City driving is awful, we’re failing at our emissions and vision zero goals, and outside of more working from home, we have not made much progress on these fronts. Fortunately, these issues are all fixable with small vehicles like e-bikes, scooters, and microcars, as well as a few thousand gallons of paint. Embracing the micromobility revolution will eliminate gridlock, reduce emissions, clear the air, and lower taxes.

Driving in the city sucks. Highways in major US cities are packed, so you’re often stuck in stop-and-go traffic averaging 20 MPH. On city streets it gets even worse: whenever you’re in a single-lane line of cars 5 deep at a red light, you’ve got a good chance that at least one of them will do a stoplight space out, box everyone out as they attempt to make a left, or box everyone out while waiting for pedestrians to mosey across the intersection when they want to go right.

Any of these situations will likely cause you to miss a phase, so you’re stuck waiting for another minute, and your average speed declines. These bottlenecks are a large part of the reason it takes forever to get anywhere in a city by car.
When you get close to your destination, you then have the joy of circling the block 5 times to find a parking spot, and your inner monologue goes like this:
 “Is that a… oh no, fire hydrant, dammit. How about right there behind the… nope, 30-minute delivery slot. Ok, let’s go down this street here. ‘No parking within 30 feet??’ Dammit. ‘Street cleaning Tuesdays.’ Today’s…wait, oh dammit, Tuesday.” You continue this game for maybe 10 minutes until you finally find something, then have to backtrack the 5 blocks to get to your destination. By this point, you’re late, frazzled, and your meeting doesn’t go well.
These are but a few examples of the frustrations of city driving. In many cities, car traffic moves below 10 MPH. In New York City, it’s around 5 MPH. Commuters in many cities spend countless hours a year sitting in traffic. The bottom line is that cars in dense and congested cities are a slow way to get from A to B. It doesn’t need to be this way.

You ever order something online like a USB stick, only to have a shoe box-sized package show up with a bajillion packing peanuts and then a little baggie with the USB stick in it? For urban transportation of 1-2 people, cars are the proverbial shoe box for a USB stick: they’re a much larger container than is needed for the payload and mission. This photo illustrates the problem well, with 60 people alternately in cars, a bus, and on bikes:

Most cars are overbuilt for urban transportation. They’re designed with ranges of hundreds of miles for speeds in excess of 100 MPH, and with safety equipment to make collisions at higher speeds survivable for their occupants. This makes them great at moving payloads around rural and less-densely populated suburban areas, but too large for cities. For urban use, these capabilities add unnecessary weight, complexity, and thus cost to the finished product. The second reason cars are too big is a range of subsidies for roads and parking that lower the direct costs of car usage while driving up costs elsewhere in the economy.

According to the Tax Foundation, the share of road funding across the United States from tolls, fuel taxes, and fees lies between 7% and 73%, with a 53% nationwide average. In other words, about half of road funding is subsidized by money from the general tax base. In my city, Seattle, we’ve got the West Seattle Bridge fix/replacement story, which looks likely to cost around $400 million, likely more by the time all the dust settles. We recently spent $3.3 billion on the SR 99 tunnel, a third of which came from state, federal, or local budgets. Many areas also have requirements for “free” parking, which according to Don Shoup is anything but: someone is still paying to provision the space to store the vehicle, in the form of more expensive products, rents, or spending tax money to store private property on public land. In sum, we’re spending huge amounts of public funds on infrastructure to move and store vehicles that are way larger than they need to be, and even then, the user experience is an awful crawl through gridlock and pollution. We can, and should, do better, by right-sizing vehicles for the mission and adapting roads to meet them.

