My recent blogs have been focused on how autonomous vehicle technology will evolve and when will it be safe.  I also touched on the regulatory framework that is evolving.  One of the most common discussion topics is their impact on cities. This blog will discuss current thinking about the physical changes on them.

While very recent numbers say that cities are losing residents to suburbs that is a residence issue not a workplace one.

In fact while one article bemoans the conversion of urban industrial space to offices instead of housing it ignores the fact that It is still workspace. So while suburbs are maybe becoming more popular again we need to figure out how AVs will work with this new displacement.


Traffic is getting much worse in downtown Manhattan,with average traffic speed at about 6 mph.  That’s 15 percent slower than 2010.  “Empty Seats, Full Streets” a report on fixing Manhattan’s traffic problem was prepared by Schaller Consultants and issued in December 2017 It was included in Gov. Andrew Cuomo’s Fix NYC advisory panel’s report in January 2018, recommended State Legislative adoption of congestion pricing in the Manhattan business district and a surcharge on taxi and for-hire trips in Manhattan.. TNC refers to Transportation Network Companies such as Uber and Lyft. The CDB, Central Business District, is defined as 60th Street to the Battery river to river. Here are some of the observations.

  • Taking into account the decline in yellow cab trips, the combined number of taxi/TNC vehicles on weekdays in the CBD increased by 59 percent between 2013 and 2017.
  • In the afternoon peak from 4 p.m. to 6 p.m., there are over 10,000 taxi/TNC vehicles in the CBD, more than double the number in 2013.
  • One-third of the vehicles are empty (“deadhead” my emphaisis), meaning between the drop-off of one passenger and pick-up of the next passenger, clogging the streets without any mobility benefit to anyone.
  • “Unoccupied” time between passengers experienced the most rapid increases, up 81 percent over the last four years. These contribute to the overall congestion without any benefits to the general population.
  • The dynamics underlying a driver-driven supply of service is likely to lead to excessive time spent between trips in cities across the country. Mixed human and autonomous driven fleets will further affect traffic.


OK, so with more cars on the street will we really get to the goal of future planners and get rid of downtown streets?  Architects and urban planners paint a picture of car-less streets open only to pedestrian and bicycle traffic. (Is that thing on the right side of the picture the tail of a vehicle driving in a bicycle lane?)  This has been done in limited areas but I don’t think it will happen over great swaths of urban space. It neglects the need for delivery drop-off spaces.

In much of Manhattan, particularly on the Upper West Side, the North/South avenues are wide commercial and the East/West streets are narrow and residential.  You might say the picture below might be typical As you move south some of the streets become semi-commercial.  You’ll never see an avenue without traffic and if you try to block off the streets you’ll put the residential buildings out of touch from delivery and repair services which tend to block street traffic as it is now. Where will the 18-wheeler truck park when a family moves into a three story townhouse in this picture?

How does and 18-wheeler moving van get in

Where will morning drop offs, evening pickups and day long deliveries occur in the commercial buildings in this streetscape that seems to go on forever?  Only if and until most people stop using their personal cars will this kind of thing be feasible and that could be three to four decades from now if ever.

Commercial Streetscape

Planners predict that the space for cars will shrink.  Below are are an existing urban street cross section and a proposed future one.  You will note that the road capacity is cut in half, 4 lanes to 2.  Why?  Because they project that fewer people will want cars in the city.  As I’ve described that’s not the case in downtown Manhattan.   Note also that the car lane width is reduced from 30 feet to eight. about the width of a standard parking stall.  There is a reason for the 30 feet.  The Ford Focus which is the basis for the Ford AV has a turning radius of 34 feet.  Eight foot wide lanes may make right turns very difficult.  Imagine trying to make a right turn out of a parking space..

Radway crosss section 2

Roadway xsection


There are scores of contradictory predictions. Here is a discussion of where most of the general wisdom lies.

