This guest post by Tim Adriaansen, an advocate for accessibility and sustainable transport, originally published on LinkedIn and cross posted here with permission.
People prefer cars, right? Probably not. Here’s why.
Granted, most people in English-speaking countries use a car to get around. This simple reality underpins our transport modelling, planning and investment decisions. Unfortunately, when used in this way, we’re succumbing to a type of bias which means that the numbers behind those transport models, plans and investments are complete nonsense.
Imagine you’re working in an office, and the team leader decides to provide morning tea for everybody. They assign the task to one of your colleagues, who is given the title of “Snack Department”.
The next day, Snack Department turns up at 10am carrying a tray loaded with cupcakes—enough for everybody. They walk around the office and allow each team member to grab one or more deliciously decorated treats. Some people decline: Perhaps they’re counting calories, they’re vegan or they avoid gluten. But all-in-all, 9 out of 10 people in the office grab a cupcake.
After delivering a cupcake to each workstation in the office, Snack Department asks everybody to lift their morning snack high in the air to take a photo.
They send out an email celebrating the success:
“Our office loves cupcakes!” they announce. “9 out of 10 people prefer cupcakes over other types of morning snack!”
Can you see the problem with this statistic?
When a cupcake is served up right in front of someone, they’re very likely to take one.
Cupcakes are tasty and people enjoy them. But to know if people would choose something other than a cupcake, that alternative also needs to be on the serving tray. Without any other options available, and with the cupcakes placed right in front of them, someone’s choice to consume cupcakes is a poor—and misleading—indicator of what economists call ‘revealed preference’.
Transport systems are a lot like serving trays. When people step outside of their home, what transport options are being served up?
Since the 1950s, our transport systems have been designed to make it as easy as possible to drive. Using public money and public land, every aspect of the transport serving tray has been loaded with infrastructure designed to make driving the convenient (and sometimes only) choice.
Providing car parking alongside every new home, shop or office was mandatory for decades. Valuable street space has been made available for storing cars at the expense of any other purpose. Intersections were designed to squeeze through as much motor vehicle traffic as possible—often making them too wide to easily walk across, or too busy to safely cycle through.
By comparison, not nearly as much public transport or bicycle infrastructure has been provided. Few people live within walking distance of fast and frequent bus or rail services. Even fewer live somewhere with a connection to a safe cycling network. For most people, these options aren’t on the tray.
Thinking that people have transport ‘choice’ is like Snack Department handing out cupcakes at people’s desks while telling them “there are also apple slices and carrot sticks available, if you walk downstairs three floors to the cafeteria and purchase them for $5”.
Could we reliably use this offering to decide if people preferred to snack on apple slices over cupcakes?
Our transport system makes cupcakes convenient, while getting your hands on fruit and vegetables requires extra effort. Given what people are being served, it’s absolutely predictable that cupcakes will be much more popular.
This situation is known as the “illusion of choice”, a cognitive bias where people are led to believe that they have more options than they actually do. In other words, people feel like they’re being given a choice between different things, when in actuality their behaviour has been pre-determined by the way those things are presented.
This bias is routinely exploited by manufacturers and marketers of all sorts of products, and in the case of transport, it is used to reinforce “motonormativity”—the idea that cars are the normal, or default, way for people to move around.
As with most cognitive biases, this one can be tough to shake. Almost every discussion about transport systems will involve somebody saying there is a “preference for driving”, when in fact this “preference” is an illusion. Whenever somebody claims that “people love cars!” when talking about the transport patterns of their city or country, they are inadvertently giving away that they have succumbed to the “illusion of choice” bias.
In a transport system that overwhelmingly serves up driving, it’s bad thinking to reach the conclusion that people prefer driving.
This bad thinking is known as a “causal fallacy”: Incorrectly assuming that one thing results from the observation of another. In this instance, somebody may believe that “people prefer driving”, because that is what they see when they look at how people move around their community. This perception is consistently reinforced by car industry marketing, which associates car use with fun, freedom and even strength or masculinity. In reality, people prefer the mode of transport that has been made easiest, and most transport resources have been ploughed into making it easy to drive.
