you know, agreeing on a transport model, which probably they also felt, you know, it’s not perfect.

Um. Let’s work with it because it’s better than nothing. The more social people tend to be more critical and so have lost debate and don’t get to an agreement on how you to measure accessibility. And there’s also too many questions to ask. 30 minutes, 45 minutes. I don’t know. What’s the proper reasonable travel time and budget is even more complex, should be linked to your income and preferably, yes.

And do we have data on income? I don’t know. Travel. By what mean then how do people combine public transport and car? Different costs, different expenditures. How many transfers in public transport are reasonable? 1, 2, 3. What is a reasonable walking distance to the bus stop? I don’t know. What [00:43:00] are the reasonable walking distance from your parking place to your destination?

There’s endless, pragmatic decisions you have to make when you measure accessibility, but basically the result has been that academics just. Agree to disagree rather than to try to push the practice to adopt one measure and say, you know, it’s a good proxy. It gives you good information about the questions you want to answer, and at least the reluctance, uh, of using it, I think, and also it also vulnerable, the moment, you know, we academics manage to convince the practitioner or the practitioner say, you know, stop talking. I’ll just take this one. You’ll be vulnerable to critique from citizen. You know, you’re measuring accessibility so meaningful. For me. What you’re measuring is doesn’t capture my life. And so I think it should measure accessibility.

It does give you good insights. It does give you. Quick insights on impacts of investments. Uh, also if you can compliment them with travel problems and you can see that there’s a link between the two, I think that becomes more powerful. You can say my accessibility measure is not just an [00:44:00] abstract notion of freedom.

It actually translates into problems if you have low accessibility. So that will be my ideal model to combine them.

Jeff Wood: Is it easier with so much data available these days? I mean, we have cell phone data, you have movement data that’s available and there’s a, there’s a fallacy in that too, is like Michael Batty was on the show talking about his book, the Computable City and is like, we created all these models with all this data.

But you know, as we try to get more complex, we’re just like kind of fooling ourselves that we’re actually gonna solve the problem. But we do have a lot of data and a lot of information out there that maybe can help.

Karel Martens: So the, the drawback of surveys is that they’re expensive to conduct and certainly if you want to do repeatedly, if you want to track your population and the mobility status over time.

Actually, one of my other ambitious plan is actually to use movement data from mobile phones, not to analyze people’s actual movements. I’m not interested in that. We have seen lots of studies, but I think if you analyze. People’s travel patterns in a different way, not being [00:45:00] interested in actually where they go.

You can actually tell whether they can travel with ease. So if people make lots of trips, they travel over long distances, it goes at relatively high speed. They go to very different places over time. They don’t just to make lots of detours. I would say that person is probably traveling with ease and I don’t have to ask them.

Whether it is, I can just, you know, based on observation of a week of data for such a protocol. You know, I think you don’t have any problems. You travel to so many different places at different times of the day. You even do it late in the evening, early morning. You travel over large distances. It’s all our indications of ease of movements.

It’s not black and white. You take somebody else that travels all the time to the same place it over. Short distances, never travels in the evening or the night. It might all be a choice. It might be a person with lots of ease of movement, but I think if you take data over multiple days a week or even more, I think you would start seeing recurring patterns if people [00:46:00] never go anywhere else.

It might suggest constraints rather than choice. And if you then can of course link it to the data you have from all kind of sources on the transport system, ETFS data on the public transport road data and congestion data from other sources. And you see there’s a match and we, we have actually evidence how that impacts people.

And so that’s my. Other line of research where I would invite other researchers to follow up. Can we use such data without surveys, which are very expensive to do, I guess theoretically, freely available data, theoretically def fact often or not rich data on people’s movement to figure out where they can travel with ease.

I now always say, you know, tell me your last three days, your travel pattern, and I can tell you where you can travel with ease. The travel that I meet is only people who can travel with ease most of the time. And then I give lectures, of course, you know, these are higher income people typically. Yeah.

There’s not so many problems, but, but I think it’s really relatively easy to capture.

Jeff Wood: Yeah. Well, so [00:47:00] my last question is, you know, how do we use the solutions that you’ve come up with for, you know, creating success? To think about what projects do we want to build? Like yeah, how do we figure out like what is the right way to spend our money that we are using for transportation?

Obviously now a lot of money goes to road building and transportation that supports cars, but how do we figure out a way to decide using those tools, what we wanna spend our money on?

Karel Martens: I think this is one of the, the big challenges, I think the attractive part of the more traditional way of doing transport planning where you do a travel behavior survey, you put it in a transport model, and it doesn’t only tell you what the problems are by identifying the problems.

The solution is also obvious in most cases, right? Ah, this link is congested. Now, well, what’s a logical solution? Well widen it, right, or improve the interchange where the congestion starts. So there will not be congestion downstream, you know, in a more creative solution. Maybe you can check public transfer, could work on the same corridor, or sometimes maybe a new link.

But the problem is very easy to pinpoint. And all the other approaches, whether you measure [00:48:00] accessibility or you measure travel problems through a survey, or you measure ease of movement through observed travel behavior, you always identify people who are struggling. So in that neighborhood, there’s a lack of accessibility in that neighborhood as a cluster of people that report lots of foregone trips in that neighborhood.

I see clusters of people with a low ease of movement, don’t make much kilometers. Few trips go always to the same places. It doesn’t tell you how to solve the problem. And so I think the next step is double so on one is, is a top down approach, and I think the other is a bottom up approach. So the top down approach is pulling transfer engineers and transfer planners at work.

You know, analyze the transfer system. You see that that neighborhood has low accessibility clusters of people that forgo trips, map how well they’re connected to their surroundings, to places with lots of destination opportunities. Check how frequent the buses are. Check how congested the roads are.

