Will driverless cars make things better, or worse?

There continues to be a lot of hype and excitement around driverless cars, with the first vehicles hitting roads in Britain recently and the NZ Herald running an opinion piece by Paul Minett earlier this week that was generally good, although perhaps a bit excitable about the need to stop all current investments in roads and public transport.

One of the big promises of driverless cars is that they will significantly reduce congestion, as their computer-controlled driving will enable much closer following distances between vehicles, alongside much more efficient operation of intersections. But how will this play out in practice? One of the most detailed pieces of analysis was undertaken by the International Transport Forum (part of the OECD), which modelled in quite a lot of detail what might happen under different scenarios involving the uptake of driverless cars.

Two types of “driverless vehicle” were analysed:

  • Taxibots – self-driving cars that can be shared simultaneously by several passengers
  • Autovots – self-driving vehicles that pick-up and drop-off single passengers sequentially

The analysis used Lisbon, Portugal as the case study city for the analysis. The different scenarios also looked at whether high-capacity public transport would be available or not, as well as how things would work at 50% and 100% penetration levels of these new vehicles. Some of the results of the analysis are pretty interesting.

Firstly, looking at mode-share, in scenarios where high-capacity public transport is retained the driverless vehicles actuatlly result in an increase in PT mode share, although it seems that they replace all “not high-capacity” PT. This makes a lot of sense, driverless vehicles could make for great first/last mile solutions and for replacing those routes that wind through the suburbs designed primarily to provide coverage. Interestingly walking & cycling mode share is projected to decline from 18% in the baseline scenario to 8% with the new vehicles.


Next, if we look at fleet-size, the projections are pretty sensitive to the different scenarios – varying from a situation where nearly 90% of the private vehicle fleet is no longer required, to other situations where there would actually be more vehicles than the baseline. Once again the existence of high-capacity PT seems to make a big difference to the totals, as does the level of penetration (it seems that most people are expected to hold onto their private vehicles until there’s very high penetration).


Perhaps the most interesting finding relates to projected overall traffic volumes, which increase under all the modelled scenarios (although to very different extents). Scenarios without high-capacity public transport are projected to see substantial increases in car kilometres travelled, from both modal shift away from PT and also the empty “re-positioning” trips taken by the vehicles.


The study highlights that while scenarios with slight increases in travel would be manageable (due to the vehicles themselves being able to travel more efficiently), scenarios with much higher increases are not likely to be manageable at all. Some further detail is provided about the extent of travel increase at different times of the day:


The most interesting trend in the above graph is that the “AutoVots without high-capacity PT” scenario’s greatest increase in vehicle km occurs at peak times, which would be when the transport system is least likely to be able to cope with such an increase. Furthermore, the greatest level of travel increase seems to be on local roads (not motorways), which is probably where we would least want it to happen:


The study then looked a bit closer at where, under the “TaxiBot plus high-capacity PT” scenario, travel increased or decreased. Obviously this would vary depending on the city, but it is interesting to see that most increases are in more peripheral areas rather than central areas. The study itself also highlights that volumes stayed constant or declined on major routes and bottlenecks, with increases mainly confined to local networks (presumably for more local trips?)


Finally, scenarios with full vehicle penetration saw significant reductions in the number of parked vehicles, although once again the reduction was far lower at 50% penetration and actually increased in a couple of scenarios:


There are a few key takeouts from this study that are really important to keep in mind when it comes to discussing driverless cars and how they might change the transport system in the future:

  1. High capacity public transport remains crucial. Scenarios without high capacity PT saw really big increases in travel demand, especially at peak times. We can rest easy that our current and future rapid transit network investments will continue to provide value in the future – even with a gigantic shift to driverless vehicles.
  2. Ride-sharing and car-sharing results in very different outcomes. A system based around “car-sharing”, where the driverless vehicles are for individuals, results in a huge amount of travel and large number of re-positioning trips. It also needs a much larger vehicle fleet than ride-sharing.
  3.  All driverless vehicle future suggest a massive reduction in the amount of land required to park vehicles. This could be truly transformational for our urban areas as this land can be repurposed into housing, businesses or open space.

The big take-away though is to note that the introduction of driverless vehicles could play out in a variety of different ways in the future. Some could be really good, others disastrous. It’s pretty important that we get it right.

Driverless cars, sprawl and density

This is a guest post from reader Frank McRae

Google Self Driving Car

The emergence of driverless vehicle technology has created much excitement, and speculation about how these vehicles will affect the development of cities. A recent article in the Wall Street Journal claimed that a major consequence of driverless vehicles will be the outward sprawl of cities (Driverless cars to fuel suburban sprawl):

Here is the weirdest thing about this hypothetical future: where you live….you will be able to escape your cramped apartment in the city for a bigger spread farther away, offering more peace and quiet, and better schools for the children.

..there is something akin to a law of nature about new transportation technology: The faster humans move, the bigger and more sprawling our cities become.

While it is true that lower transport costs and faster travel speeds will generally incentivise the outward spread of the city there are other forces at play. I want to put forward the case that the outward spread of cities will not be an inevitable feature of driverless cars and that these vehicles can complement the ongoing intensification of cities.