Unfortunately, this fallacy frequently shows up in transport modelling. A transport agency will (historically) serve up widespread driving infrastructure, take a snapshot of people with their cars, and say “90% of our residents choose driving! We need to accommodate this level of driving as our population grows!”
This erroneous thinking—where the model is baselined against a serving tray offering little other than driving—will always produce misleading results. Transport models built on user data from car-centric systems suffer from in-built biases, illogical assumptions and false thinking, and should not be relied upon to estimate the impacts of transport system changes.
This applies when, for example, investment is being considered into cycling or public transport networks. Current rates of cycling on unsafe or incomplete networks, or bus patronage on infrequent routes, cannot provide useful information on how many people may want to use these options if they were to be made available on the same ‘serving tray’ as driving. If it is currently easy to drive but does not feel safe to cycle, the ‘choice’ being provided is an illusion, not a useful piece of data.
We also see this fallacy at work when we consider repurposing on-street parking, for example to install bus-priority lanes. Where on-street car parking (car priority) exists but a bus lane (public transport priority) does not, cars are on the serving tray but catching the bus is ‘three flights down in the cafeteria’. Counting the number of people currently choosing to drive (and park) is misleading, and will not give us useful data about how people will behave when we change what is being served up.
Critically, the “illusion of choice” bias and associated causation fallacy often rears its head when cities are planning how to reduce transport emissions. If false thinking is used to assess the potential for behaviour change, targets may seem unachievable and costly. When planners and decision makers understand how current patterns are a result of what is being served on the tray, and how that offering can be changed by using street space differently, they can more confidently embrace plans to shift car trips to low-pollution modes like walking, cycling or shared transport.
When the Snack Department serves people nothing but cupcakes, then goes on to conclude that most people prefer cupcakes over other snacks, we can see the illusion of choice and associated logical fallacy.
To avoid succumbing to the same problem, transport professionals must measure and evaluate what is being served, rather than focus on what is being consumed.
How easy is it to access different networks (including driving, shared transport and active transport)? How complete are these networks—what can be accessed with each mode? How can we quantify currently observed behaviour based on the availability and quality of these different networks?
Importantly, transport professionals need to be careful that we don’t fall into the “illusion of choice” cognitive trap, and suggest that existing levels of car use are indicative of revealed preference.
Processing...
The Judicial Review on Decisions around the 2021 RLTP was misinformed by an affidavit by an AT staff member who suffers from exactly this problem. He deeply misunderstood the value of the traffic modelling, and declared that because so many people would be driving in the future, to take any road space from driving would result in inequity, poor wellbeing and bad economic outcomes.
Council’s Transport Strategy Team tackled the fallacy in the TERP, but AT have stuck to their guns and continue to waste money on projects that are grossly misinformed by poor applications of traffic modelling, due to this cognitive bias.
It’s at the heart of Auckland’s problems; it’s pushes the costs of projects up, yet ensures they deliver little.
Was his name Simeon Brown ,?that is his mentality as he is owned by fossil fuel and auto industry .
TERP
Transport Emissions Reduction Pathway
It may have been funded if it was a Roadway not a Pathway.
what a wasted opportunity.
And having the big cup cake manufacturers and their ingredient supplies being the biggest advertising group across our media outlets helps maintain cupcakes for all!
When Steven Joyce was the newly appointed Minister of Transport 15 years ago he famously said that because 85% of Aucklanders drove to work then that should mean that 85% of transport spending should be spent on roading improvements. The idea of investing in alternative transport options in order to create a more balanced (less car-dependent) system was anathema. Whle he later softened that view a bit he was the inventer of the RONS concept that is under his protege Simeon Brown central to current transport planning and investment.
Steven Joyce has always been a provincial buffoon.
Simeon Brown said something very similar recently “the census data tells us that 70% of people wake up in the morning and get in their cars …so that means 70% of people want more roads etc etc “ – seriously!
Definitely prefer cup cakes to the alternative. PT would have to be far better than it is now to get people out of their cars. Stop whinging and pour a lot more resources into making sure the motorist is well catered for.
