With the small talk concluded, Jarrett Walker’s first question to me before we start the interview is whether The Currency belongs in Ireland’s neoliberal or socialist camp.

I was thrown – I hadn’t really thought of it that way before. Was this an American thing?

He went on to explain that BusConnects, Walker’s reimagining of the Dublin Bus system, was the most politically-charged project he’d ever been involved with. 

His projects, he says, are always contentious to some extent. He has designed bus systems all over the world. So he’s used to arguments. But only in Dublin did the project become politically polarised:

“The only thing that has been different for me is having a network plan get caught up in partisan politics. Where political parties were taking positions and making it an example of a larger struggle.”

BusConnects was a messy project, even by the standards of contentious bus network redesigns. It’s fair to say the first draft incensed Dubliners – it inspired an amazing 72,000 objections and suggestions when it came out in 2018. That’s “a different scale” of feedback than Walker had ever gotten on a project.

Walker’s first BusConnects draft was focused maximising the speed at which Dubliners could traverse the city. But to speed up Dublin’s bus system, BusConnects proposed fewer secondary routes and more changes. This angered Dubs — particularly mobility-impaired Dubs — who valued fewer changes and shorter walks to bus stops.

In the early stages of the project, Walker had insisted that only minor changes to his draft could be made, or the network would “fall apart”. But, 72,000 pieces of feedback later, he has softened his stance. “The fact that the final plan reflects public feedback means it’s better by definition, as far as I’m concerned,” he told me.

The final BusConnects plan, incorporating the mountain of feedback, was unveiled in late September. Jarrett Walker has signed it off and it is now in the hands of the National Transport Authority (NTA) for implementation. 

Over a Zoom call from his office in Portland, Oregon, Walker looked back on the project. We talked about what made BusConnects different, how to move people around a city, and how differing political systems produce differing transit outcomes. 

We discuss: 

  • The “politically horrible” task of changing buses
  • Why our system of government is biased against long-term transport projects
  • “Forcing your grandchildren to live by your values”
  • Why the benefit of a redesign “is exactly proportional to the number of people who will be angry about it”
  • The purpose of the city


Sean Keyes: Irish people care a lot about their front gardens. But you’ve designed bus networks in cities all over the world, and you probably have built in your share of front gardens. Has the pushback been different in Dublin than elsewhere?

Jarrett Walker: The only thing that has been different for me is having a network plan get caught up in partisan politics. Where political parties were taking positions and making it an example of a larger struggle. Making it an example of a larger socialist versus neoliberal struggle in which for some reason we got assigned the neoliberal role, even though I wouldn’t identify myself as a neoliberal. It was all very funny.

SK: That leads me to the big question I have about bus transport. We know bus transport is cheap and efficient. But is bus transport weakened by its dependence on buy-in from politicians? By comparison, if a city builds a train line, once the city starts to build the system it’s committed to it. It has to follow through and run it. Whereas bus services can get chipped away year by year, if the political buy-in isn’t there. 

JW: That’s not actually the problem. Changing bus services is very difficult. And changing bus services in a comprehensive way that achieves large-scale benefits for a city is extremely difficult. You can study the politics around the BusConnects network and reach that conclusion yourself. 

Yes, infrastructure locks down the network pattern. But I could point to plenty of examples where infrastructure locked down the wrong network pattern, and became a barrier to providing the most effective and liberating kind of service.

So for example, we have this problem in the United States very dramatically, because there’s a lower level of ambient understanding about public transport. Infrastructure gets built that gets in the way of providing the best public transport service. 

So that becomes a barrier, that becomes a thing we have to work around in trying to design a public transport network that is robust and useful to everyone. I can point to fewer examples in Europe of infrastructure built really badly, in ways that form barriers to good mobility, although there are certainly some. A good way of thinking about it is, infrastructure is a way of forcing your grandchildren to live by your values, even after you’re gone. We are all trapped in our grandparents’ vision of what a good city would be. 

“It’s the fundamental measure of the functionality of a city – the whole point of a city is proximity to opportunity.”

SK: But does a bus network bind your grandchildren less tightly than a rail network?

JW: No, because bus services do not change recklessly and consistently. They change in two ways. They change through major redesign projects like the one we do. And that’s the sort of project you do every 20 years. But even that project doesn’t change the most successful patterns of service within the city. And it mostly works around the edges really. 

And so I want people to recognise that a lot of bus services are at least as permanent as any rail service. There is always going to be lots of frequent bus service coming down Drumcondra road. Because it’s a fantastically strong market. And it is a strong market because of the development pattern. And the permanence of the development pattern is the basis of the permanence of the service. See how that works? 

