New "Shared Mobility Principles" have too much 2018 thinking.


A new group has released a document called the "Shared Mobility Principles" for livable cities. It was started by Robin Chase (who built companies like ZipCar and others) and has had several of the mobile app taxi companies like Uber, Lyft, Didi and others sign on, though not Waymo, Cruise or the automakers.

Some of the documents principles are fairly "motherhood," but a few of them, especially the last, drift away from goals and into implementation and ideology. That's an unwise drift because we of 2018 don't know enough about how to do specific solutions. In addition, some of the goals are goals which have always seemed good but have turned out to easily be distorted to the wrong answers.

Here's the document with a critique of some of its tenets.

We plan our cities and their mobility together.

The way our cities are built determines mobility needs and how they can be met. Development, urban design and public spaces, building and zoning regulations, parking requirements, and other land use policies shall incentivize compact, accessible, livable, and sustainable cities.

The sentiment is good, but urban planning tends to have a fairly poor track record, especially on transportation. Especially on rapidly changing technology. Planning, such as it is, should be very dynamic and adaptable to change, laying down few rules and instead observing and learning and correcting only when things go wrong. In particular, planning is well advised to push the concept of "virtual infrastructure" not just because it is the right one, but because it is the most flexible. If you learn something new, you can change the city by changing the virtual layers, not the physical ones. The physical layer is best as bare pavement with few rules built physically into the pavement. Create the ability to allocate space for things like transfers between vehicles and modes. Do not allocate any corridor to one type of vehicle (as is the case with rail.) Disolve physical barriers and replace them with virtual ones.

We prioritize people over vehicles.

The mobility of people and not vehicles shall be in the center of transportation planning and decision-making. Cities shall prioritize walking, cycling, public transport and other efficient shared mobility, as well as their interconnectivity. Cities shall discourage the use of cars, single-passenger taxis, and other oversized vehicles transporting one person.

This is a good principle but sometimes it is applied in a way that forgets that most vehicles travel in the service of people. The idea that vehicles have been given priority over people is false; rather it is that the people using those vehicles got a priority. That can be addressed.

The document stresses "public transit" when my own models suggest that will become a thing of the past, and to the great benefit of the people. We should not lose sight of the goal -- good mobility for the needs of people - by naming any of the methods (like public transit) which we might hope with 2018 thinking to attain those goals.

We support the shared and efficient use of vehicles, lanes, curbs, and land.

Transportation and land use planning and policies should minimize the street and parking space used per person and maximize the use of each vehicle. We discourage overbuilding and oversized vehicles and infrastructure, as well as the oversupply of parking.

Fortunately, with the benefits that come from robocars, we don't have to be so efficient. We can value comfort, privacy and luxury and get away with it in the robocar world. In fact, my calculations show that smaller vehicles, especially outside of peak hours, are actually not just more pleasant for riders but more efficient, because they are only used fully loaded. Larger vehicles tend to get too much use at poor load factors and waste energy and rider time.

Efficiency is a great goal, but it is not the only goal. Fortunately, accurate pricing of road use and externalities naturally leads to the right mix of efficiency and other factors like comfort, time, privacy, safety and luxury. You never have to force people to be efficient if the real costs are allocated.

The reality is that we use only a small fraction of our available road and car capacity today. We just use it really badly, so we see a world of congestion and streets packed with 5 passenger SUVs with 4 empty seats. We're going to use so much less parking in the future that lot owners will beg cities to rezone their lots to other purposes, not because they are told to. More solutions to congestion become workable.

We engage with stakeholders.

Residents, workers, businesses, and other stakeholders may feel direct impacts on their lives, their investments and their economic livelihoods by the unfolding transition to shared, zero-emission, and ultimately autonomous vehicles. We commit to actively engage these groups in the decision-making process and support them as we move through this transition.

Here we see again a statement about methods rather than goals. There are no zero emission vehicles, not even electric cars powered only by solar panels.

I am always a bit wary of the concept of the "stakeholder." I think the only stakeholders are the residents and visitors of a city. In so many processes we have seen instead that corporations, political groups and others become "stakeholders."

We promote equity.

Physical, digital, and financial access to shared transport services are valuable public goods and need thoughtful design to ensure use is possible and affordable by all ages, genders, incomes, and abilities.

A motherhood statement, but actually much more complex. For example, Lyft and Uber arose by creating transport services only available to people with fancy smartphones and credit cards. And they needed to! A demand that innovators meet equity goals would have left us without the innovation that is making the better world possible. In the long term, we do want to reach this equity goal, but we should be careful how quickly we demand it.

In addition, some of the old ideas about equity will need rethinking. Is it better to try to make the whole fleet accessible, or to take advantage of the way cars can deliver themselves to create a "separate but superior" sub-fleet? Should everybody share vehicles or should women have the right to ride alone rather than with a creepy stranger?.

We lead the transition towards a zero-emission future and renewable energy.

