7 ms·
What will self-driving cars do to cities?
- lukethomas 9y agoA common theme with the self-driving car discussion is the potential to reshape cities. While I'm sure this will happen, I'd love to see thoughts/ideas on the impact in suburban & more remote areas. For example, will people be willing to commute longer distances? Will smaller cities outside larger metropolitan areas become more popular?
- noir_lord 9y agoI think a lot will depend on the average speed attainable in a system of wholly autonomous vehicles, if you can commute an hour each way at 90mph that's a catchment area 2.6 times larger (than at 55mph). That would have a huge effect. 90 mile circle around London covers the entire south east of the UK pretty much. 90 miles from Leeds covers Manchester, Liverpool, Birmingham,bradford and Hull which is much of all the major cities outside of London in the midlands/north.
- tyingq 9y agoI suspect by the time there are enough self driving cars that it lowers drive time, other factors will kick in as well. Automation to that degree would likely have wiped out many other non-driving jobs in the city, or made them such that doing the job remotely was possible.
- pmiller2 9y agoI already commute an hour each way by public transportation. The only benefit a self-driving car could theoretically offer me is that I don't have to deal with other passengers.
- jrs235 9y agoOr extra stops. Or aligning your schedule to the shared vehicles schedule.
- hn_throwaway_99 9y agoWhich would be huge, at least for me. I'd love to be able to pull out my laptop and get rid of my "busywork" (e.g. emails, review daily reports, etc.), which would be a hell of a lot easier in a self driving car than, say, the BART.
- EternalData 9y agoInteresting subpoint here: people who are able to work remotely are going to add about an hour or perhaps a bit less to their productivity but those who cannot (let's say a hypothetical janitor/sales clerk) will not, further adding a layer of inequality to those who work with their minds versus those who work with their hands. This is all things equal, and assuming that the advent of self-driving cars doesn't also bring a whole new economy/job market.
- cm2187 9y agoI wouldn't overdo the remote working. My experience is that it is terribly inefficient in my business. It's fine at small dose to help people cope with other obligation but I don't see that being the future of office work.
- majewsky 9y agoIt depends on the person. Some people are amazingly productive while working from home (I have two such colleagues), while others (e.g. me) are clearly less productive and more easily distracted at home.
- daurnimator 9y agoYou usually pay a premium (in real estate/rent) to live near fast/accessible public transport. As self-driving cars are introduced, places that previously had terrible public transport access, yet are still within a 45 min drive of CBDs via driving will have mini-booms.
- brightball 9y agoI am most interested in what self driving RVs will do.
- majewsky 9y agoThat would be an interesting alternative for business flights, potentially. Instead of flying from NY to DC and staying in a hotel there for the meeting next day, just hop into the autonomous RV, lay in a course and go to bed.
- djyaz1200 9y agoAll the dependencies that make urban density relevant are being eroded by technology. Self driving cars will be one part of a trend that makes remote living a lot more pleasant and popular. Amazon will deliver most of what you need to live (and soon eat). SpaceX is working on low orbit and therefore low latency high speed global internet so you can work and/or get entertainment anywhere. Tesla can sell you solar panels and a battery to power your home. You can do a septic system instead of a sewer hookup. If you insist on gas you could get propane delivered inexpensively. If you want to mingle/meet people you can do that online. Then with a level 5 self driving car you can send your kids to far away schools everyday easily, or work at your "desk" during a long commute in the car. Or say if you want to drive 90 minutes into town for a basketball game that's no big deal if a computer does the thinking while you watch movies in the back. I live in Oregon where I expect a profound impact. There are so many amazing places here that are incredibly beautiful/cheap and within 90 minutes of a major metro area/airport. I think the real driver of this change will be cost, younger people (myself included) are getting priced out of owning a nice home in a major metro area. I'm seriously considering moving to Sweet Home Oregon and building a mansion on acres for half the cost of a modest townhouse in portland. High speed Internet is the missing piece for me.
- ghaff 9y agoIt's at least plausible that the current infatuation with mostly a handful of dense urban cores by mostly a certain young educated demographic will end up being a short-term blip. It's pretty recent--which is one of the big reasons for all the issues with housing and transit infrastructure in those areas. It wouldn't surprise me at all if people look back in 20 or 30 years and see migration to the densest urban areas in the US as this temporary aberration.
- majewsky 9y ago> If you want to mingle/meet people you can do that online. There's a huge difference between meeting online and meeting offline. Online might become a venerable substitute for offline if VR/AR technology advances to the point where we can have the full experience of the other people sitting around the same table, i.e. their whole body language etc.