All good engineering projects start with a clear understanding of the problem. According to Table 6b of the USDOT’s 2017 NHTS, average trips are 10.5 miles all-up, and 12.8 miles for commutes. 76% commute alone, though this rate is lower in many cities. The NHTS puts the total car occupancy for all trips at 1.55 in 2009. Pulling these together gives us a handful of requirements for a commuting vehicle:

• Range: 10-30 miles
• Payload: 1 person (100 kg), occasionally 2
  
• Speed: 20-30 MPH cruise (many cities don’t allow much more anyway)
• Size: no bigger than needed to accomplish the mission

If you send a bunch of engineers away with these requirements and tell them to design something affordable, they’ll come back with an e-bike: it’s a cost-effective way (Under $2k out the door and cheap charging) to achieve the range and speed goals for the average rider, potentially one with a bit of additional cargo. They help the rider stay healthy by requiring some level of effort, but have a configurable boost level to make hill climbs easy or to prevent the rider from being too sweaty when they reach their destination. In addition to having lower acquisition and operating costs, e-bikes have several other advantages over cars: because they are roughly 1/10 of the gross weight of a car, they consume much less energy to move their payload, which means they require less lithium for their energy storage, which in turn means lower pollution and less strain on electric grids to charge them. Around town, e-bikers can make use of bike paths, which often have fewer signals, and thus help people get places faster. Parking e-bikes is also much less of a hassle: many cities have racks or metal signs to lock up to, which means you save the frantic search for a curb spot and the ensuing 10-minute delay and jog. Sales show more people are agreeing that e-bikes are a better urban mobility solution, and countless more are expected to jump in the saddle in the next decade.
As e-bike adoption increases, we’ll need to focus more on right-sizing roads to accommodate the changed blend of vehicles.

One major adoption hurdle with e-bikes and smaller micromobility vehicles like scooters and powered skateboards in general is safety. Many people I’ve spoken with about cycling or e-biking in cities are nervous about getting hit, and even confident riders avoid certain areas they deem dangerous. I’ve written in the past that city planning employees should use the infrastructure they’re responsible for building and maintaining, which in places where implemented should accelerate the pace at which problems get resolved. One cheap and elegant solution to increase safety as well as to accommodate a growing number of e-bikes is lane doubling: cities should take a 10-foot-wide lane and paint a centerline stripe down it to make two smaller micromobility lanes for the growing number of small vehicles. Such a change will immediately double the throughput by allowing e-bikes to go two-abreast. Bottlenecks, like the frustrating left- and right-turn box outs will now be cleared:

Signal phases can be made shorter because each column of vehicles will be shorter by at least half, further increasing average speeds. In streets where there are two travel lanes and two parking lanes for a total of 4x10’, these can be redrawn to 8x5’, with e-bike/micromobility parking on the sides. Shrinking the width of the parking lanes would allow a third travel lane each way, which combined with the flow efficiencies would yield in excess of a 3x improvement in total throughput. Some lanes will still need to remain wider, for buses, trucks, and other vehicles.
Overall, lane doubling and intersection enhancements will allow a combination of increased urban densification and lower road spending. Wherever the throughput requirements can be met with narrower roads and bridges, we can spend less on construction and maintenance and in turn avoid many of the costly boondoggles like the SR 99 tunnel, West Seattle Bridge fiasco, and further examples in other cities. The billions saved in Seattle alone can be returned to taxpayers or reinvested in solving other problems like housing affordability. Smaller parking requirements will mean business owners won’t need to spend as much on that space, which will lead to lower prices for goods and services. Less traffic, cleaner air, lower prices, and less tax: let’s sign up for the micromobility revolution.

There may be a few of you who are not 100% sold on the urban micromobility utopia (shall we call it “umu?” Maybe not, this is why I don’t get invited to Marketing meetings…). As I’ve developed this idea, I’ve encountered a few objections that I’ve listed below, and several of you will come up with more. When formulating the objections, please consider the following question: “Does the cost of addressing [objection] make the entire idea bad on net, or is it a solvable problem that will eat up a small fraction of the billions in savings?”