SK: You mentioned political polarisation as a particular problem for BusConnects. How was it different from other projects you’ve worked on?

JW: The polarisation of the project around the political parties was a significant political obstacle. It meant that people’s view about the plan became not entirely but partly bound up with partisan identity. And I don’t think this necessarily affected the positions that lots of citizens took, but it sure affected the positions that elected officials took. TDs especially, but to some extent, councillors also. And that made the process more difficult. 

The process is always difficult. Because a network design changes things that people are used to. And people get very angry about that – always, and everywhere. And we always have to start with the fact that the amount of new access to opportunity that we’re going to be able to deliver with a network design is exactly proportional to the number of people who will be angry about it. Because that goes to how substantially are we going to consider changing the network to get better outcomes. And that directly governs how many people will be angry because you’ve changed something that they’re used to. And that’s just almost a lot of sort of political physics. 

And so our work, fundamentally, the storytelling work around a network plan is about telling a story about city-wide benefit. That has the potential to outweigh 10,000 stories about individual anger about changes. How that plays with the politics, I can’t control. But my job is to build that story. And to make sure the government has the tools necessary to tell that story.

SK: So you don’t just see your job as the guy in the office with the pad who is drawing the map and moving the buses around. You try to tell a story and convince stakeholders.

JW: I have competitors who think about it that way. But we are about building the story as we build the plan. And they are ultimately the same thing. And by the story, I don’t mean just narrative. I mean, the right facts, the right description of the plan. 

In the first draft of the Dublin network plan, we were able to say the average Dubliner can get to 20 per cent more useful destinations in 30 minutes. And although that was easy for that to get drowned out, because The Irish Times would much rather talk about people, angry people at meetings, it did penetrate among a lot of decision-makers. And that’s ultimately the kind of metric I am trying to plan for when I can. Because I know that that’s not only the basis of how you get lots of people to use this service. But it’s also the fundamental measure of the functionality of a city. Because the whole point of a city is proximity to opportunity. So that’s an example of trying to tell a story about city-wide economic benefit that, to some decision-makers, may actually be worth having 10,000 people angry about.

SK: You had, I think, 72,000 submissions? How does that compare to the amount of pushback that you’ve received in other places?

JW: It was different in scale, for a couple of reasons. In a lot of the places I work in, a smaller share of the population cares about public transport at all. I don’t think any Dubliner can conceivably believe that public transport is unimportant to the life of the city. Whereas in American cities and Canadian cities, even in Australia, New Zealand cities, there’s a percentage of the population that can believe that, and therefore, it just tunes out. 

“Part of my job is to warn elected officials about how politically horrible this is going to be.”

I think the biggest thing was the decision of political parties to take positions and to make this a partisan issue. And also related to that was the decision of a labour union, one of the two labour unions, to seize on it and make it an issue. And that, of course, took the level of conflict to a whole different level, because now people were acting out of partisan identity. And let me just say quickly, I’m not saying that anything went wrong in Dublin. Because I don’t come to this with a feeling of entitlement to have my plans implemented. That’s not my job. My job is ultimately to set the options before people and let the political process figure that out. Now, my plans usually are implemented in some form, after a public process that leads to a lot of improvement. But in terms of the political difficulty, I’ve worked in other cities where there is lots and lots of interest and lots and lots of feedback. The difference, I think, because the opposition was so organised, is that there was more polarisation. 

SK: How many years will it take to implement the plan? And realistically will our political system deliver that plan in that timeframe? If, for example, the plan takes five years and we get a left-wing government in three years.

JW: I defer to the NTA on the question of how long. NTA has a phasing plan. I defer to them on the wisdom of that. 

And that doesn’t mean I think it’s wrong, it just means that I’m not in the position to have all the information necessary to make that decision. And so I’m not going to tell you, I’m not going to second-guess their decision. 

I will say – as somebody who works a lot in the United States, but has also worked in countries governed in the Westminster system, Australia, New Zealand, and Canada – that while there are plenty of things that are wrong with the American way of organising things, one of the challenges of the Westminster system is that typically when government changes, it is easy for the government to throw everything out that the previous government did. I’ve seen that become a real problem. Toronto is a poster child for this kind of behaviour. And it has really set the city back enormously. Because no plan moves forward, because nobody’s in government long enough. 

In Ireland, however, you have a buffer against that, which is the semi-autonomous nature of the NTA. And the NTA has been constructed to a large degree out of recognition of that problem. And out of a widely supported understanding that some sort of set of intentions about transport and commitments to projects about transport need to survive changes of government. 