Driverless cars are not viable without road pricing

Driverless technology will remove the labour cost of driving, and electric vehicle technology will significantly reduce the cost of fuel. While we don’t currently have road pricing, fuel and labour impose costs on the user that practically limit the amount of time their vehicle can be on the road. If there were driverless vehicles but no road pricing then the vehicles could be left on the road at almost no cost. Indeed if parking is priced, and road space for moving vehicles is not, then leaving vehicles circling the block with no occupant would be the rational thing to do. What would be the point of paying for parking or storage when empty vehicles could be left roaming the streets for free? It’s easy to imagine roads quickly descending into gridlock when the cost of leaving a vehicle on the road is so low. Road pricing would be a simple way to clear the roads of circling drive-bots.

While road pricing is a difficult political sell, the politics may shift when the alternative to pricing is completely dysfunctional roads. People generally don’t like paying for parking either, but it is politically palatable to charge for parking in centres when the alternative is the unavailability of parking spaces. Road pricing may become politically palatable when the alternative is roads gridlocked with autonomous vehicles.

Driverless cars will also improve the viability of road pricing by making it practically easy to calculate the distance travelled and to implement time of day charging. And driverless vehicles would remove the privacy case against GPS based road pricing as, for better or worse, users of on-demand driverless vehicles will already be giving up their privacy to the service provider.

Finally, electric powering will require a new source of funding to replace the fuel excise which goes towards funding the maintenance and upgrade of roads. Distance based road pricing can replace the fuel excise in a way that provides a much better link between funding and demand for infrastructure.

This distance based road pricing will provide a disincentive against living in outer suburbs.

Driverless cars will disrupt car ownership

Driverless cars will mostly be used through on-demand services. Uber has already provided the model for this, and it will not be a huge stretch to extend Uber’s model to driverless vehicles. Indeed Uber is already a major investor in driverless technology and is launching an autonomous taxi service in the US city of Pittsburgh.

If it is possible to get an affordable ride on-demand then why would anyone bother with the storage, insurance, and maintenance costs of car ownership? It seems likely that a major effect of the driverless revolution will be the end of car ownership for the majority of people. This disruption of car ownership will significantly reduce the need for car parking spaces. This has significant implications for the development of the city.

Driverless cars will remove the parking and traffic constraints on dense development

The need to accommodate parking sets design limitations on development, and minimum parking requirements create a regulatory barrier to intensifying housing where demand is highest. These limitations can reduce the financial viability of intensified developments. If on-demand driverless vehicles disrupt car ownership, development can be freed from these constraints.

Additionally, a major source of “community” objection to development is the effect of new dwellings on local parking availability and congestion. For example, a recently proposed development for an apartment tower in Glen Eden was opposed by the Local Board because of the potential traffic generation.

This is just one example but almost every development in Auckland (and elsewhere) is objected to on the grounds of traffic and parking. Also parking and congestion are a significant justification for the planning rules that limit density in the first place.

On-demand driverless vehicles will remove the real and perceived constraint that parking places on development. And the increased efficiency of driverless vehicles combined with road pricing will undermine using traffic as a reason for limiting and objecting to intensive development.

Electric vehicles will improve the amenity of central suburbs 

A major drawback of living centrally, and at density, in a car dominated city like Auckland is the air quality and noise disamenity caused by cars. The electric technology used in driverless vehicles will remove these problems making inner city suburbs a more pleasant place to live.

On-demand ride services will be better in the inner city than outer suburbs

Driverless technology will not change the fact that trips to outer suburbs will take longer and be more expensive than those in the inner city. And while passengers will be freed from the burden of driving themselves, driverless cars are unlikely to change people’s ultimate tolerance for being stuck in a vehicle for more than an hour.

Though it is difficult to predict what a ride in a taxi-bot will cost, an article in Bloomberg suggests that the average cost could be as low as 44 (US) cents per mile (1.6km). But the cost per kilometre could be much higher in outer suburbs to reflect the reduced likelihood of the vehicle picking up a return fare. So outer suburban travellers will not only have to pay a higher fare to reflect the greater distance, they may also have to pay for it at a higher rate.

The service is also likely to be of a lower quality in outer suburbs, with longer wait times due to the lower density of potential passengers. Higher density inner suburbs will have a larger pool of potential passengers and hence shorter wait times for a ride.


Trying to make predictions about an unpredictable future can be a foolish task, and many of the predictions made about driverless cars have been foolish indeed (Dump the cycleways – how driverless cars will save the world). But the impact driverless cars will have on the development of the city is not inevitable. As always, much of this depends on the policy settings we adopt. Driverless cars will not necessarily accelerate exurban sprawl and with the right policies there is plenty to suggest that these vehicles can complement the intensification of cities well.

Doomsday Driverless Scenarios

Driverless cars are presented by some as a utopian solution that will solve our transport problems. I’m not convinced that the changes driverless cars will bring about will happen as fast or be as vast as those most enthusiastic about the tech. This is not to say driverless cars won’t happen or won’t have an impact, in fact I think some areas such as taxis will see considerable change.

Some of the changes might be great either. For example one prospect is that driverless cars will be zipping about in all directions meaning roads are busier off peak than they are now when cars are all parked up.

Google Self Driving Car

Twitter user Queen Anne Greenways from Seattle created a list of over 50 different doomsday scenarios that could occur if driverless cars eventuate. While they’re a bit of fun some also seem scarcely accurate.