1. Emergency vehicles: how will fire trucks/ambulances get through the umu? Same as always, sirens blaring and going through intersections when everyone’s cleared out. In fact, having fewer bottlenecks would probably help here.
2. Delivery vehicles: how will I get all my IKEA junk delivered? How will downtown restaurants get food supplied? Many European cities with central pedestrian zones have solved this problem: schedule time windows with lower traffic where larger vehicles are allowed to make deliveries outside of rush hour. Within a lane-doubled road, you’d need to have a sufficiently low speed limit to make it work out. Also, how many companies still pack USB sticks in shoe boxes when they could put things in smaller containers and vehicles? Looks like FedEx is already fixing that.
3. Paratransit or children: I have n people in my group who lack the mobility or ability to operate an e-bike or e-scooter, how should this be addressed? Small four-wheeled vehicles with tandem seating like the Renault Twizy, maybe stretched versions for an adult and two kids or driver+wheelchair, should cover most use cases. The Twizy is 4’ wide, so it would fit in a 5’ lane, but barely.
   
4. How do you expect me to ride an e-bike in the heat or cold? Cold weather is easy: bundle up. I wager some e-bike manufacturers have plug-in heated jackets and gloves, or are working on it. As far as extreme heat, I wouldn't want to ride outside in summer in Phoenix or Vegas either. For hot deserts, a Twizy with AC is probably the best solution.
5. I frequently drive longer distances and need the bigger car I have now, how will that work? Your situation is pretty rare. With the freedom to operate a larger and more dangerous vehicle around vulnerable road users comes an elevated responsibility. We’d rather you enter/leave the city during off-peak times, but when you are in town, please be aware of our new <30 MPH speed limit, the fine schedule for exceeding it for heavier vehicles, and the penalties for injuring others.
6. Re-fleeting for low-income drivers: how will someone cough up 2 grand for a new e-bike if they can’t afford it? If a city is moving to a lane-doubling model, most people should have enough time to sell their car and get at least $2k back, or save part of that in reduced gas and insurance. A city could create a loan program to cover the e-bike capex that would get paid back by saved vehicle opex. Doing some napkin math, if Seattle took the $400M from the West Seattle Bridge (just the tip of the iceberg), it could afford to buy 200k e-bikes at retail price and hand them out for free. At that volume you’d get a discount, and you wouldn’t need to buy that many anyway, so even with extremely pessimistic assumptions you’d come out ahead.

All of the objections I've heard thus far are either solvable for a price point far lower than the savings, or deal with with fringe use cases that should not be the primary issue to be solved.

Have you ever gone through a city and asked yourself “Who the hell designed this, and what were they thinking??” I have this thought all the time while cycling around the Seattle area, and often can’t help wondering, “Has anybody involved in planning or building bike infrastructure ever actually ridden a bike on said infrastructure?” The answer in many cases must either be “No” or “They must have a death wish.” Look at this gem:

The bike lane goes all the way up to the light and has two bikes stenciled onto it, which clearly means they intend cyclists to be there. On the opposite side of the intersection, there’s an ambiguous shoulder lane that just sort of tapers off on a 6% climb. Meanwhile, the speed limit sign in the distance reads 35 MPH (56 km/h), which means that people in cars are likely going at least 40 MPH (64 km/h). A moderately healthy cyclist climbs a grade like that around 10 MPH (16 km/h), which means the closing speed between car and bike are around 30 MPH (48 km/h) as the cyclist is unceremoniously forced into the “general purpose” lane where most motorists are not looking for them. This is a recipe for disaster, as getting run over at that speed is usually fatal, as seen in several studies and infographics:

Vehicular energy goes up as the square of the speed, so anybody on foot, on a bike, or in a wheelchair will have bad chances in such a collision.
This makes you wonder what the discussion in the traffic planning department actually looked like when they designed this mess.

“Say, Randy, what should we do about the bike lane on the other side of that intersection? Should we continue it?”
“Don’t know, Craig. The bridge tapers ahead, so we got a constraint, maybe the cyclists will just levitate to the other side of it.”
“Brilliant, why didn’t I think of that? I was thinking of making trade-offs like merging the GP lanes to keep the bike lane, but levitation is much better.”
“Yeah, we’ll save half a bucket of paint and a few posts by not restriping that. Sure a few cyclists unpracticed in levitation might get flattened, but the EMTs need practice using their spatulas to clean them up.”
“Sweet, let’s go to happy hour.”