In the beginning most manufacturers are planning for fleet sales to take advantage of the ride-hailing trend. But according a 2015 Slate article, most cities are suburban densities and the timing for ride hailing to your home for your morning trip to work may not seem so convenient.  Also can anyone actually predict how much sharing will be acceptable during rush hours, particularly for suburban riders? Can you imagine a commuter agreeing to going to two other locations for pickup and not being the first to be dropped off? Also as suggested by “Unintended Consequences” what happens to the AV after it makes it suburban delivery while its riders are eating dinner?

A website FutureCar quotes a Forbes article saying that AVs will reduce greenhouse gases.  I’ve not been able to find that article but since there’s no plan to park all these vehicles when they’re empty and they’ll continue to be operational unlike current practice it’s unlikely.  The only good news is that they’ll certainly be electric.

The elimination of parking lots and garages begs the question at the beginning of this blog- where do they go after rush hour?  The International des Professionnels de L’immobilier which translates into the International Market of Real Estate Professionals (who knew real estate translated into immobilier?) in an annual event in Cannes, France. At the most recent meeting of real estate dignitaries and designers an architect Carlo Ratti recommended building new urban properties parking above grade with no interior ramps and greater slab to slab dimensions to be convertible to future occupied space. He also predicted the demise of subways due to cheaper per mile for a driverless AV. On what basis does he price subway fares?  This ignores the fact that subways won’t have traffic jams and in general move at a faster speed with point-to-point travel.

There’s been much discussion about how millennials will address AVs.  Will they embrace them as part of the sharing economy and forego ownership for Uber and Lyyft?  Will they want to own them like their older generation?   Will there be enough living in the cities to change the overall ownership broadly?

Here are some of the problems in the predictions.


Really?   Certainly not in the short run.  Remember the existing number of cars currently running, 129 million passenger cars?  How fast do you really think that will disappear?  This inventory will take decades to shrink and every survey shows part of the population will still want cars they can drive.


E-commerce is still less than 20 percent of total retail.  Malls are going through changes that will continue to attract people. At the height of shopping hours traffic in and out is often jammed. Will customers have to wait for their Uber like theater goers when leaving what’s the point?  Of course if you can call your car remotely that will be great, but you’ll still need a parking lot.

Basically the AV “chauffer” mode is a as yet not understood supply and demand problem.


This headline posted by GeekWire quotes a study done by QuoteWizard a website that compares car insurance.

“Two commuters — one in Seattle and one in Denver — agreed to travel using only Uber and Lyft for a month to determine about how much it costs to get around that way. QuoteWizard compared that data with AAA’s estimated costs of personal car ownership and factored in the price reductions analysts expect to see when rides-hare services adopt self-driving cars. QuoteWizard estimates that ridesharing will be cheaper than owning a car by 2027, assuming Uber and Lyft are operating autonomous fleets by that time.””

QuoteWizard can be forgiven because they are clearly not in the survey business and this is statically so wrong. Opinions about car preferences typically include over 2,000 participants Too short and too small a survey pool to be statistically significant.


The industry’s response to the recent fatalities may not be the most effective.  Annual total deaths are news but I don’t think this is on commuters’ minds when the start off each morning. For that reason they may not see AVs as a major advance.  Also, they may revolt against being unable to buy their own semi-autonomous car, which has been the “sweet spot” in most surveys.

A Wired Magazine article by Jack Stewart “After Uber’s Fatal Crash, Self-Driving Cars Should Aim Lower” I believe correctly says that after the fatalities the technology companies will need support: from the public, politicians, and from regulators. The mantra that the technology will save lives has been reported so much that it’s lost its impact.  The benefits would only be confirmed after decades of driving when the presence of AVs has reduced the number or conventional cars.

However, if the industry promoted some common sense benefits they will be able to get the public’s attention.  Keeping drunk drivers off the road is understandable although how this would be enforced is not clear.  Keeping drunk teenagers safe is more likely to be effective at least for those who need to borrow their parents cars.  This could be a free service offered by Uber and Lyft.

The change in driving is understandably sucking up enormous time on the media. The future, however is further away than its messaging is telliing us.