  1. To save money on parking, people let their cars circle the block downtown all day.
  2. Unexpected re-boots kill thousands.
  3. Rich people buy dozens of cars and send them to run errands all day.
  4. Instead of dispersed crashes like we have now, systemic software glitches cause thousands at once.
  5. Terrorists hack every Prius, turning them into killing machines.
  6. Computers gain self-awareness. Self-driving cars make perfect hosts.
  7. Unfettered by safety concerns, cars become massive rolling entertainment capsules.
  8. Governments spend billions on research and infrastructure but the benefits never materialize.
  9. Auto traffic becomes more like *internet* traffic: mostly unnecessary and 40% porn-related.
  10. Self-driving doesn’t mean self-fueling. Stalled, driverless vehicles litter the roadways.
  11. Self-driving doesn’t mean self-repairing. Poorly maintained equipment still kills thousands.
  12. Everything works great until a sensor wire comes loose.
  13. Braking distance is the same whether the car is self-driving or not. Peds learn the hard way.
  14. Peds crossing everywhere causes permanent gridlock, which is “fixed” by restrictive ped laws.
  15. Stopping all traffic is as simple as placing a lifelike doll at a few intersections.
  16. Now that you can sleep in your car, a three-hour SOV commute becomes feasible.
  17. Long, private, pleasant SOV commutes make sprawl more attractive.
  18. Traffic deaths drop to 0. Yay! So, no reason to address pollution, noise or ugly streetscapes, ever.
  19. Efficiencies gained by “bunching” induce demand that completely offsets all efficiencies.
  20. Self-driving delivery requires menacing robotic solution to the “curb-to-door problem.”
  21. Human-driven cars quickly outwit the automated ones, completely neutralizing their benefits.
  22. Only the worst drivers keep their human-driven car.
  23. Enthusiasm for self-driving cars fuels outcry for more dedicated car space.
  24. They divert investment away from things we already know work. Like transit.
  25. Instead of mailing parcels across town, everyone just sends cars there and back.
  26. Carmakers develop new horns to make loud, complex announcements: “I’m here to pick you up, Janice!”
  27. They become the ideal platform for rolling billboards, which become cheaper and proliferate.
  28. Driverless but not carbon-free. The mechanism causing climate change has simply been automated.
  29. Two words: Road drones.
  30. Instead of taking shape of regular car seats, American butts take shape of *self-driving* car seats.
  31. Chevy unironically unveils model equipped with stationary bike so you can exercise while you commute.
  32. An hour on hold with tech support figuring out how to park in a field at the fairgrounds.
  33. Interaction with emergency vehicles proves problematic. Police given universal kill switch.
  34. Volkswagen decides to make one.
  35. To save money on parking, cars commute twice for each person, once at 8:00am and once at 5:00pm.
  36. Car manufacturers work hard to write algorithms that outwit competing algorithms.
  37. US government demands “back door” into all car sensor data. Opens new data mining possibilities.
  38. America’s “love affair” with the car is rekindled for yet another century.
  39. Two words: Apple Maps.
  40. Almost nobody buys one, so I just wasted 40 tweets.
  41. people stop buying houses and buy self driving RVs which endlessly circle city centre. (this one was by user @CarpenterBernie)
  42. Occupant-activated “emergency mode” lets cars drive in protected bike lanes, on sidewalks.
  43. All taxi, bus, delivery and long-haul truck drivers lose their jobs. Executives all get bonuses.
  44. Private cars are driverless yet, for some reason, buses and metro rail still need operators.
  45. Cities stop building bike infrastructure in anticipation of impending driverless utopia.
  46. Vehicular cyclists finally win (and they gloat about it to the rest of us).
  47. Cyclists banned from the road to prevent “system inefficiencies”.
  48. Without threat of death, cars must now honk continuously to get bikes and peds out of the way.
  49. Google and carmakers lobby Congress to severely limit their liability.
  50. Government adopts “standard algorithm.” Tea Party-type groups protest for more speed, less safety.
  51. Consumers prefer cars with algorithms that prioritize occupant safety, not bike and ped safety.
  52. The Self-Driving Indianapolis 500.
  53. New kind of political protest: the “vehicular DoS” attack. Sending 1,000 cars to the exact same spot. (as opposed to the analog version we currently have).
  54. Google writes the perfect algorithm but fails to consider the ways people will use it in the wild.
  55. Google achieves six sigma reliability: 1.5 trillion US trips * 0.00034% rate = 510 million “defects.”These ones I added
  56. To help pay for your trip you’ll be forced to watch in car advertising for the length of your journey.
  57. Prior to each trip you’ll have to watch to an airplane style safety video even if you’ve used that car before.
  58. You’ll have to click a button accepting a 1,000+ page EULA before the car will move.
  59. The term blue screen of death takes on a whole new meaning.


I could actually see quite a few of these becoming reality, what do you think?

The Ministry of Transport and Driverless Cars

Automated, autonomous, driverless or self-driving – whatever you want to call them cars that can drive without the interaction of a human are increasingly talked about as the next big thing in transport and the Ministry has recently highlighted what’s happening in New Zealand around them.

They say that while the talk going around is about fully autonomous vehicles, there are actually different levels of it and increasing levels of technology are already in some cars. A good description of the different technology levels comes from the United States National Highway Traffic Safety Administration which breaks things down to five levels as shown below. It’s noted that other jurisdictions have similar classifications.

Levels of vehicle automation identified by NHTSA

As for actually implementing driverless cars they note that they will pose a number of challenges although one advantage is they likely mean there is no need to change existing roads. Instead the challenges are more likely to be policy/requirement related

The Ministry of Transport has a work programme to clarify the current legal situation that applies to the deployment of autonomous vehicles in New Zealand. Section 4.14 (page 25) of the government’s Intelligent Transport Systems (ITS) Technology Action Plan [PDF, 431 KB]specifically relates to autonomous vehicles. This includes the following action:

Government actions to promote New Zealand internationally as a test-bed for new technologies

The Ministry of Transport, in conjunction with the NZ Transport Agency, will review transport legislation to clarify the legality of testing driverless cars in New Zealand. This will specifically consider the issues of liability associated with testing, but will not consider liability for general use.