The reason we end up with such bad infrastructure is that Randy and Craig don’t have a reason to care about what they’re doing. To be fair, they likely face a thicket of red tape and competing departments that won’t give them half a bucket of paint and a night to slap it on the road. Maybe the cone-laying union is on strike as well, who knows.

The solution to shoddy infrastructure is for cities to start dogfooding it: if Randy and Craig experienced the terror of merging their bikes in car traffic on that climb, they’d figure out a better solution faster. On their other rides around town, they’d start finding all the other death traps, like bike lanes in the door zone, tram tracks, and bomb-crater sized potholes on neighborhood greenways. They might look at their city’s budget and conclude that the Vision Zero project allocates more coin to its website than real improvements. In many of these situations, it’s more of a priorities problem than a money problem.

Within the first few weeks of the dogfooding initiative, Randy and Craig would experience firsthand where the most urgent problems are, and then work out how to address them. They might have a bit of a budget for speed bumps and signs on a neighborhood greenway, but, having ridden on it, decide that they should skip the signs in favor of spending that money on filling the bomb craters because that’s a greater help to making it safe and usable within their budget. They could likewise use budgets more prudently by alpha testing ideas with temporary solutions like toilet plungers and biodegradable paint, then getting a group of beta testers from the local community to play with it and provide real user feedback, instead of wasting millions on yet another ineffective design.

Infrastructure issues are not exclusive to cycling, though that’s where I personally see it the most. Sidewalks could be made a lot more useful with dogfooding as well. Here’s a wonderful specimen of a root bulge in a sidewalk:

The ADA spec calls for an 8.33% gradient for curb cuts for navigability, presumably because anything steeper is hard or impossible to climb. This bulge looks a wee bit steeper than that, not to mention the lateral unevenness. If Wilma’s in charge of sidewalks and is required to dogfood it in a wheelchair, this would become a higher priority. She might see the well-paved strip on the other side of the green and ask these types of questions:

• Why do we allocate budget for people to store private property (cars) in public?
• Why, three decades after the ADA, do we still not make wheelchair accessibility a priority?
• What problems have Craig and Randy found in their studies of this road, and how can we pool budgets to better achieve our goals?

They've obviously found enough budget for city-funded parking space at the expense of sidewalk usability, again showing us that this is a priority problem, not a money problem. The answers Wilma finds to her questions would lead to a re-prioritization of investment to ensure that sidewalks and bike lanes become navigable for all users.

Through dogfooding, infrastructure teams will form better empathy for vulnerable road users, and will become better internal change agents. If those they report to remain unconvinced of the new proposals, they should implement “Skeptics Day,” where the dogfooders can invite any city leader for a day out on the “need fixing” roads or sidewalks. If higher ups experience the misery themselves with the cameras rolling, it will be harder for them to retreat into their offices and ignore the problem. If they do try to duck their responsibilities, maybe Randy, Craig, and Wilma could make a solid case for that person's salary to be reallocated to something useful.

The examples here are Seattle-centric, but the problems and lessons apply to plenty of other places as well. If you want to introduce this in your area, get involved with your local cycling/walking/wheelchair advocacy groups, see what they’re doing, and pitch the idea of putting infrastructure dogfooding on an initiative in a local election. Voters across the political spectrum in many places think that infrastructure is shoddy, so the idea of focusing their city’s attention and having their leaders put skin in the game should be broadly appealing. If you’re in infrastructure planning and want ideas on where to find high-ROI areas to address, check out the Strava Global Heatmap for pointers on where heavy bike and foot traffic are, then do a segment search on keywords like “death trap” and overlay those with the high traffic areas to find some areas to put near the top of your list. Lastly, form a user group of locals to give feedback. The fact that someone will finally listen and commit to doing anything should energize locals to provide more than enough actionable feedback.