I think you have missed the point of the article.
Except that the train I took to work this morning had limited standing room. There are a lot of us out of our cars already. Which there wouldn’t be were there no train to catch.
While buses drive around West auckland and the North West (not sure about other areas) most of the day empty (or max1-3 passengers), all subsidized by rate payers. This has been going on a long time. Yes sure there may be requirement for them at peak hour but the rest of the time it is chewing up ratepayer funds that would be better spent elsewhere
I’m in West Auckland too , and looking over a huge expanse of land that is full of empty cars that are there for 8hrs or so all subsidised by ratepayers.
the war on single occupancy vehicles does not seem to include buses, go to Westgate between 10 and 2 and count all the buses driving around with no one in them except the driver, (may get a few extra customer over school hols)
One of the drawing elements that encourages use of a public transport network is its convenience. Part of that convenience is the ability to use it throughout the day and that means that there will be times where it is underutilised. It is not appropriate to view those empty or underused buses in isolation, rather the full context should include the service throughout the day and how the network at large serves the area and how regular services – even at non-peak hours – allow people to get around the city.
under utilized is an under statement, you have to see the no. of empty buses for yourself to believe it. Later in the day I doubt it is much better, just stating what I have witnessed on numerous occasions for years. Maybe utilization has more to do with cost and need, people just are not needing access to places like Westgate every 5 mins with the population we have.
Glad we’ve got Thomas’ reckons here. Most big cities run PT all the time, I had many late night tube rides with very few people in a city of what 8 million or so?
Also what about those roads that have no one on them at random hours of the day?
Shouldn’t be spending any money on those…
there is a difference between an nearly empty late night tube ride and buses being empty half the time as is happening in some areas. From memory, one of the buses we were paying for out of our targeted rates was being subsidized $34 per passenger per ride. I was prepared to give the ideology a chance but as my rate bill keeps going up I think about all the things where $ could be saved. Back in the day before everything had to be instant, you coordinated your time around public transport and the buses were full
oh and those nearly empty roads, your buses use them too. I am not opposed to buses, just think it makes fiscal sense to run them with demand in mind. Over supply of a service without patronage just pushes fares up and what rate payers have to subsidize.
also you are straight up never going to encourage or enable people to catch public transport if you have a bare minimum frequency route to “meet demand.”
when people can walk out their door to a bus stop and never have to wait more than 10 minutes at absolute worst or consult a time table, public transport can stack up better against cars for “freedom of going whereever, whenever”. patronage has grown every time headways were improved here in Auckland, ditto for Te Huia, and there’s abundant evidence & research that higher frequency service boosts ridership more than price reduction does.
https://digitalcommons.usf.edu/cgi/viewcontent.cgi?article=1608&context=jpt
https://www.sciencedirect.com/science/article/abs/pii/S2214367X19302911
social needs should come well before economic considerations; unfortunately end stage capitalism is all about hoarding wealth, not investing in infrastructure while subsidizing big oil, big tobacco, big automobile, parasite landlords and middle eastern genocide. i guarantee you there’s more money being wasted to fill the troughs of pigs than there is in running buses at convenient frequencies off-peak; and the culture war being forced upon us by the right makes you think public transport the problem.
And ok, how about presenting some hard stats on patronage and what rate payers are subsidizing these bus services. I do recall seeing some stats in the past and it was not impressive.
your last comment 5.41pm, yes the trough is a much bigger issue (as are all the other things you mention). Maybe fares should be even more subsidized then to get more butts on seats?
yes, and the subsidy should come out of actually fair and proportional taxes on petrol and tolls for congestion. Car dependency is heavily subsidized, far moreso than public transport
https://www.sciencedirect.com/science/article/pii/S2214629620300633
good ideas re subsidy should come out of actually fair and proportional taxes on petrol and tolls for congestion.