Maybe it’s the roads. It could be the roads. There are always people without cars that suffer even more than those [00:49:00] on congestion. I was argued. So non-carb based solution is always more inclusive than a car based solution. And so the top down approach would start, you know, from the neighborhoods and start mapping.

Okay, we have here in hospital, we have here at the school, we have here. Employment center there. Shopping center. Are these neighborhoods well connected to these opportunities? And compared to what is reasonable travel times, travel speeds, and if not well. You can start improving them. The order is to go into the neighborhoods and say, you know, Hey, we figured out that you have travel problems.

Is this correct? Most likely they will tell you yes. And so what are the places you wanna travel to? You know, what are your key destinations that you would really benefit from or you would improve them And uh, and how we could best improve them. And so that might sometimes be cycling, that might sometimes be public transport.

Sometimes it’s just improving, a walking connection somewhere, a bridge somewhere, or changing a traffic light. And all of a sudden it’s easy to get over a big road. Slowing down traffic, it could be very small and measures that gradually improve the system. So I think these two can go together where you can also maybe check your [00:50:00] top down proposals with bottom up insights to enrich it.

Now, if you want to zoom even more out, there will be many neighborhoods I suspect with people with travel problems. I mean, any neighborhoods. There will be people without cars in US cities for sure. And actually very close to the the city center near. Quite struggling to get anywhere else but to the city center.

And so you would have to start thinking more about systematic investments. And then you really need a advanced analysis of how different investments can and enhance accessibility for people with low accessibility. And I revert back to accessibility measurement and, uh, using tools to measure how certain.

Substantial investment, like a new light rail line or a VOT corridor or simply, you know, putting paint on the roads and creating a, a free bus lane or along a long arterial how they can benefit multiple neighborhoods and connect them to multiple destinations. I think that’s the way to go and I’ve developed a approach, but it’s not finished yet.

So, uh, [00:51:00] maybe you can talk in two years next time.

Jeff Wood: Yeah. Next time.

Karel Martens: Now your, your smart question would be, what was missing in the first version of your survey? Why did you actually develop a new one?

Jeff Wood: What was missing in the first version of your survey?

Karel Martens: I think three key things. Maybe we had questions that were too vulnerable to expectation bias, especially the, the questions about time where travel time was too long, and so we actually got, took inspiration for Alexandra Murphy and her way of asking about travel time and frustration about time.

We ignored. And I, I guess it’s because I’m a man. I ignore travel, safety, uh, concerns about social safety, uh, maybe also traffic safety. I feel also on my bike, even in the crazy television traffic. I feel pretty safe, but it’s not safe. Of course, it’s irrational behavior. So both social safety and traffic safety not captured at all in the survey.

And the third one is like, we only looked at three days and we. Completely missed structural transport problems and the structural transport problems. Actually, I think where I would say the gradual [00:52:00] revolution in thinking about transport started is like realizing that there are people that are so poorly served that they’re being excluded from society because of transport.

Uh, I think you know that you came up in the nineties, in the u uh, late nineties in the uk. Got a policy agenda led to new, uh, accessibility planning at a local level in the uk. It really started to change the agenda and thinking about transport and those questions about structural problems. You know, does transport prevent you from getting a job?

Do you feel you health has been affected because you have difficulty? To go to clinics because of lack of transport. Do you feel you’re isolated because of lack of transport? Those structural things we didn’t ask in the first survey. And so we added a set of six questions, if I’m correct about this, more structural dimensions of problems.

And we did find that between six and 8% of people said, yes, this really affected my life. So it’s, it was something we didn’t capture. So the 6% we found in the first saveway of people reporting foregoing trips. If you have a structural problem, you don’t even forego your trip anymore. It’s not that I [00:53:00] forgo my trip to work, I just don’t have a job because I cannot get to jobs that are relevant for me, and I don’t forgo the trip to my family because I can’t get there.

I didn’t even plan to go on the weekend to my family because I have no clue how to get to the people. And so that questions we forgotten. And so I think. That’s why I also think the 6% of people with serious problems is an underestimate. I think Alexander Murphy came to 10%. I think 10% actually also applies to countries with much better transport systems, which is in a way surprising.

I would say. I would expect the US much more people struggling with transport.

Jeff Wood: Anna VARs, we had her on the show talk about her book. Uh, you know, basically said like there’s 30% of people don’t drive. Whether they they have the ability to drive or not, they don’t drive. And so that number is fairly, uh, spectacular in the sense that that’s a lot of people that don’t have access to the transportation system that we’ve built for everybody.

I

Karel Martens: always repeat that a say it’s one in three. You understand one in three, look around you in your class. How many people don’t it blow people’s

Jeff Wood: minds, right? It

Karel Martens: always sounds like [00:54:00] a man minority, right? But it’s like, but

Jeff Wood: it’s a lot of people. It’s a lot of people. It’s long. Yeah. Well, so where can folks find out more about what you’re doing and reach out to ask questions or anything along those lines?

Yeah,

Karel Martens: they’re very much welcome to reach out. I’m happy to talk, to Share Insights, to share even data. I have a website for my lab, which is one the Fair Transport Lab in the in the ion. We like to say the MIT of Israel. And on my website of the lab, you can find all the ongoing studies. You can find my publications directly into the full versions.

Most papers are freely accessible, they’re open access, and on the website of course, also our contact details are there. And so just drop me a email and you will get a response.

Jeff Wood: Awesome. Well, Carol, thanks for joining us. We really appreciate your time again.

Karel Martens: It was my pleasure. Always wonderful to be here,