The work programme has a range of possible outcomes – one being a law to set requirements for driverless vehicles. However, there are no immediate plans to do this. The relevant excerpt from Section 4.14 of the ITS Action Plan reads:

Internationally there is a great deal of thought being given to what laws will be necessary for the general operation of driverless vehicles. Their widespread operation will pose complex legal challenges, especially to determine liability in the event of any accident. It is not proposed that the New Zealand government will explicitly look at these legal issues at this time. Rather, the government will continue to monitor international developments and draw on this knowledge once international thinking has developed further and it is clearer if or when these vehicles will be commercially available.

One interesting thing I’ve been noticing with the driverless car debate is that many countries, states and cities all seem to be falling over themselves to be the test-bed for these new technologies – presumably in the hope that some of the big players will turn up invest money. The MoT say they have started to look at the legal issues associated with testing these vehicles. However they also say there are “no obvious legal barriers” towards testing driverless cars as “NZ Law has no explicit requirement in our laws for a driver to present”. Despite this the MoT say there have been no formal requests to test driverless cars on our roads.


While the MoT seem to be mostly thinking about how the legal issues of testing driverless cars, perhaps they should also start to have a think about other associated issues too. Some of these are highlighted quite well in a few pieces I’ve read recently.

The Ethical issues

There’s a greater issue that just liability should an incident occur but a debate that needs be decided about how driverless cars deal with accidents in the first place. This article highlights the issue well

He soon came to see both its significance and its painful complexity. For example, when an accident is unavoidable, should a driverless car be programmed to aim for the smallest object to protect its occupant? What if that object turns out to be a baby stroller? If a car must choose between hitting a group of pedestrians and risking the life of its occupant, what is the moral choice? Does it owe its occupant more than it owes others?

When human drivers face impossible dilemmas, choices are made in the heat of the moment and can be forgiven. But if a machine can be programmed to make the choice, what should it be?

Coming up with an answer to these issues could have wide ranging implications for society – but then again we’ve changed society in the past in the way we allowed vehicles to dominate our cities. The implication of driverless cars on cities is the next point

Are we solving the wrong problem?

A great piece from Peter Norton, the author of Fighting Traffic highlights that driverless cars could be fantastic and fix many of the issues we associate with our auto-dominated society – which he says came about because we focused on how best to move cars and not what is best for people or the city. Unfortunately he says that if we don’t address the issue of what we want our cities to be then driverless cars could actually make our auto-dependency worse.

In autonomous vehicles and other intelligent transportation systems, we may have a solution so powerful that we fail to pause and ask what problem such systems are best suited to solving. We may fail to ask whether the problem formulation we inherited is the right one.

If engineers continue to seek to accommodate all of motorists demands, and f they accommodate such demands much more efficiently, each car may make much more efficient use of road and parking capacity, but total demands may rise so much that even more space will be needed for road surface and for parking. In the fully autonomous vehicle, as the driver need pay little or no attention to driving, driving time may become work or play time in effect negating the time cost of travel. Autonomous cars might also safely travel much faster. Such changes might turn the 50-mile commute of today into the 100-mile commute of tomorrow. Today, people trying to travel by other modes such as walking or bicycling must contend with urban sprawl governed drivers perceptions of distance. How will they reckon with distances that have doubled again? Presumably many of them too will resort to driving. However reluctantly they turn to it, their decision will be taken as a vote for driving. Finally, as the skill demands of driving fall well have more drivers. Such trends would mean that we would continue to rebuild the world for drivers, instead of asking what world we want to live in and conforming driving to it.

If we do get the right implementation of driverless cars then the impact they could have is absolutely transformative.

The transformative potential?

This piece from Vox looks at some of the massive potential driverless cars offer. He openly says he’s being a bit utopian in his thinking on some of the benefits that could be delivered. These include that we can “right-size” both vehicles and the infrastructure needed to support them such as roads and parking. Doing that could mean more space made available for people. They also could fix the suburbs, freeing up space and allowing more development to occur all while improving air quality (assuming they’re electric).

The piece also notes that this utopia is a long way off. There are a lot of vested interests in the transport system and many of those will be reluctant to change forcing path dependence along a route we don’t necessarily want. The article also highlights that there are likely to be some substantial privacy implications.

But “smart” means information, and a city filled with sensors and trackers will accumulate a lot of information about every citizen within it. Who owns that information? Who can access it? How much will basic services like transportation hinge on the surrender of personal data? How will all that data be protected from the copious cybersecurity threats that face smart cities?

As people continue to look closer at driverless vehicles they will also continue to see there are a lot of issues that needs to be addressed. As such it could be some time before we really start seeing any serious driverless car proposals.


Driverless cars – What about the Pedestrians

An interesting article on Citylab highlights something I’ve been saying for a while when it comes to the hype around driverless cars – what about the pedestrians. It’s based of the video (below) from MIT’s Senseable City Lab showing how an urban intersection might work in a world of driverless cars.

Imagine a city without traffic lights, where lanes of cars merge harmoniously from one to the next, allowing traffic to flow smoothly across intersections. This futuristic vision is becoming reality. The development of autonomous driving promises to revolutionise the landscape of urban mobility.