Nicolas Reid did an article on what congestion tolls could deliver – an estimated net $200 million per year just from tolling the isthmus & north shore at rush hour only, half of which could go towards infrastructure maintenance and expansion and the other half towards increasing service frequencies. Long story short it would be possible to boost 15-minute frequency public transport to cover 75% of Auckland’s population within a 5 minute walk of the bus stops/train stations.
https://www.greaterauckland.org.nz/2021/09/07/road-pricing-a-boost-for-aucklands-transit/
But if you haven’t yet been offered a variety of alternatives, how would you know?
More?! Is billions upon billions for decades not enough?
“PT would have to be far better than it is now”
Sure; so lets ‘pour a lot more resources into making sure’ Public Transport gets better
Motorists are already well catered for; the vast majority of money spent on transport in Auckland goes to motorists. And we know from every city overseas that ‘mor roads’ is not the answer; we need choice including PT and active modes
“Stop whinging and pour a lot more resources into making sure the motorist is well catered for.”
The good news is we are about to make it much easier to drive at peak. It’s called a congestion tax.
You’re welcome.
Screw motorists.
Great piece Tim.
To misquote a fantastic waiata tamariki…”cupcakes are not a fruit”.
I completely agree with this. What do you see when you catch a bus, a train, or ride a bike? CARS!
What do you see on every roadway…petrol stations, car yards.
What is the coolest part of the LS Onehunga train? That it outpaces the Southern Motorway most of the time.
We are a motor normative, car obsessed nation. When I was fourteen, all I wanted was a car, and my dad gave me half a car when I was fifteen and taught me to drive. Since that car was sold, I have never owned another car, and that was more than two decades ago.
I do not worry about car related things, and there are so many things to worry about with a vehicle.
Yes I rely on buses and trains, and there are people that worry about the mechanics of those machines too, but they are professionals.
And not driving is the best thing ever. Riding a bike is fun, sitting in a train is comfortable…but not driving, not being stressed by traffic, not concentrating to avoid hitting something or someone, all of those things associated with control of a deadly weapon; they do not waste my time.
I only wish that more people could understand that life is better without the bad air, and bad manners, that traffic provides.
bah humbug
I don’t know if you have a family and if they sports or other youth groups, but if you do, how often does someone else provide transport for you kids? To and from practice. To and from games. Away for weekend adventures and activities. If you don’t own a vehicle, for these things you are relying on others with the vehicles you seem to abhor. There has to be a happy medium here.
Would a happy medium mean give walking, riding a bicycle or using public transport the same advantages we have given to the car over many decades?
I really really wish I wouldn’t have to drive my healthy, capable 14 year old to her training and games (luckily, in my part of Auckland, I don’t always have to). When I was a 14 year old in a European city, my parents had very little to do with my everyday activities – I would just get myself to them, or go with friends. Including carrying a full-sized cello on a tram across the city to get to my music lessons.
And actually in my part of Auckland we have reasonably good PT options, but many other parents are so used to driving their children they don’t even know what PT options are available, and don’t realize that advocating for safe cycling would make their children’s lives healthier and more independent (not to mention that parents would have more time and perhaps be more productive themselves…). Which is funny, because I always thought that this is the New Zealand ideal – being strong, fit, and independent.
So true. I think some people need to make more effort to make PT & cycling work for them at least some of the time. It can only help make it better for us all.
This reminds me of the talk-radio-level-of-claims that New Zealand has been massively over-investing into “gold-plated bikeways everywhere” in recent years before National came in and restored sanity.
Ugh. We didn’t even provide enough funding (as a percentage of the overall budget) in the last decade to keep up with existing cycle mode share. But because media and opponents keep blowing up the “issues” of cycleways in the news, some people think we’ve spent money like water on bikeways.
Not helped by the fact that they treat some costly proposals – like NZTA’s bike bridge over the harbour – as if they actually happened, while ignoring all the motorway projects that actually DID happen and on which we DID spend enormous amounts of funds.
Enormous amounts of funds spent with, apparently, little or no quality control at either the design or construction phases.
Since the Waikato Expressway was completed I cannot think of a single occasion when I have been able to drive along it without encountering several kilometres of lane closures for remedial repairs.
Was just going to type the same thing. I always avoid the section north of Hamilton for that reason. Join it at Taupiri or Ohinewai.