Public transportation and shared-use fleets will accelerate the transition to zero-emission vehicles. Electric vehicles shall ultimately be powered by renewable energy to maximize climate and air quality benefits.

There is no such thing as a zero emission car powered by renewable energy, not if you want to be green. Electric cars must be connected to the grid for proper environmental benefit. To deliberately not connect the power sources for cars to the grid, so that their excess power is discarded rather than used to offset grid fossil fuel power, is an environmental crime.

There are no solar electrons, no renewable energy electric cars. (Biofuel cars can be renewable energy cars.) This pernicious myth makes people feel good but must be stopped. We should make our grid cleaner, and we should have electric cars, but these are two orthogonal things.

We support fair user fees across all modes.

Every vehicle and mode should pay their fair share for road use, congestion, pollution, and use of curb space. The fair share shall take the operating, maintenance and social costs into account.

This is actually about the only principle needed. It is not trivial to get this right, but the more you get it right, the more good results you get, including, most importantly, good results you had no idea would happen when you wrote the principles.

It won't be a world of all-shared transportation, though. And it's certainly not easy. Pure user fees bother people. They suggest the roads are for the rich, and congestion charges have failed politically in most places. There are interesting ideas which allocate unalienable road rights to all and use fees only to tweak things a bit.

People's desire for private transportation won't go away, nor does it need to. Private vehicles which take half a lane and carry 1-2 people can be very efficient in energy and road space. People want and deserve to travel privately door to door when the capacity to do so is available, and that's most of the time. And since robocars don't need to be parked close to their owner's destination, but instead can be densely packed, valet-style, in locations inconvenient for people but fine for robots, capacity will be plentiful.

We aim for public benefits via open data.

The data infrastructure underpinning shared transport services must enable interoperability, competition and innovation, while ensuring privacy, security, and accountability.

This is a very tall order. Innovative, competing companies need to have proprietary edges, and since value will be so much in data, they need to have proprietary data and systems. The need for interoperability is only modest here. While it might be nice if you had an app that would summon the closest Lyft, Uber or Taxi, the pain of these services operating independently is modest, and the benefit of robust competition is greater.

We work towards integration and seamless connectivity.

All transportation services should be integrated and thoughtfully planned across operators, geographies, and complementary modes. Seamless trips should be facilitated via physical connections, interoperable payments, and combined information. Every opportunity should be taken to enhance connectivity of people and vehicles to wireless networks.

People are going to be super-connected no matter what we do, and that's good. Vehicles will also be connected, but much less so, because of overwhelming computer security concerns. We just minimize the attack surface on these fleets of 1500kg robots roaming our world.

The need for planned interoperability is much less in the virtual world. In spite of all our instincts in this area, Skype, which refused to interoperate with the standardized SIP world of internet telephony, defeated SIP almost entirely by doing it better. We are not nearly far enough along to even dream of standards in this area yet.

On the other hand, because physical connections (transfers between vehicles) do need a place for them to happen, this is one of the few things we should be thinking about today. Fortunately, the philosophy of virtual infrastructure helps us. We just have to be sure that space is there as bare pavement at the intersections between major routes. Fortunately the old parking lots we won't be needing that much of offer a source of that space.

We support that autonomous vehicles (AVs) in dense urban areas should be operated only in shared fleets.

Due to the transformational potential of autonomous vehicle technology, it is critical that all AVs are part of shared fleets, well-regulated, and zero emission. Shared fleets can provide more affordable access to all, maximize public safety and emissions benefits, ensure that maintenance and software upgrades are managed by professionals, and actualize the promise of reductions in vehicles, parking, and congestion, in line with broader policy trends to reduce the use of personal cars in dense urban areas.

This is the most incorrect of the principles. This is not critical at all, and in fact it's a bad idea. We want a world of heavy competition and innovation for the next several decades (if not forever.) We want individual tinkerers and small players creating their own vehicles and selling them to small markets of rich, early adopters. We don't just want that, we need that. This principle that only services like the future versions of Lyft and Uber and city transit should operate in the future could not be more wrong. Shared fleets will probably dominate, because they are efficient, and because the user fees of principle 7 above will push towards that efficiency. If we price in the externalities of emissions, then lower emissions vehicles and shared vehicles will naturally result -- until somebody comes up with something even better.

Because parking is going to be ridiculously cheap there will still be plenty of private vehicles for those who want them.

One can't avoid having some cynicism over Lyft, Uber, Didi and others pushing for a rule that only big shared fleets can operate in cities.

Goals vs methods

I believe the right approach is to discuss and agree on our goals. Our goals are more stable. The methods we use to get those goals change constantly as technology changes, and we can't predict those changes. Goals include things like clean air, reduced global warming, pleasant urban spaces, pleasant (fast, convenient, predictable, luxurious) and low cost travel, productivity, safety, fitness, peace, social bonding and many others. Things like public transit, shared vehicles, electric cars, less parking and central planning are methods, not goals. The arrival of robocars should teach us how rapidly those methods change.