As Citylab note:

The first thing to notice is how truly terrifying it would be—at least initially—to ride in a driverless car going that fast through an intersection. Seriously: pause the video at 44 seconds and see how narrowly the car turning left avoids being slammed by another going straight. When you ride in a self-driving car, you quickly learn to trust it; in fact, Google has said its early test riders trusted the car too much on highways. But having faith in a computerized intersection overlord to orchestrate so much city traffic at such great speeds will require a steep period of public adjustment.

The second thing to note is far more important: Where are all the pedestrians and bike riders? (Hat tip to Columbia University planning professor David King for bringing this to our attention.) Keep in mind this wasn’t some remote crossing being modeled; it was the intersection of Massachusetts and Columbus avenues in Boston.

There’s an obvious reason why an “intelligent intersection” would want to eliminate people crossing on foot or by bike: they’d slow things down. But it would be a huge mistake for cities to undo all the progress being made on human-scale street design just to accommodate a perfect algorithm of car movement. If the result is that driverless cars need to move through cities at sub-optimal speeds, then so be it. We won’t be losing as much productivity to traffic as we do today, anyway.

So yes driverless cars might improve vehicle throughput a bit over what we have now but almost certainly not as much as the claims made by many about the technology unless we take the retrograde step of marginalising pedestrians (and cyclists).

I must say I hadn’t thought about the issue of how it would feel being in a driverless car and going through an intersection and seeing another car approach at speed from a different direction. That would definitely take some time to get used to and would be very scary if you were on a bike.

All up it just adds to my view that while driverless vehicles are coming, they won’t be here as soon or have as big an impact as some people claim. In the mean time we will likely continue to see aspects of autonomy increasing such as technology that helps improve safety or assist with parking.

Driverless cars to increase congestion

Proponents of driverless cars often suggest the technology will make all sorts of significant changes to transport. Gone will be car ownership with people just hiring a car when they need one, like taxi’s only easier and cheaper. As such they say gone too will be the need for public transport, especially when you take away some of the benefits PT currently has like being able to do other things such as work, read, use a phone or even sleep. Further congestion will also be a thing of the past with these smart vehicles able to better work together rather than the randomness of humans. Of course the biggest and likely most accurate prediction will be safety as for a start these cars will obey the road rules so no speeding, no running red lights or any of the other bad habits human drivers have.

That all sounds wonderful however an article from CityLab highlights research showing that at least for some time driverless cars could actually make things worse on the roads.

A new simulation-based study of driverless cars questions how well these two big secondary benefits—less traffic and more comfort—can coexist. Trains are conducive to productivity in large part because they aren’t as jerky as cars. But if driverless cars mimic the acceleration and deceleration of trains, speeding up and slowing down more smoothly for the rider’s sake, they might sacrifice much of their ability to relieve traffic in the process.

“Acceleration has big impacts on congestion at intersections because it describes how quickly a vehicle begins to move,” Scott Le Vine of Imperial College London, who led the research, tells CityLab via email. “Think about being stuck behind an 18-wheeler when the light turns green. It accelerates very slowly, which means that you’re delayed much more than if you were behind a car that accelerated quickly.”

For their study, Le Vine and colleagues simulated traffic at a basic four-way urban intersection where 25 percent of the vehicles were driverless and the rest were standard. In some scenarios, the driverless cars accelerated and decelerated the way that light rail trains do—more comfortable than, say, riding in a taxi, but still a little jerky at times. In other scenarios, the cars started and stopped with the premium smoothness of high-speed rail.

Within these broad scenarios the researchers also tested alternatives that reduced speeds but improved smoothness, such as longer yellow lights or following distances. All told they modeled 16 scenarios against a baseline with all human-driven cars. The researchers then ran each simulation for an hour, repeated it 100 times, and calculated the average impact that scenario had in terms of traffic delay and road capacity.

In every single test scenario, driverless cars designed to create a comfortable, rail-style ride made congestion worse than it would have been in a baseline scenario with people behind every wheel.

So cars with fast acceleration and deceleration are obviously easy to make but that’s not what people are likely to want if you’re also trying to do some of the other activities mentioned earlier. Regardless traffic generally moves at the pace of the slowest vehicles so all it takes is one slow driver or driverless vehicle and many others will be slowed down too. I bet they won’t say that in the marketing brochures.

The interior of the Mercedes-Benz driverless car concept


I suspect this isn’t the only aspect of driverless cars that could create congestion. As an example the driverless taxi model that most people say will happen, is likely to result in a lot more vehicle movements as cars reposition themselves to pick up additional fares. That means that where roads are generally congested in one direction only, with driverless cars congestion could occur in both directions.

Will people choose to buy new vehicle technologies?

Last year we started to take a look at an emerging technology that some claim will revolutionise urban transport – driverless cars. My view is that they aren’t all they’re cracked up to be – if we wanted to, we could easily get the purported benefits by investing in existing, proven technology:

While driverless cars (or hoverboards for that matter) sound exciting, we can’t afford to pin all of our hopes on them. The pragmatic, proven way forward for transport in a big city is the same as it’s always been: Give people good transport choices by investing in efficient rapid transit networks, frequent bus services, and safe walking and cycling options.

If we want a safer, more efficient, and more environmentally friendly transport system, we can achieve it now by making smart policy changes. We don’t have to wait.

But, for the sake of argument, let’s say that we did want to wait for driverless cars to solve our self-imposed problems. How long would it take, exactly?

