Uber Uber Alles?

Because most of the existing U.S. housing stock is built in the suburban pattern, that is where the majority of Americans are going to have to live, for several generations. It is a high cost lifestyle because of the high cost of getting around in one automobile per adult, and the even higher cost of providing mass transit service to such low-density areas. For younger, poorer generations to live in these places affordability, different ways will have to be found to get around. To allow young singles and couples, and seniors, to live without owning their own cars, families with children to get by with just one, and moderate-income households, a growing share, to afford that one car. For trips outside of walking and bicycling range, this probably requires some kind of carpooling.

This is something I realized 20 years ago when, as an employee at the NYC Department of City Planning, I was asked to come up with a transportation proposal that would work for Staten Island, New York’s suburban borough. I suggested dynamic carpooling, with drivers and riders matched for trips using the technology of the time – touch tone phones and conference calling – with a fixed fare. As it happens, Uber has introduced a new carsharing application of its own, using the more advanced information technology now available. According to the Daily News:


Uber is introducing a carpooling service that will match riders in the five boroughs with other travelers going in the same direction, bringing down prices by as much as 50%.” Unfortunately, however, Uber’s computer app, even if it works great, would not come close to what is required to meet the needs of those seeking affordable transportation in suburban America. That would require a different plan.

I envision dynamic carpooling as something moderate- and middle-income households would provide for each other, making the cost of transportation lower by sharing it. The drivers would not be professionals seeking to earn a living by driving, but rather people making trips they would have made anyway and taking other people along, seeking only enough income in exchange to cover a share of the cost of the car. For that reason I asserted dynamic carpooling receipts would not be taxable income, and dynamic carpooling vehicles would not have to be insured as commercial vehicles.

For the riders, the cost of dynamic carpooling would have to be low enough to be a huge savings over owning their own cars. That would require a fixed fare that was only slightly higher than a comparable mass transit fare. And, since riders would be relying on carpooling for time-sensitive trips such as travel to work or picking up children at school, it would have to be 100 percent reliable. Because in most of the U.S. there is no Plan B available.

Now consider Uber’s new system, as described by the press.

It matches a driver with up to two riders, each of whom would be allowed to bring along an additional passenger for the trip. Fares vary by distance. According to the NY Post, which more accurately called the system cab sharing rather than dynamic carpooling,


The disruptive car service company says the plan, which has already launched in Paris and San Francisco, could lead to fares as low as $7.50 from Williamsburg to the East Village and $10 from Nolita to Lincoln Center.”

Here are the problems.

Let’s imagine that the average fare a rider would pay was just $7.50, and they made only 14 trips per week – five round trips to work, and two others, to shopping, doctor’s appointments, religious institutions, etc. That would be $105 per week, or $5,460 per year. That cost approaches the cost of owning an automobile, or an additional automobile, even in high cost Brooklyn let alone anywhere else. The rider would get almost no savings from relying on Uber instead of buying and insuring their own car, or adding another car to the family.

Then there is reliability. How about what I called the “chicken and egg problem?” If tens of thousands of people were already engaged in dynamic carpooling, riders could be reasonably certain of getting a match from their point of origin to their destination at a given time. But unless dynamic carpooling already existed with tens of thousands of participants people could not rely on it, and thus could not give up their own cars, and thus could not save money by using it.

Uber handles the problem of necessary trips and insufficient drivers by “surge pricing” — charging more at peak times to limit trips to those who need them the most, and are thus willing to pay the most. To Uber management, I suppose, this seems fair. But imagine someone working at a part time, $12.00 per hour job and needing to get to a five-hour shift. The other spouse has the family car at his or her job. Suddenly that $15.00 ($7.50 each way) Uber cost to get to that shift that will provide $60 in wages (minus taxes) turns into $50 ($25 each way) Uber cost. Or could at any time.

Bear in mind that according to the American Community Survey, in 2013 the median household income was just $52,250. The median cash earnings of full time wage and salary workers was $48,100 for males, and $38,100 for females. That means half earned less. And the median income for all workers, including part timers (voluntary and involuntary) was just $30,500.

Moreover if there was no match and people were stranded Uber doesn’t really even have someone to call, or not enough of them, because actual workers cost more money than bits and bites running through a server. So in most of the U.S. people who used Uber to get around would also need to have their own cars as back-up. But that would make no sense, because most of the cost of having one’s own car is fixed – purchase, financing, insurance, maintenance. Aside from a few places where parking is extremely expensive, such as Manhattan and San Francisco, once someone already owns a car (or a second car), and is already paying those all fixed costs, it doesn’t make economic sense to get around any other way. Paying to carpool can’t cost less than paying for gas.

At this point Uber drivers are generally professionals, not people earning a few extra dollars taking their neighbors alone. But even at a price moderate- and middle-income riders cannot afford as their primary means of transportation, Uber drivers barely get by on the money they earn. This has made Uber a political target for professional drivers who see their standard of living going down either as Uber drivers or drivers competing with Uber.