- Fricken 9y agoThere was a period where I was commuting in a private coach bus 1.5 hours each way. It was a really nice bus, clean, air conditioned, wifi, reclining seats. Nonetheless, spending 3 hours a day in transit still sucked.
- ghaff 9y agoYeah. A while back I had a sometimes commute that was about 1.5-2 hours, mostly by train plus a short drive and a short walk. It was somewhat manageable but I would have hated to have done it every day. Yeah, you can read etc. but it's still a big chunk out of your day.
- jaclaz 9y agoI must be missing a link. Whether there is a driver or not (or whether the cars will be electrically powered or not) the cars will still occupy the same physical space, please read as parking, and if the same number of people will need to go from point A to point B and viceversa, roads will be needed as well.
- johnking 9y agoWith regards to the parking space article reference. In the long run with self driving cars we can reduce the overall number of parked cars through greater vehicle utilization. Instead of them being parked for long periods of their life, they can join fleets (think the Tesla fleet idea) and service many people in one day.
- SomeStupidPoint 9y agoOnly if you can negate the need for a personal vehicle, which self-driving cars aren't close to doing. Self-driving cars do nothing about the fact I need somewhere to put my gym bag during the work day or my gym and work bags while I go out in the evening. Or that I might need a carseat for a child. Or that I might like to keep an umbrella and change of clothes in my car. Or.... Any benefit to not having the static car is quickly mitigated by the inefficiency of storage shuffling for even a moderate amount of things.
- cm2187 9y agoThere is another objection. People who use a car to go to work need that car at the same time. Playing lottery on whether I will be allocated a car or not to get to work this morning isn't an option. That's why farmers usually own their hardware. They could rent except that they all need it the same days. It will free up the streets nevertheless. The reason you park your car nearby is so that you can walk to it. But if the car can go park itself and come back 10-15min from where you live in some big car park, you would probably as happy as if you had to walk 5-10min yourself. Plus ownership isn't fully rational. Otherwise people wouldn't bury themselves into debt to own their house, pushing property prices up. Instead you would have cheap property you can rent.
- fullshark 9y agoStar Trek style Teleportation is the real game changer regarding real estate prices.
- jaclaz 9y agoAs long as the manufacturers of the teleportation devices won't do OTA updates, otherwise there will likely be a sudden drop in demand for real estate, before or later ...
- seanmcdirmid 9y agoUntil people find out the teleporter is really just making a copy of the person at a remote destination and then killing the original.
- majewsky 9y agoActually, it's the other way around. You make the copy after killing the original. :)
- seanmcdirmid 9y agoThat would be considered a bug, like copying a deleted file.
- madengr 9y agoIt may raise prices at the outer suburbs as a commute will be easier.
- specialist 9y agoCar sharing will be the catalyst that reshapes cities. Reducing car ownership 90% (let's say) will reduce the space allocated to parking and traffic proportionally. When that happens, land use (zoning) will become much more dense. Maybe self-driving cars will hasten car-sharing. That seems to be Uber's long-term bet. We'll see.
- mmcconnell1618 9y agoI don't think car ownership will drop by 90%. People like their personal space and think about a family who needs to put a car seat into the self-driving vehicle every time they use it. What about all the other junk you have in your car? People who can afford their own self-driving vehicle will probably have one (but maybe not multiple cars per family). Cars will likely still be seen as an expression of personality or status symbols.
- seanmcdirmid 9y agoThis is a very American mindset. I agree with what you said if we restrict the world to the USA.
- cagenut 9y agoIMHO self driving cars will predominately be a fleet/subscription-service model eventually, and in that case it will actually further incentivize shorter commutes (denser living). Right now, when you buy, finance or lease a car you pay a fixed rate and the variable component is almost negligible. That means a 10 or a 50 mile commute are really just a question of your personal time, they don't have much price-cost. Conversely for a shared-fleet-subscription model you will pay a much more utility-like rate (or they will price the subscription based on utility like analysis). Now there will be a much more direct correlation to the distances you travel and the price you pay. Especially since commuting will have the peak-utilization effect on pricing that all shared services from frame-relay networks to ec2 autoscaling exhibit. People will be able to directly price the cost of living in location X vs. location Y and make a much more informed decision (I can pay $250/month on my mortgage for a 3rd bedroom in a home closer to work, or I can pay $300/month for my 45-minute rush-hour service charge). Overall more accurate pricing will dramatically skew toward density.
- jstanley 9y agoBut if living further from work costs you $300/mo extra, then it won't actually be cheaper to use a subscription-service model than to own the car yourself, and therefore people won't do that.