The wait would be a function of three factors:

  • First, how long it takes until driverless cars are proven and widely available for purchase in New Zealand. Most people agree that the technology is improving and may be ready for wide deployment sometime in the next decade. (Obviously, regulatory barriers could slow uptake as well.)
  • Second, how long it will take the New Zealand vehicle fleet to turn over – i.e. how long until the cars that’s currently on the road is scrapped and replaced. At the moment, the average NZ vehicle is around 13 years old, meaning that we’d expect it to take at least 13 years for half of the fleet to be renewed. Full replacement of every car on the road could take 25-40 years – a quick glimpse at Trademe shows that people are still buying and selling cars built in the early 1980s.
  • Third, and possibly most importantly, how rapidly driverless cars gain market share. Even after the introduction of driverless cars, most people will continue buying self-drive cars, which will dramatically slow the transition to a driverless fleet.

People have spent a lot of time thinking about the first two points, but I haven’t seen any commentary on the third one. Fortunately, we can draw upon some real-world data to get a sense of how rapidly consumers take up new vehicle technologies. Over the last decade, hybrid and electric cars have become commonly available, with cumulative global sales figures in the millions. While they tend to be more expensive to purchase, they offer savings on fuel costs and improvements in environmental performance.

So: How have consumers responded to recent technological transformations?

In short, they have hardly noticed. People are not rushing to give up their petrol (and self-driving) vehicles, even though there are now viable alternatives. A recent study from the US has found that hybrid vehicles’ market share has stayed low, even though car-makers have introduced many more new models. Over a decade after the Toyota Prius first arrived on the market, hybrids account for only one in every thirty new car sales in the US:


Source: IHS/Polk

Obviously, uptake of hybrid and electric vehicles has been faster in some places than others. However, a 2013 New York Times article on new vehicle technologies found that alternative vehicles have failed to capture a majority of the market even in the most favourable environments:

SANTA MONICA, Calif. — It would seem to be a good time to own an electric car in Santa Monica. From the charging stations dotted around town to the dedicated public parking spaces — all provided at no cost by the city — Santa Monica has rolled out the welcome mat for electric cars.

But even here, in this wealthy, environmentally conscious city of 90,000 west of Los Angeles, only a core group of owners has switched from traditional gasoline-powered cars.

Less than 4 percent of registered cars run only on battery power, according to an analysis by the industry researcher Edmunds.com of data from R.L. Polk, which records vehicle registrations nationwide. Hybrids, which run on some combination of gasoline and battery power, account for 15.5 percent, the data says, but many of those are traditional hybrids, which do not require a plug-in cord for recharging.

In other words, after a decade, over 80% of Santa Monica’s car fleet is still composed of conventional petrol cars. And that’s about as good as it gets anywhere in the US, which is on the leading edge of many new trends in vehicle technologies.

The picture isn’t much different outside of the US. Research on vehicle fleets in 19 countries shows that there are only two countries where hybrids and electric vehicles account for more than 1% of vehicle fleets. Norway (largely electric cars) and the Netherlands (mostly hybrids) were far and away the leaders in uptake, due to extraordinarily generous subsidies for buyers. Everywhere else lagged far behind:

In short, people don’t seem to be rushing out to buy new vehicle technologies. Although we all have the option to buy electric now, few people do in practice. It is very likely that driverless cars, when or if they become readily available, will follow a similar pattern. Initially, at least, they will be costlier and seem riskier than self-drive cars. Current rates of uptake for hybrid and electric cars suggest that it could take half a century or more for petrol cars to vanish from the road. Why should the transition to driverless cars be any faster?

All in all, recent market realities should encourage caution about driverless cars. Slow rates of uptake for new vehicle technologies mean that they aren’t going to solve our problems any time soon. A 2014 London School of Economics report on the state of urban mobility (pdf) described the dilemma of vehicle technology innovation well. They noted:

Regarding the development of new transport technologies, key actors (above all the automotive sector) have failed to convert technological progress into substantive improvements in energy efficiency and vehicle emissions or more broadly transform modes of accessibility in cities.

The clear implication is that if we want better transport outcomes, we must implement better transport policies. The data shows that waiting on new technologies is not a sensible option. If we want to lower the road toll, we must invest in safe roads, including protected cycle infrastructure. If we want a workable solution to congestion, we must build rapid transit infrastructure, bus lanes and walking and cycling improvements to give people the choice to avoid it.

There is no realistic alternative – so why don’t we get on with it?

What if a technology revolution happened and nobody noticed?

A new NZIER research report, entitled “Disruption on the road ahead! How auto technology will change much more than just our commute to work“, makes the case that new technologies will upend urban transport systems:

Near autonomous cars followed by driverless vehicles (smart cars) will transform our commute to work and much more over the next two decades.

Car-based technologies hold the promise of reducing the billions of dollars we spend on roads by improving how we use them and by saving lives.

We need to rethink our reliance on infrastructure solutions to transport problems and look at how to effectively embrace the new technologies.

If you read on, the report implies that we should stop investing in public transport and count on driverless cars to allow roads to flow more efficiently. The report does not grapple with the question of where the driverless cars will be stored – in spite of the fact that parking is one of the costliest elements of a car-heavy urban transport system. Space is expensive in cities and cars, even driverless ones, do not use space efficiently.

Now, let me be perfectly clear: this is a lazy analysis. As I have previously argued, waiting on unproven technologies to solve our problems is a bad strategy, especially when there are proven technologies that can be implemented right now. It is more realistic to invest in frequent bus networks, rapid transit infrastructure like Auckland’s rail network and the Northern Busway, and safe cycling facilities like the separated Beach Road cycleway. (Auckland Transport, like many other transport agencies, understands this and is getting on with it!)