And for the short time I’ve been watching during this project, I can see that working. The NTA is, of course, ultimately answerable to the voters, but there’s a little bit of a damper on the impulse to just throw everything out and start over. 

So I think you’re fortunate. And I think the NTA structure works well. And a lot of other cities, through trial and error, have evolved that structure. And Toronto is really an interesting city to look at in terms of one that is really struggling because it doesn’t have that structure, because new mayors and new premiers keep blowing up everything that was there before.

SK: Is that something you’ve observed before – where you spend years, you craft a plan that you deliver, you say goodbye, you go back home, and the plan gets shelved because the politics have changed?

JW: It can happen because of political parties. It can also just happen, because the elected officials were not properly prepared for the feedback. Even in the context of the polarisation that we saw around BusConnects, most elected officials seemed to be able to stay the course with it.

But it’s important to be clear, part of my job is to warn elected officials about how politically horrible this is going to be. It’s part of my job to warn elected officials about how many people will be yelling at them. I’m very clear, this is what it’s going to be like. This is who’s going to be yelling at you. And if you want to plan that achieves anything, it’s going to make some people mad. 

“One of the challenges of the Westminster system is that it is easy for the new government to throw everything out the previous government did.”

So tell us clearly, because if your priority is to not have complaints, then let’s not proceed. And we might as well stop now. And so I’m always very clear, as a consultant, I’m always asking that question right at the beginning. And I want that, and I try to get that question asked of elected officials. Because if the definition of success in a network design is that nobody complains – then, you should do nothing.

SK: And you’re telling them that the city won’t see the benefit potentially until after they’ve left office?

JW: Well, that’s a little more complicated. I mean, there’s, there are lots of different ways that implementation can proceed. US agencies tend to implement faster, but of course, you could argue they’re much smaller agencies, it’s this, the stakes are lower, you can’t compare that situation. But it’s also true that the larger-scale benefits don’t appear overnight. And so people have to be used to that. You can’t evaluate the outcomes of a network design for a couple of years, because it takes a while for everyone’s behaviour to respond to it.

SK: You’ve seen our bus lanes, and they’re full of cars. Does your plan depend on better compliance with fully segregated bus lanes? Can it work in the way you intend given our bus lanes?

JW: The question of network design and the question of bus lanes is largely separable. In the context of the infrastructure you have, there is a way of organising the bus services better or worse. There are things we can do to make the organisation of the bus system better in the infrastructure you have. In that sense, the issues are separable. That doesn’t mean bus lanes aren’t critical to getting ultimately the kinds of outcomes you want to public transport. 

Many of my friends, particularly coming from the green space, are really trained in saying everything is connected – you can’t talk about this without talking about that. It’s a very easy thing to say that. But you can’t actually do everything at once. And so to actually do things in a bureaucracy, to actually do things politically, we are always looking for separability. Right? We’re always looking for the ability to say, “yes, this is connected to that. But we can do something about this separately from doing something about that.” So that’s how BusConnects worked. We were working on a track that was about the network design. There was another track that was about the infrastructure. We talk to each other. Does that make sense?

SK: You said at the outset that there was a limit to the amount of changes that could be made to the draft plan. What do you think was lost between the first and the last iteration of the plan?

JW: I don’t think anything was lost. The thing that is not understood about what happened is that the plan we put out was a draft. And the key powerful figures in the committee chose to respond to it as though it were about to happen. And that, therefore, we needed to have a revolt, and people waving signs in the streets and everything else to stop the government from doing this thing that it was going to do. That’s never what it was. It is a normal part of the process to put out a draft. And the purpose of the draft is to elicit feedback so that we can make the draft better. The fact that the final plan reflects public feedback means it’s better by definition, as far as I’m concerned.

“The amount of new access to opportunity that we’re going to be able to deliver with a network design is exactly proportional to the number of people who will be angry about it.”

It’s not my job to decide what good and bad news is. I, as a consultant, go to great pains to not answer questions like “is this better?” I will answer questions like does this serve this goal? That, I can explain. But whether something is better depends on what your goals are. So we started with a plan that was very much geared toward maximum total access to opportunity. In order to do that, we sometimes did things like removing a bus route that came once an hour and wandered through a neighbourhood and pointing out that the way people can get the most places fastest is to ignore that bus route and walk out to the main street where there’s a bus coming every five minutes. 

And so, in that first round, we questioned whether that little bus route was needed. It is important to do that in a first draft, because sometimes it will turn out one of these bus routes wasn’t needed. No-one will complain. On the other hand, where people do come out and say “we need this bus route”, we can put it back. This is the process.