- seanmcdirmid 9y agoYou ignore places like china with severe traffic problems in their cities and an autocratic government that can push things through. They'll make self driving cars mandatory inside the fifth ring road just to optimize road and parking resources and ignore the western habits you describe above. That will be a huge deal for megacities, and not much of a change from personal habits where taxis are already ubiquitous. Also, see Singapore, Hong Kong, Tokyo, Bangkok, Manila, ....
- pjc50 9y agoI expect to see fossil fuel vehicles banned from London in my lifetime, so the city could well bar non-autonomous private cars once the autonomous tech is good enough. It would make it a lot safer for cyclists, although it's the trucks and buses that are the real danger there.
- mherrmann 9y agoI've been wondering if electrical cars (self-driving or not) will affect prices of properties which are - currently - next to noisy and polluted streets. Does anybody have insight into that? Or maybe smog-ridden cities?
- johnking 9y agoInteresting idea, and probably will have an impact. But definitely a long term bet, given the time scale that ICE cars will be replaced.
- patkai 9y agoAs much as Skype did.
- rsp1984 9y agoIf you look at the big European cities, real estate is expensive wherever there is an acceptable train or subway connection to the city and prices fall sharply outside of these zones. Self-driving cars will equalize this and it's a good thing. If you can take an affordable self driving taxi to the city you don't care about train connections any more and the key factors for choosing a home outside the city will change.
- dredmorbius 9y agoWould you have any data or maps showing this? Or specific examples in mind?
- simonrobb 9y agoI currently live in East Melbourne (~15 second walk from Hoddle Street) and pay through the nose for the privilege. And yet a few years ago lived in one of those big Victorian terraces on Hoddle just 50 metres away, where a mate and I paid AUD $240/week each for the entire place. What you say about the disparity in property value caused by the road is spot on. Funnily enough, a little while back my neighbour made the exact same point about Hoddle Street - in (say) twenty years the property values are going to see a massive shift because of changing technology in our cars. However I think it will be the move to electric cars that makes the real difference. No more fumes, no more grime covering every surface, far less noise. Add in reduced traffic because of self-driving tech and suddenly that property will be worth a packet.
- jwilliams 9y agoTotally agree - I think self-driving will be a major impact. But the electrification of cars will be much, much sooner.
- pouetpouet 9y agohttps://www.youtube.com/watch?v=utnPEbDNbrE&t=24 https://www.youtube.com/watch?v=utnPEbDNbrE&t=24 Jeff Speck: Autonomous Vehicles & the Good City 10 rules
- brudgers 9y agoThe politics of cities will trump the politics of self driving cars, and the politics of self driving cars sits a very long way from a future where enough cars are self driving that removing super arterial roadways will be a reasonable consideration or the costs of going underground are palatable...the Big Dig replaced one form of transportation infrastructure with another.
- sharmi 9y agoMost discussions seem to take for granted that the self driving cars as a service would equally service all participants at all times. The operators will be incentivised to service maximum people with minimum number of cars. At any point where there is high demand, there would not be enough cars to service everyone. So again, those who can afford it would opt to hace their own cars. If everyone is forced to use the self driving service, there comes the question of how the system prioritizes pickups, young vs old, healthy vs sick, the vulnerable etc. So, I believe it will still settle into the more wealthy using an elite service or owning their own car and others using a system not too different from a public transportation system, only self-driven.
- dredmorbius 9y agoThe operators will be incentivised to obtain maximum revenue with a minimum cost. That will likely lead to similar dynamics of other transport modes: 1. A high-priced, high-amenity premium "first class" (perhaps "white glove") service, catering to upper-income / high-discretionary-spending households. 2. Possibly: a moderate-tier, principally professional-employment / shopping-oriented tier. The primary support for this will come from the businesses served by the livery, notably professional employers (a transport/commute perk) and upscale shopping districts (to increase in-store traffic). I'm presuming that brick-and-mortar retail will remain a thing, and given a less-than 10% incursion by e-commerce, this seems likely. 3. A bare-bones "economy" tier. This will be made intentionally bad, not because of price, but because any amenity enhancement will erode demand to the higher-priced tiers. This will be offered to low-income households, and possibly to noneconomic sectors (children, students, handicapped, elderly). One characteristic of transport services dating to the dawn of large-scale passenger ship travel and rail service is that the lowest-tier amenities are made intentionally worse than necessary, for the logic outlined above.
- fulafel 9y ago100 years ago large passanger ships and railways were enormous investments that made it hard for competitors to enter the market, but the barrier to entry in driverless taxi service can be quite low. Regulation is another way that intentionally bad service can be fixed, it's widely done in rail and air travel.