> This bad thinking is known as a “causal fallacy”: Incorrectly assuming that one thing results from the observation of another.
This isn’t even a cognitive fallacy anymore. It’s literally the National Party’s media engagement strategy (Appendix A – https://www.tiktok.com/@newzealandnationalparty/video/7412518200635657489)
We need a change of language – “travel forecasting” is what is needed to test future scenarios and make the strategic decisions – the real fruit vs. cupcake vs. “No thanks, I need to diet (= work from home or go somewhere else).”
“Transport modelling” and its one-eyed illegitimate brother “Traffic modelling” are used to compare design options for a pre-selected choice of modes and numbers. They have nothing to do with forecasting, which needs to be carried out first.
Confusing these two very different things by using the same name for them perpetuates the shoddy thinking.
people are still going to need to travel for leisure, shopping, and other activities you can’t do at home. travel forecasting can’t be based entirely on commuting, and there should be efforts to shape urban form and public transport operations so that less people have to drive regardless of whether for work or pleasure.
Great post. What more can I say?
“We’ll need make other options just as easily available before we can evaluate what people will choose to consume.”
Yes we can ignore revealed choice and do what some plonker on LinkedIn wants. Oh wait we impoverished ourselves paying for CRL so we can’t actually do anything.
or we could do what the resident libertarian and live in self-sufficient survival bunkers completely isolated from each other, no social fabric or infrastructure, all in the name of frugality.
oh wait, that’s the future we’re locked into because public transport and environmentally beneficial projects always get unfair scrutiny of their “economics”
“so we can’t actually do anything.”
Looks around at all the motorway projects actually proceeding, or which have proceeded since we started CRL. How sad – they must all be imaginary.
Central government still has money but Auckland Transport are done. They shredded all theirs digging a gold-plated tunnel. Maybe they should have listened to their model.
They shredded theirs using car-based transport planning, which is very expensive.
Yes – Penlink & Eastern Busway jump to mind as massive projects that reinforce and accommodate car dependency. No wonder AT doesn’t have much left over.
This sums up the current government and council .If we dont build bus lanes ,rail tracks and cycle and walking lanes then the voters will only have one choice .
Subversive authoritarianism – the western way.
Great analogy
Can’t believe how good Quay Street looks as a cupcake now vs the old image – like comparing something from a fine patisserie with a stale mass-produced and seal wrapped offering from a gas station. Kudos to the designers and team that made this happen, and thanks Grant Dalton and Peter Burling winning the cup back that was the catalyst for this.
Rest of the inner city streets need to have planting of Quay St quality; it would make the central city hugely more liveable.
Pity “economics”, construction time, and refusal to give up an inch of automobile space will hamper that
Would it be possible to do a survey of how many people contributing to this largely acedemic discussion have actually done any transport modelling in their lives and therefore have some practical experience of what it is about?
Transport model user here: I’m probably a bit late to the party, but in the currently fashionable trend of denouncing transport models as the cause for all that’s wrong in our transport system, this one is downright bizarre.
After the mandatory photo of gridlocked traffic and multi-lane motorways that apparently models are responsible for (talk about false cause!) and a discussion on the fallacy of government mostly catering to car drivers because most people drive, and I 100% agree with the author here, things take a turn.
Models suddenly get lumped in the argument, but I cannot see the author actually make any point that explains what models have to do with this problem. Indeed the issues seem to be with counting rates of cycling, or bus patronage, or counting the number of people currently choosing to drive, with apparently none of it being a useful source of data.
I think I sort of get what he’s trying to get to, which is that using a snapshot of the mostly car-based status quo to justify investment in underdeveloped modes can only lead to underestimation of benefits, something along those lines.
Seems to me then that transport models of the kind that can estimate potential redistribution of trips and shift between modes are exactly the tool you’d need then. Yes these are calibrated against current conditions, they have to be as they must be capable of representing the current conditions sufficiently well to get a check mark, but their whole purpose is to then allow testing different scenarios, including investment in other modes.
