Benedict explores the higher order consequences of a shift to EV’s and full autonomy, and is well worth the read (also a podcast with Russ Roberts is here). Benedict makes the case that electric doesn’t so much as get rid of the gas tank as kind of rip out the spine of the car. He prefers the term ‘autonomous’ to ‘self-driving cars.’ because the term ‘self-driving car’ is like saying ‘horseless carriage.’ US and Australian urban design was driven by the post-war adoption of the motor car, and a shift to autonomy may redefine the urban landscape, with implications for land-use, real estate, big-box retail, and more.
The first thing that strikes me about Benedict’s analysis and another perspective from Tony Seba, is that the people making the case for EVs and autonomy usually adopt a techno-optimist narrative – technology is making the world better and electric autonomy is integral to the reshaping of a (mostly) better world. In the Seba world, there seems to be an underlying assumption that there is an enormous latent demand for electric, and that once EV models reach a cost-range threshold, basically everyone will adopt electric. I’d like to challenge those narratives.
First of all, EV’s are going to work for many people. The combination of Autonomous-Electric Uber-style vehicle has enormous potential in an inner-urban landscape – if an on-demand ride is the cost of a coffee, why own a car? But what about suburban, the urban fringe and rural? This is where the Autonomous-Electric paradigm breaks down. The utopian vision of lanes of platooned Autonomous-Electric vehicles bringing outer urban workers into the city at high speed is a long way off.
Having worked closely with the automotive industry many years ago, the thing that is obvious from within the auto industry is the primacy of having cars on the showroom that motorists are going to actually buy. Tech websites and the media talk about EV’s and Autonomy to sell readership – the auto industry makes money from selling cars. Nowhere is this more obvious than the most recent sales data for EV’s in Australia. EV’s (and autonomy) are a gold mine for tech websites, but nearly no-one is actually buying them. From publicly available data, I estimated that there had been around 3,300 electric vehicles sold since 2010, with 1,140 sold in 2014 and 942 sold in 2015. The latest data showed that only 219 EVs were sold in 2016 out of 1.2 million new vehicles. Yes, that’s 0.018%. How does this compare with media coverage?
I was also alerted to the high number of low-kilometer Tesla Model S vehicles for sale in Australia. A search on Car Sales found 28 vehicles, many with low km’s. Given the small new vehicle sales, this seems high (or does it ?). Having driven a Tesla Model S, my impression was of a tight, sharp and powerful car. But I was also surprised at how quickly the novelty of electric wore off. Is this what’s happening? Enjoy the drive for a while but off-load the vehicle while the battery still has longevity?
Despite the large commercial gap between battery-electric and hydrogen-fuel cell-electric, many car manufacturers have substantial and ongoing investment in the hydrogen cycle – e.g. Honda, BMW, Toyota, Mercedes. Why? Because the low energy density of the redox-chemical pathway and underlying limitations of charging will constrain the penetration of battery-electric. It’s possible of course that a new chemistry will emerge, but most engineers I’ve spoken to don’t see lithium as being able to provide the sort of combination of energy density, cost, charge time and safety that is going to completely displace liquid fuels.
EV’s are often promoted as zero tailpipe emission, but there are two issues. Obviously the source of electricity is critical. An electric vehicle powered by hydropower is clearly cleaner than powered by coal-fired electricity. But the embodied energy of EV’s is also far greater than conventional vehicles. When the full life-cycle embodied energy is aggregated, the life-cycle benefit of electric diminishes significantly. Adding in solar panels increase the footprint further still. It’s still the case that a small, fuel efficient diesel/gasoline vehicle is generally more environmentally friendly than a medium-sized EV. A world full with one billion Tesla Model S’s (or equivalent) would be an environmental disaster.
The Seba hypothesis – that the auto market will completely switch to electric – is way off. The auto-ecosystem is much more complex. After 100 years of fine-tuning, the auto industry produces cars that motorists will buy. People care about cup holders. People are not going to buy an EV that doesn’t work for them. Full-sized diesel and gasoline hybrid can be bought with a range of 1,000 km and take 5 minutes to fill. How is electric going to compete with this?
The future of urban design and mobility is far more complex than often portrayed by a techno-optimist narrative. I see a more variegated world of different vehicle types, cycling, walking and public transport. The most efficiently-effective mass mobility type remains heavy rail – yes, the old nineteenth century, steel tracks, and inflexible routes. Nothing shifts more people more effectively. Urban form built around railway networks, with urban centres structured around cycling, walking and limited car travel is the most sustainable and livable. People who live on the urban fringe and further will continue to rely mostly on owner-driven liquid-fuelled motor vehicles.