- mcculley 9y agoI'm intrigued by thinking about what will happen when all of the parking lots and garages get turned into more useable space. When we hit the tipping point of self-driving cars being commercially feasible, it will make sense that they be parked where the land and electricity is cheapest, when they are parked at all. If we transition to mostly Uber-style car usage, the cars will be in motion as much as possible. That will lead to lots of parking lots and garages being razed in favor of more useful buildings. That means more density in urban areas and potentially more pedestrian friendly cities. Higher density means more electricity, water, and sewage usage. I was looking for new office space a few months ago. I asked every real estate agent and commercial builder what they were thinking about self-driving cars. They all said that they weren't thinking about it as it was too far off to worry about. I was surprised at their lack of vision.
- ghaff 9y agoI'm not sure why I would care as a real estate agent or commercial developer. I have to function in the here and now with clients (and zoning) in the here and now. That there might be changes brought on by autonomous driving or any of a slew of factors in a few decades is not really on my radar.
- mcculley 9y agoIf I were a real estate agent, I would be interested in what my market looks like in a few years. The commercial property developers I spoke to are building new properties right now and not thinking about how they might want to refit them in a few years. I agree that most of them aren't thinking ahead. I see buildings all around me being razed that were built just a few years ago. It's a myopic industry.
- Fricken 9y agoI believe that the first publicly available robotaxis will arrive in less than a year, as small scale pilot projects. It will take much much longer before they have a measurable impact on real estate and infrastructure.
- dredmorbius 9y agoA good premise, though I'd strongly suggest a retrospective view looking at how cities have been shaped in the past, by automobiles, lifts, and other technologies, in technological and pre-technological eras. Cities are dendritic or networked structures. They can be defined as a locus of points with connections between them, both physical (as with transport networks), or logical -- the financial, cultural, social, political, communications, and other links between nodes. Cities are characterised by greater nodal density than non-urban areas. Cities provide a superlinear return to scale, Cf. work of Geoffrey West (Santa Fe Institute) and others. This is subject to limitations, ultimately an irreducible inter-nodal cost factor. An example of this would be the disease risks of major cities such as London and New York in the mid-19th century. London actually had a net-negative population replacement rate, excluding in-migration. That is, the city could not replace its population faster than it was dying out. The principle cause was disease, quite notably cholera epidemics, responsible for deaths up of to 50,000 people annually. It took the creation of the London Sewerage System (1870) for this to be addressed. A similar history can be told of New York City. It had its first sanitation canal by the 1840s, but didn't have its first sewage treatment plant until 1903, and was still regularly dumping raw sewage into the Hudson River in the 1970s. (It now does so ... only irregularly.) http://toilet-guru.com/nyc-sewer-system.php http://toilet-guru.com/nyc-sewer-system.php New York (and Chicago) also saw the first use of the lift (elevator), which decreased the cost of vertical real estate, along with structural steel. The consequence being that those cities built up. By the 1950s, cheap petrol and booming auto sales made horizontal space cheap, and subsequent cities built out, notably also the Chicago area (which had land to expand into), Los Angeles, San Jose, Atlanta, and the Washington / Alexandria region. New limits were reached with smog and traffic congestion levels, in suburbs. At the same time, crime and social problems were driving middle-class and wealthy residents from city centres. Policing costs and requirements, as well as greenfield infrastructure developments made sprawl more attractive than densification. Self-driving, potentially electric, vehicles affect these dynamics a few different ways. They may allow for more effective roadbed utilisation. Given traffic density in many cities, this may well be limited, and still likely cannot meet to density of urban bus, light-rail, or subway / heavy-rail systems. They should reduce the requirements on residential and destination parking, by shifting parking to non-central regions. Parking may become denser and/or less deterministic -- it's not necessary that a specific vehicle is extracted, only that some vehicle is extracted. Utilisation costs of transport likely shift toward marginal rather than fixed costs. That is, much of the present cost of private transport is a sunk value, whilst a distance- or time-based rental front-loads the full direct cost. This may discourage casual travel. Transport use is very unevenly distributed through the day. Peak-load requirements are either un-met, are responded to with demand-pricing, or require a large unutilised capacity in off-hours. A likely development of a multi-tiered service-class offering, as with rail, air, or ship transport. The lowest-level tier offering will be intentionally disincentivised to foster utilisation of higher-level tiers, in a private-enterprise system. Elements of the service might be more or less reliable than current private transport. I suspect on balance that denser rather than less dense construction will be incentivised, though it's hard to account for all factors, or arising disincentives.