Needless to say models come with their limitations and there are whole arguments to be had about their use and misuse, what forms of models are best, etc. But the current trend of blaming them for all the ills of the transport system is troubling.
The same models that are used for assessing road infrastructure now were also used under the previous government for PT infrastructure and emissions reduction (with some arguments then about their limitations for representing active modes or large shift to PT, and these are worth discussing). So modellers are not exactly a shadowy network of motorway enthusiasts, but in my experience people who are quite passionate about improving the transport system, and themselves much more likely to be PT and active mode users than the general population. And nowhere near as powerful as some people tend to think.
I’ll finish by using the conclusion from the author who does not seem to realise that he makes a very good argument for what transport models can be user for:
“To avoid succumbing to the same problem, transport professionals must measure and evaluate what is being served, rather than focus on what is being consumed.
How easy is it to access different networks (including driving, shared transport and active transport)? How complete are these networks—what can be accessed with each mode? How can we quantify currently observed behaviour based on the availability and quality of these different networks?”
Could not have said it better myself.
So seems you almost in agreement with the post author. They are not really calibrated to predict other non-car modes properly (especially in NZ perhaps). I’m sure you do a great job at what you are doing, just the tools or the person’s in charge telling you how to use it.
This article is a great description of motornormativity in the real world, but not so great on transport modelling.
I think what’s really being said here is that for people to not choose the cake (car) every time, there needs to be an equal offering of fresh, tasty and free fruit, all on the same plate. Or alternatively make the cakes a bit stale, sit them on a high shelf and charge to use the steps.
Without models, it’s easy to say “we’ve done enough, the options are there”. With a model you can demonstrate clearly that the choices are NOT balanced, and show what you need to do.
(Yes, bad ways to use models also exist)
“Transport models built on user data from car-centric systems suffer from in-built biases, illogical assumptions and false thinking, and should not be relied upon to estimate the impacts of transport system changes.”
It would have been helpful to explicitly support this statement with specific examples or evidence. What are the inbuilt biases? What are the illogical assumptions? What is the false thinking? In the absence of examples, it’s hard to engage with the article’s core claims. That’s not to say they’re incorrect, just that they’re not supported by evidence contained in the article.
If the claim is simply that people’s true preferences over transport outcomes are latent, or unobserved, then I think 100% of modellers would agree with you. The empirical question then becomes one of how we can uncover people’s true preferences, which is the topic of a lot of research that often seeks to blend revealed preferences with stated preferences.
Or, uses information on preferences from other places that have access to transport options that are not available elsewhere.
One thing to add that I thought about afterwards: I think the article would benefit from engaging with the variation in transport outcomes that we observe within Auckland and how that compares to other places.
Like, even though Auckland has high-ish car mode share on average, there are many locations (e.g. central suburbs) where we observe high shares for active and public transport modes.
As the behavioural parameters in transport models exploit differences in transport outcomes between locations (and to a lesser extent over time) the key question is whether there is sufficient variation in transport outcomes within Auckland to estimate the parameters reliably.
That would be interesting to explore.
Thanks for the post – i realise its moment has passed, but im documenting this here anyway.
AT consulted with my community re pedestrian crossing upgrade. New raised crossing went in and its brilliant – working perfectly, that 1 in 50 crazy driver now slows down around the kids.
The kicker, the crossing goes to across the road where there is no houses, every kid crossing has to cross another very busy confused road. Every submission said, Make McKenzie road safe.
The follow up on the consultation had the following confusing no demand with too dangerous to use.
Why now, we just had a very near miss with a school kid on this stretch of road. His quick thinking kept him out of hospital or worse when a car stopped and waved him across, and a following car pulled out and overtook for unknown reasons.
AT – Response follows
Request to widen the footpath or install a barrier on the footpath on Mackenzie Road. Concerns that the existing footpath is too narrow, dangerous on the bend and runs parallel to the road with no safety measures separating vehicles and pedestrians.
Our pedestrian survey data indicates that the pedestrian demand crossing Mackenzie Road is very low (two pedestrians during the peak hour 2:30pm-3:30pm). Therefore, we are unable to justify footpath improvements at this location at this stage.