With new tech coming online, cities need a department of urban testing

INSUBCONTINENT EXCLUSIVE:
Shaun Abrahamson Contributor Shaun Abrahamson is a managing partner at Urban Us Ventures and serves on the
Investment Committee at URBAN-X. More posts by this contributor How should startups work with city governments The design
and operation of cities is the province of urban planning
But an explosion of startups in cities means a lot of new products and services for urban areas
The problem is, we don&t really know how people are going to use these new products and services
&The company launched a trial service in Santa Monica just last year and when I first saw the scooters (parked literally outside of our
office) I was convinced nobody would want to ride them…The volume grew so steadily that I finally hopped on one, rode down to Bird offices
and pleaded with Travis to take money from us
I had literally never seen a consumer phenomenon take off so quickly,& says Mark Suster in All The Questions You Wanted Answered about Bird
Scooters and Their Recent $300 Million Funding
There is no doubt Santa Monica scooter usage has benefited from a significant investment in bike lanes, as you can see from this map
If you doubt this, take a look at what happens as you travel a little outside of Santa Monica, where bike infrastructure doesn&t exist
Scooter riders take to sidewalks, just as they do on bikes
(I suspect data would also reveal less usage in these areas and that there are a lot of complaints in these areas about the scooters.)
Adapting to new behavior Just four-and-a-half years ago, people were stunned by the then huge valuation of Uber at over $3.5 billion
It took Uber just over four years to get to that valuation
Now Uber is acquiring electric, dockless bike companies and investing in shared scooters, as they dial back their self-driving activities
The chart below explains why. The practice of sharing rides came out of nowhere five years ago, and two years ago, docked bike share
seemed to be rewriting what was possible with on-demand transportation
(Let see if/where scooters show up in Q1:19.) It hard to appreciate the impact of these new mobility models until you look at them against
some well-established existing modes of transportation like taxis and car rentals
These industries trace their roots back to the explosion of the automobile with Model Ts a century ago and to when entrepreneurial companies
like Hertz decided taxis should be yellow to enhance their visibility for hailing
(And yes, that the same yellow Hertz car rental sign.) As the chart below shows, these 100-year-old industries are rapidly melting away in
the face of on-demand competition. Revisiting city planning What does this mean for city planners For a long time in the U.S., cars have
so dominated public spaces that we&ve mostly focused on parking and roads
We forget that companies lobbied for the car — they hadn&t always dominated public spaces, and now cities around the world are steadily
pushing back
With new mobility options come new corporate alliances
&San Francisco has more than 250,000 on-street parking spaces for ‘dockless& cars, so why does the addition of a few thousand dockless
scooters spark such a heated debate& says Roelof Botha, who is also an investor in Bird
Parking for small shared EVs is just the start
What about the impact of last-mile options on public transit usage Lyft &friends with transit& policy could be viewed cynically as a way to
win favor with regulators, but last-mile trends are also playing out with smaller on-demand vehicles
In Germany, DB figured out the relationship between bikes and transit more than 10 years ago with the first dockless shared bikes, and the
massive bike parking infrastructure connected to Dutch transit. In the city most associated with car culture, you can now stop by any Metro
station along the Expo Line in LA and see a growing number of bikes and scooters
Is this associated with an increase in transit usage Is this another reason to sort out parking for small, last-mile vehicles With more
venture dollars in the mix, there are now strong alliances seeking to test new approaches
Maybe we just forgot history Just before cars finally took over most public spaces for transportation, there were more public transit
options provided by some weirdly familiar looking vehicles
You can almost see how this 1922 Austro Motorette would eventually be given a seat to become the more familiar-looking Vespa
There are some important differences now
Electric vehicles need new charging infrastructure, which has led to debates about who will pay for it
On the other hand, scooter-sharing companies have learned infrastructure doesn&t need to look like charging stations
Small vehicles can have their batteries swapped out (as with the scooter share Coup) or companies can offer incentives to people to take
vehicles home to charge (and then return them)
Electric vehicles will continue to drop in price as the main cost — batteries — become cheaper
Finally, shared vehicles have never been more available, and electric scooters are almost 100x cheaper than cars
What happens to behavior when availability is so high that it doesn&t require you to leave your block to find a vehicle Bad old
assumptions It seems inevitable that AVs will replace drivers at some point
While we wait for the takeover, there are some approaches we can take to improve safety
Unfortunately, some of these seem to have unintended negative consequences
Growing up I used to visit Zimbabwe, where speed bumps were called &sleeping policemen& because it was believed that they caused drivers to
slow down and obey posted speed limits
But data from a recent NYU and Urban Us portfolio company,Dash, reveal that &drivers have a tendency to accelerate quickly after traffic
calming infrastructure like speed bumps, which can lead to dangerous situations for pedestrians.& This is potentially a new issue as more
drivers discover previously quieter streets via navigation apps and it should still force us to revisit how we try to manage drivers who are
going too fast
Beyond mobility With NYC L train shutting down in 2019, people have started to plan their housing options and new commutes to and from
Manhattan
The city has responded by saying they will add additional ferries and CitiBikes
But is there a way to add more options quickly How might they fare in the winter It hard to know, but the consequences go far beyond
mobility to impact everything from restaurants to residential real estate
Mobility ultimately has a lot of consequences for real estate
Are people going to live in smaller private spaces in exchange for better shared amenities Will they use robotic furniture to get more from
smaller spaces What might cause homeowners to purchase backup power in the form of a home battery system How will this impact the grid Will
people be OK sharing sidewalks with delivery robots With so many questions about how our behavior will change, we need to find better,
faster ways to test new solutions
Maybe alongside departments of urban planning, we need departments of urban testing