One factor in low driver pay and high rider cost is the very high $2.00 fee per matched ride Uber charges. This in a sector, the information technology/new media sector, where most “successful” companies give away their services for free, and hope to make money on advertizing. To the point where for current stock valuations to represent something other than another dot.com bubble, the U.S. will have to evolve into an economy where everything is free, but everyone who produces anything makes money by selling advertising, for products and services that people no longer need to purchase. The manic conditions created by zero interest rates have driven the market value for Uber’s stock to $41 billion, from what I read.


At that price investors seem to be assuming that Uber will be able to charge a $2.00 toll to everyone who goes anywhere anyplace on earth.

Finally, for dynamic carpooling to work as a mass system for ordinary people, it would have to be usable by non-college educated, working class people, seniors, and even non-English speakers. Not just hip college-educated twentysomethings seeking to early-adapt the latest app on their smartphone. The people who need an alternative way to get around the most need something very simple to use, and someone to help them with it. They can’t be expected to find the latest information technology easy to use.

In reality, Uber’s high cost per ride and limited reliability means it will be a niche player in the U.S., and not an answer to mobility issues for millions of moderate and middle income suburbanites, and for seniors trapped in the suburbs with decreasing ability to drive. At these prices Uber is really only useful as a way for affluent people who don’t like to drive to get driven around by working poor people. And for less affluent riders seeking an occasional more luxurious alternative to mass transit – as for a night on the town or a trip to the airport. An alternative to a conventional taxis ride every not and then, not everyday transportation relied on for basic mobility.

Uber might actually work in San Francisco and Manhattan – where there are plenty of conventional taxis and mass transit to be cannibalized, and to provide back-up when Uber cars are not available. It might also work in India, where there is a growing affluent, often non-driving population and plenty of desperately poor people willing to work for little. Uber is not a solution for those seeking to live affordably in the suburbs.

So let’s get back to dynamic carpooling as I envisioned it.

Right now the NYC transit fare is $2.50. So the dynamic carpooling fare would be a fixed $3.50, with $3.00 going to the driver, 25 cents to the information technology provider, and 25 cents to the marketing and service organization, something Uber currently lacks (more on that later).

Why would drivers accept the same fee for a one-hour trip from Suffolk County NY to a job in Brooklyn, as for 10-minute trip to the supermarket within the same suburban town? Because each trip would be a trip the driver would have made anyway, not a trip they are making to earn a living as a driver. The only extra driving done specifically to accommodate the riders would be the pick-ups and drop-offs. The pick-ups and drop offs could take just as long for the 10-minute drive as for the 1-hour drive. So the fee would be the same in either case.

For the riders, the seven round trips per week (14 trips) would cost $49 per week or $2,550 per year, less than the cost of owning a car (or an additional car in the family). For the drivers, taking two riders per trip for 14 trips per week would earn $4,370 per year – more if three riders were occasionally matched, less if only one were matched from time to time. Not enough to live on, but enough to make owning that car more affordable. And, of course, the same two-adult family could participate in dynamic carpooling on both ends, with whoever had the car pickup up riders, and whoever did not have the car using dynamic carpooling to ride as a passenger, rather than having the family own a second car.

The information technology provider doing the matching would get just 25 cents per trip, a fraction of the amount Uber now charges. But with far more people using the service, it could match far more trips.

And what about the sales and marketing organization? My view is that at the metropolitan area level, dynamic carpooling should be organized by a non-profit membership organization, a dynamic carpooling club. There are two reasons for this.

First, dynamic carpooling is what economists call a “natural monopoly.” It makes sense for riders to use the system that has all the drivers, and for drivers to use the system that has all the riders, to make it certain that those traveling from particularly places to particular places at particular points in time could be matched up. Overcoming the “chicken and egg” problem would be difficult enough, at least at first, for one dynamic carpooling system, let along two or three competing systems. The fact that in reality only one system should survive in any given metro area might explain why the competition between Uber and Lyft has been so fierce.

Monopolies have a tendency to become abusive, whether they are single corporation, business association, government, union, or even non-profit. If dynamic carpooling were adopted on a large scale, the dynamic carpooling provider would have the power to strand a large share of the workers/consumers in a metro area. It could double what it charged, or shut down and demand subsidy money from the government to restart, perhaps enough that everyone in the organization got to retire with a big pension after working just 20 years. Like a certain public transportation union in New York City, an area that depends on a public transportation monopoly for travel to Manhattan, did not long ago.


Given the way those who control our public and private sector organizations have been abusing their power in recent decades, it would be a mistake to hand monopoly power to an organizations whose leaders are seeking to earn uber money so they can have uber sex with uber women, from what I read.