"You never have to force people to be efficient if the real costs are allocated"
I like this. Using clear evidence to publically explain the real costs (congestion, infrastructure costs and pollution) and then charging for those costs in a transparent manner should be a goal. Road pricing is an efficient way to do this as it gives good feedback on where infrastructure money should be spent rather than the more blunt tools of fuel taxes, registration and license fees.
A couple of thoughts: As changing to road pricing is politically fraught, it could be made voluntary. There should be good savings to be had as TaaS reduces the need for endlessly expanding and upgrading the road system. If those savings where applied to reducing road pricing costs but traditional revenue streams where left with normal inflation adjustments, then over time most people would go with the cheaper model. With TaaS companies it would be best if they could be encouraged to move directly to this model before they begin displacing private cars.

One problem often raised with driverless cars is that as the marginal cost of driving further is low, especially for EVs, and the trip is more relaxing than driving, they will increase congestion by encouraging people to choose to live a lot further from their common destinations. It's a fair criticism as in effect it is a contributing problem to congestion and the requirement for more infrastructure. While 'time of use' and 'area of travel' are obvious targets for congestion charging, some method of capturing these other costs within road pricing could discourage more urban sprawl and congestion.

An increasing distance based charge could offset some of this although it would be socially unfair to many low income earners who sometimes have little choice but live a long way from their employment. Apart from reducing house moving costs and moving them to more direct land taxes so the cost of , I cannot see an easy way to compensate or help these people.

Yes, I wrote about this early congestion problem before. Traffic gets worse before it gets better.

Overall, robocars are going to be more efficient, so you can't charge them and not change charging for the old cars. But the old cars don't pay that much -- gas tax which has not gone up as fast as it should, and a few hundred dollars in vehicle tax.

Right now we subsidize electric vehicles, both with credits and by having no gas tax. In time that will fade out.

The key is to make the cost congestion based. The way to do this is with 2nd price auctions. With a second price auction, if you are the only bidder, then then 2nd price is zero -- free. As soon as 2 people want the same resource, then you start paying, and you pay as much as they actually value it.

I have proposed in my other articles how to swap this from paying cash to having a lottery. The daily lottery gives you superior road access. And if you do efficient road access things (like ride the train, or an efficient vehicle) you always win the lottery.

My favourite hoped for outcome of Transport A Service is drastically reducing vehicle weight per passenger. Here in New Zealand and probably in the US, families take (mostly) summer vacations where the whole family and a variety of toys (surfboards, jetski, camping gear etc).
This represents a surge in demand for large capacity vehicles which if catered for by TAAS appears to simply shift the excess capacity from the private owner to the fleet.
I'm not sure if you have written on this topic before, but whenever I tell people about the huge potential benefits of right sizing, this bugs me.
It's probably not all doom and gloom, I'm not sure how representative the U.S and NZ are in their holiday habits.
Some possible solutions or outcomes suggest themselves.
The benefits of TAAS both reduction in cost and the removal of the many small burdens of ownership mean that over time people forgo trips which require larger capacity, and focus on holidays which don't require it.
People switch from ownership of summer toys to rental, the quantity, quality and selection of rental equipment all increase to enable this switch.
This still requires the ability to carry that equipment from rental store to accommodation and on to beach, lake, or trail head.
There is an implicit assumption that families need to travel together, wheras a 2 child 2 parent family could have 1 child and 1 adult in each vehicle.
For holiday seasons a percentage of vehicles could be reconfigured for luggage only. This would probably work if the daily work commute peak was greater than the holiday peak. If the capacity is not there then pricing could be used to spread the peak.
An additional complication is that when at the location where the leisure activity is taking place is that the car is no longer available for storage of optional extras brought along just in case. Security also becomes an issue, if you want to hail a ride on demand rather than at a pre booked time, are you comfortable withn your iphone in your bag rather than in an alarmed car? Would this have the effect of moving such apps to wearables?

Those who predict massive fleet reductions to 1/10th of the current fleet are overdoing it I think. However, 4 to 1 reduction seems not unreasonable. This applies to summer touring fleet too. Yes, everybody wants it only in the summer but not everybody is on holiday the same week. And this excess summer fleet (which is larger group and family vehicles) can still be used at other times, it is just a bit less efficient. It provides the peak demand buffer at a higher cost.

Delivery robots will exist though I don't think people want to divide themselves from touring luggage all that much. And while I think families and other groups will be fine with splitting their party among smaller vehicles for short trips, that is less likely for a road trip, though you could have things like video links between the cars to make it more acceptable. People want guaranteed availability. They are not going to accept being stranded on holiday because all vehicles are in use. There must always be a bit of extra supply. Extra supply is not a big problem if vehicles are otherwise in heavy use. Those wear out by the mile, not the year, so having extras only costs interest and parking, and parking is cheap out in the country, and interest is low at present.

Add new comment