However, if we set aside NZIER’s technological utopianism, they are making a reasonable point: When a transformational technology emerges, governments must make complementary public investments to enable society to benefit.

With that in mind, I would like to point out that a technology revolution has happened over the last decade – and gone largely unnoticed by New Zealand’s transport agencies. I’m talking about electric bikes, which are now proven, readily-available technology. Several companies are selling them in New Zealand, with basic models going for under $1000, which is price-competitive with a new road bike. In the Netherlands, 19% of all new bikes purchased in 2013 were electric.

Why is this so revolutionary? Simply put, because electric bikes flatten out all the hills on a cycling route. By providing a bit of extra oomph when riding up inclines, they remove a major barrier to cycling in hilly cities like Auckland and Wellington. Suddenly, the vertiginous climb out of the Queen Street gully might as well be pancake-flat Christchurch.

electric cycle 07122012.JPG JPEG 0540287993

Easy as… climbing the world’s steepest residential street on an electric bike (Source)

Even on flat sections, the additional power provided by the electric motor can make cycling much more relaxing and gentle. That may not matter to the young and/or fit, but it’s a boon to people who are less fit or only starting to cycle.

Consequently, electric bikes have the potential to majorly disrupt New Zealand’s urban transport markets. According to my calculations based on 2013 Census journey to work data, one-third of all commutes in Auckland are under 5 kilometres. At present, only a very small minority of those trips are done by bike. Recent technological change means that could shift, and rapidly. Taking all those short trips on bikes would have a much more fundamental impact on congestion than driverless cars.

However, there are some big barriers to getting the full benefits of this transformative technology. Simply put, our roads often feel too unsafe to ride on. People on bikes often must compete for road space with cars, buses, and trucks. They have to look out for cars backing out of driveways and drivers opening doors into their path. Over a lifetime these risks are more than balanced out by the health benefits of cycling, but they can feel a bit intimidating.

Copenhagen cyclists wikipedia

If only there was something we could do to make streets feel safer for cycling… (Source)

There is a strong case for public investment and policy changes to unlock the benefits of electric bikes. It is relatively easy to make cycling safe and common by investing in a complete cycle network. This means:

  • Implementing more off-road cycle paths like the successful Northwestern Cycleway and Grafton Gully Cycleway, and the impending Nelson Street and Glen Innes to Tamaki Drive cycleways
  • Putting separated on-street cycle lanes, like the excellent Beach Road cycleway, on every major road where there is enough space
  • Slowing speed limits to improve safety on side streets and alternative routes like the Dominion Road parallel routes.

This can all be done immediately at a relatively low cost. It will enable us to benefit from a transformative technology that actually exists right now, rather than waiting decades for an unproven technology. So why aren’t the enthusiasts for “disruptive technology” taking notice?

Dude, where’s my hoverboard?

The 1989 sci-fi comedy film Back to the Future II gave audiences a glimpse of what transport would be like in the far-off future date of 2015. Flying cars would merge and flow seamlessly in midair motorways, allowing for cities to be large, low-density, and free of congestion. (Something that is ordinarily a geometric impossibility.) And, of course, there was Marty McFly’s famous hoverboard:

Back to the Future hoverboard

Now it’s 2014, and I’ve got to ask: Dude, where’s my hoverboard? Why haven’t any of the promised revolutionary breakthroughs in transport technology actually happened?

Or, more generally: Should New Zealand wait for new, unproven technologies to solve its transport problems, or should it invest in existing technologies instead?

A few people, including the new Minister of Transport, have called for New Zealand to look to driverless cars as a solution to road congestion. Matt reviewed the state of the debate in a recent post. I’d like to look at the issue from an economic perspective instead.

Economists often use the concept of a “production possibilities frontier” to display the alternative production and consumption options available to an economy at a point in time. Here’s an example of a hypothetical PPF for an economy which only produces two goods (“guns” and “butter”). Points inside the curve (like point A) represent underutilisation of resources, while points outside the curve (like point X) are impossible to sustain.


Source: Wikipedia

Importantly, the shape and position of an economy’s PPF depends upon the technology (and techniques) that are in use. Over the long term, economic growth means shifting the curve outwards by adopting better, more efficient technology. And I’d like to stress “adopting” rather than “inventing”. While some countries have done well by developing new technologies, the best way for underdeveloped countries to raise their living standards has been to adapt proven technologies. Technology transfer is almost always easier and cheaper than inventing totally new technologies.

This is how east Asian countries like Japan, South Korea, Taiwan, and (most recently) China have become wealthy. They didn’t wait around for unproven pie-in-the-sky ideas to come along – they just figured out how to do the same things that successful countries were already doing. If Japan had decided, in the 1950s, to await the semiconductor revolution rather than figuring out how to make cheap copies of American consumer electronics and cars, it would still be a poor country.

The same is true for New Zealand’s transport system. In economic jargon, our transport system is far behind the technology frontier. While we have an extensive and relatively well-managed road network, we have failed to adopt a number of technologies and techniques that have been proven elsewhere. As a result, Auckland has:

  • A suburban rail network that can’t provide metro-style levels of service due to the lack of the 3 kilometre City Rail Link tunnel – in spite of the evidence from numerous overseas cities that frequent rail systems are efficient ways of moving a lot of people without congestion
  • A serious lack of safe cycling infrastructure – in spite of the fact that cities that have made cheap, efficient investments in cycling have reaped the rewards in terms of health and wellbeing
  • A single busway, and slow (but steady) progress on further busways – in spite of the fact that they are a cheap and easy-to-implement ways of getting high-quality rapid transit to underserved areas of cities.