More importantly dynamic carpooling would have to become a social movement to succeed, not just a business or an app. It would require people to talk with other people, to show things to other people, to make them feel like part of a movement with other people. Carpooling would have to be seen as a way to be part of a community, and meet one’s neighbors – the ones that also worked in the same area, shopped in the same area, or attended the same religious institutions. Could that happen through social media alone? Perhaps for college-educated people in their 20s, but not for the working poor and seniors.

A social movement is what Uber lacks. According my proposal 20 years ago, the social movement would be led by the people who received 20 cents every time someone they had recruited to dynamic carpooling took a ride in someone else’s car. Someone who lived nearby, in their neighborhood or suburban town. These people would be professionals working on dynamic carpooling, and in exchange for their 20 cents per ride they would not only have to recruit and show people how to use the system, but also pick people up and drive them – for the fixed fare – whenever a match could not be made. They would be the back-up in the suburbs and small cities out in the middle of the country, where transit and taxis are either not available or prohibitive expensive.

To understand what is required, and how it could work, I suggest this documentary on Tupperware – which needs to be viewed before mid-February.


Tupperware was sold directly to neighbors by housewives seeking to earn a little extra money. Dynamic carpooling could be sold directly to neighborhoods by younger retirees who spend some of their time driving older retirees around, and late career job losers who need an income source to keep them going until they can afford to retire.

Even with a large, home-based group of leaders, the dynamic carpool club would still face the “chicken and egg” problem – it would already have to be large and established to be useful. So it would be helpful if it were organized by an existing group that already has many members. Back in 1994 I had proposed NYC public employees, who are concentrated on Staten Island, but that was before all the retroactive pension increases and their consequences took the public employee unions out of solidarity with ordinary people. One might as well ask Goldman Sachs workers to share rides with busboys and dishwashers as to expect the free parking placard holders to share rides with anyone else. How dare you!

So who else could serve as the foundation of the club? Perhaps public employees in places where the unions are less predatory and the government is more progressive. Raleigh-Durham, Austin, Madison Wisconsin or the Twin Cities for example. Perhaps members of the military and retired military in places where they are concentrated, such as San Antonio or San Diego. Perhaps religious organizations with many dedicated members who are willing to help each other, and the broader public, out. Maybe if enough people who weren’t in the one percent, or profiting from an insider connection to the government, started meeting up with and helping out each other, things could evolve from there. Put a little soul in the suburbs, with the kind of social interaction so many young people are desperately trying to move to the cities to fund.

Once the dynamic carpooling club was organized for a given metro area, it would hire the information technology provider, offering that 25 cents per ride. Might Uber take that role?

If Uber’s new ridesharing app works that is a substantial achievement. Matching more than one combination of origins/destinations in an efficient way is a rather complicated piece of mathematics, particularly if an element of time is introduced (and those who fail to make an optimum match are nonetheless matched with someone after waiting a certain period). I’d be interested to know how they do it, and if they stopped at matching two riders rather than three for business reasons or because the calculations would otherwise be too overwhelming. Moreover, Uber has background checks, handles payments, and allows riders to avoid future rides with drivers they were not comfortable with (and vice versa?), all features I identified as necessary long ago. I expect the firm will improve all these features if it survives.

In the end, however, I now expect that the primary dynamic carpooling service is not going to be Uber, due to its public relations problems. It has been a long time since WWII, but choosing a name that many associate with the Nazi party was not a good start. I’m not sure what its other public relations problems are, but I do know that any organization seeking to take on vested interests with their hooks into the government needs overwhelming public support to have a chance. It has to be pure as the driven snow, because it will come under attack from those interests, and from the general anti- anything new clickbaiters in the media.

As it stands Uber is a threat to the regulated taxi industry, and is thus under attack by that industry. If it had instead attempted to create the sort of service I had proposed, it would have instead threatened to reduce the number of private autos sold, insured and gassed up, and thus would have come under attack from a whole different group of powerful interests.   Who would like nothing better that to turn the public against such threats, to provide cover from their lobbied politicians to stamp them out.

The example of what might befall Uber is Myspace, the first social media platform. As its popularity surged, the media reported that creeps were using it to stalk underage girls. I remember telling my daughters they couldn’t have a Myspace account. Sometime later, those daughters told me everyone at school was getting a Facebook account. It’s just for students! they said. We’ll need to use it to work together on our homework! But in the end Facebook was no different than Myspace. Facebook just benefitted from coming along later, and avoiding the backlash.

That same generation is Uber’s current market. When my daughter moved to another city, she asked if I had heard of Uber, and if I thought it would be a good idea to use it. I said that since she was new to that city, and didn’t know her way around, she should use a regulated taxi for a trip she didn’t want to make on mass transit. Rather than take a risk with a service and drivers for which there were no assurances. She tried the service anyway, since her peers were using it. But after negative news came out, she told me, she deleted the app and will no longer use the service. All the information that led her to start using Uber and stop using Uber was circulated within that age group, with us older folks knowing nothing about it. I guess the lesson is if you aren’t a social movement, but rather a $41 billion corporation, there is always the possibility of a social movement against you.