Because these technologies already exist, we could implement them now and reap the benefits immediately. If we want to rapidly improve our transport system – i.e. move our PPF outwards – we should look to existing but underutilised technologies rather than awaiting this decade’s version of Marty McFly’s hoverboard.

A recent CityLab series on the future of transportation makes this point clearly. The cities that are moving ahead rapidly are largely investing in proven technologies like bus rapid transit, light rail, and cycling infrastructure. (As well as low-cost improvements to existing technologies like bike share.) They are not holding off on smart investments while awaiting new technologies.


Auckland desperately needs technology transfer (Source: Wikipedia)

By contrast, there are three big reasons why waiting on unproven technologies like driverless cars will make us worse off.

  • Waiting for new technologies imposes a significant opportunity cost. If we use the future promise of driverless cars as a reason to avoid investing in the City Rail Link, safe cycling infrastructure, and busways right now, we will sacrifice years of better transport outcomes.
  • Waiting for new technologies is risky, because new technologies may not ever succeed. An in-depth Slate article on driverless cars highlights how far they need to come before widespread deployment. There are big problems to solve in navigation and safety in poor weather conditions. Computers are still not very good at dealing with unpredictable situations.
  • Even if the technology was available today, it would still take decades to replace New Zealand’s entire vehicle fleet. New Zealanders own 2.7 million passenger cars, and it would cost tens of billions to fully replace them. New cars entering the fleet today will still be on the road in 15-20 years. In other words, we wouldn’t see the benefits of a fully driverless vehicle fleet until the 2040s or so.

In addition, many people actually enjoy driving quite a lot and wouldn’t really want to be chauffeured around by a robot. Driverless cars are not a good substitute for a V8!

While driverless cars (or hoverboards for that matter) sound exciting, we can’t afford to pin all of our hopes on them. The pragmatic, proven way forward for transport in a big city is the same as it’s always been: Give people good transport choices by investing in efficient rapid transit networks, frequent bus services, and safe walking and cycling options.

It may seem boring, but as an economist I’ll take tried-and-true over utopian fantasies any day.

Rail vs Driverless Cars

A few days ago there were two major transport stories, the first was about a new record for rail patronage and the other topic was about the government looking to make it easier for driverless cars to be on New Zealands roads.

The prospect of cars travelling New Zealand highways with no one behind the wheel is moving closer says new Transport Minister Simon Bridges. Officials are reviewing legislation allowing for the testing of umanned autonomous vehicles on public roads.

Mr Bridges has pledged to work with environmental interests while also pursuing the Government’s road building programme.

Mr Bridges said he was committed to “a balanced approach” and ongoing investment roads were important even from a green perspective, “over time as we move to electric vehicles and autonomous vehicles”.

Mr Bridges said the Government was not doing a great deal to accommodate autonomous vehicle technology, “but I don’t think there’s any doubt that if you look at what’s going on internationally, maybe not in the next couple of years, but over time we will see driverless vehicles and that will have implications, like for example less congestion because vehicles can travel closer together”.

We’ve discussed driverless cars a bit in the past so I’m going to try and not rehash those arguments too much. What I do want to touch on is the odd relationship between them and rail. By that I mean opponents of rail investment often like to claim that rail is an old technology – despite the fact that modern rail systems involve some very sophisticated tech – and that we should instead look to the future which they see as being driverless cars. Of course some rail systems have been driverless for decades.

One of those opponents of rail investment is Phil McDermott who runs the ironically named Cities Matter blog which argues for low density and auto centric cities. He was a guest on Radio NZs The Panel talking about both rail and driverless cars however his contradictions were huge and probably about 0.8 on the David Seymour scale. The section on The Panel starts from ~11:15 and McDermott comes in from ~14:10

or listen here

He starts off by dragging up the old cliché that trains run on fixed routes but that roads allow for flexibility and then says that if the city develops as expected that people will be travelling across the city between centres. Of course everyone traipsing across town to dispersed centres is something we’ve been doing for decades and has only led to more and more congestion. The intention of the Auckland Plan is to focus growth in and around the central city, a handful of major metropolitan centres and a wider range of local town centres which are all linked by high quality public transport. It’s that public transport network, of which rail is an integrated part, that will be key to moving a huge volume of people around and doing so free of congestion.

Auckland Plan - Development Strategy

Here are some of the other points and contradictions he made.

  • Trains in Auckland are full by the time they get to the CBD but that we shouldn’t build the CRL as he thinks the trains won’t carry a lot of people. You really have to wonder what’s going on in his brain as it works through that logic.
  • That the CRL doesn’t do enough for the transport system despite the fact it doubles the capacity of it.
  • That we shouldn’t have intensification near rail lines as somehow it creates a high marginal cost for each extra trip however later he says if we want rail to work in NZ he says we need lots of cars on the road with people driving to stations with big park n ride facilities.
  • That road building is ok because in his view the marginal costs are low which of course conveniently ignores that we’ve exhausted all the easy road building options and are now faced with massive costs for projects, often for not much gain in capacity or mobility.
  • That driverless cars will increase car use and that it will make congestion worse, but it’s all ok because they might not be as polluting as our current fleet.

Many of the views he expressed are downright odd and I get the impression that what people like McDermott are really after is to preserve the status quo which they likely currently benefit from. Some might wonder why bother to even discuss the interview but unfortunately we still see the likes of Phil trotted out on a regular basis.