Bo the Independent

When Bo cycles, wrapped in her leather jacket and airpods, she feels free. She listens to the same music while biking as she does while programming: progressive metal, chamber pop and occasionally Hans Zimmer soundtracks (depending on her speed). She just moved to a new city for a job.

She likes the job but hates the commute. It's too far to bike and they don't really seem to do transit around here. Her company has this shuttle that could get her to the office, but she doesn't like the idea of being on someone else's schedule. Or being cooped up every day with a bunch of randos.

She drove down in her parents' old car and she's been using it for now until she figures something out. She's looking to buy her own car, but she'd honestly rather go through the agony of a second snug piercing than deal with a car dealership and figure out the financing and insurance. Why isn't there like Netflix for cars?

She hasn't really worked out the food situation. Mostly she's been getting microwave dinners from the nearby supermarket and ordering delivery through Caviar. She's going to have to figure out how to cook before her mom visits next month, because she knows she'll make a fuss if she sees the kind of junk she's been eating (she's coming under the pretense of taking the car back, but everyone knows this is an inspection). One of her coworkers just told her about a service called Freshly that delivers meal prep kits - maybe she should try that!

Kari the Nurturer

Yoga on Tuesdays and Thursday is when Kari feels zen. But it's HIIT on the other days that gets her through the week.

The boys have a busy schedule. She likes to joke that people call her a helicopter parent, but that would be silly since she drives an SUV. Most people don't get her joke but she tells it anyway and still finds it funny.

The boys are both really into soccer. Sven's already pretty good and has a game most weekends. Hanson has less of a natural talent for sports, but he's pretty good at the violin. It's challenging for Kari on afternoons when she has to stay late at the lab. Joren helps out during the week when she asks him to and has the kids every second weekend - they're very civil and focused on what's best for the boys. She's also been doing a carpool with a few of the other parents in the soccer league but she doesn't like that she's usually the one arranging the schedule. Just because she's a physicist doesn't mean it's quick for her to figure out this periodic table (another one of her jokes that doesn't always land). Recently she's been looking into a service called HopSkipDrive to help with some of the extra trips. But she's not crazy about this idea of outsourcing parental responsibilities to an app with such a silly name.

During the boys' soccer practice, she usually spends a bit of time browsing Pinterest and will occasionally buy something that catches her eye. She likes "forgiving" (i.e. low-maintenance) houseplants and has a thing for colorful headbands (she usually goes straight to Etsy for these).

She's also recently started ordering groceries through Instacart which saves her all that time (and stress!) going to the supermarket. Most of her other shopping she just does direct through Amazon. Amazon Prime is already a lifesaver since she hardly ever goes to the mall. They also do takeout Tuesdays (and sometimes Fridays when she wants to spoil the boys).

If only she could also order Thai massages to her home. She remembers their vacation two years back in Indonesia where she used an app called GoJek to do this. It was just before the split and there's not much else about that vacation she remembers fondly. But the massages were great.

Espen the Stoic

Espen is happiest playing with his granddaughters, carrying them on his shoulders or spinning them around as they roar with laughter. It's become harder to play these last four months since he started dialysis.

When he was first diagnosed with CKD, he would drive to his doctor's office for checkups. But as his kidneys got weaker, he started having bouts of fatigue and nausea. Ame insisted on driving him to these appointments. He enjoyed these trips with his son, especially their conversations about political history. He had been working his way through Julian Jackson's biography of Charles de Gaulle and he found the idea of French resistance inspiring at a time he was fighting his own battle. But when the dialysis started and he needed to be at the clinic several times a week for hours at a time, it was clear they needed to find a different solution.

He considered taking the bus, but it was far for him to walk to the nearest stop and Ame fretted about him taking public transport with his weak immune system. His doctor suggested a paratransit service. It got him there, but it was slow and unreliable.

Then Ame's partner Britte read that his health provider could cover the cost of Lyft rides classified as non-emergency medical transportation (NEMT). He likes the service so much he even uses Lyft to get to the nearby park he used to walk to before it became too hard. He now meanders around, supported by the cane his granddaughters gave him as a present. He stops periodically to drop little pieces of bread, enjoying the way the baby ducks peck at it and then scurry off behind their mother. He also loves having a Kindle since it makes it easier to take a book with him to read in the park or while waiting for the dialysis machine to do its work.

Espen is quite excited about all this newfangled technology and how it has made things easier for him. He recently discovered that he can do his check-ups with his doctor through the computer and then get a company called PillPack to deliver the medicine. And he likes to read about autonomous vehicles in the news, but he's not sure how soon they'll be ready. He's gradually moving up the transplant list - maybe he'll live to benefit from them. If not, at least his granddaughters will.

Nissa the Magnetic

When Nissa dances, everything around her stops. Others are drawn to her and start moving to her beat. As she moves, her necklace moves with her. It's made of pearls, a kind of retro-cool she manages to pull off.

Nissa loves theater and improv. And she's excited about this new Unagi scooter she got for her birthday. She likes Unagi in spite of the fact that Billie Eilish has one. It makes it much easier to get to theater rehearsals and to her friends' houses afterward without having to depend on her parents for rides. Nissa's friends are mostly theater and dance crew, except her boyfriend - he plays basketball. He's the one who got her into Fortnite. She thought it was silly before she tried it, but it's not just about shooting. It's basically a place to jam online and she's not that bad at it.

Theater parties are the coolest parties. Nissa's parents know she drinks, but trust her to be responsible. But her mom insists that she takes Uber back from parties because she wants her to be safe.

Nissa's friends love that she's grounded and is a good listener. She also cares and not in a fake, I want to have more social media followers kind of way. She volunteers at an animal rescue center. She actually doesn't like posting on Instagram and TikTok - she feels sucked in and it gives her anxiety - but it's a good way to promote her shows and help raise awareness about the rescue center.

It's weird to think that her parents grew up without smartphones and met at a bowling alley at the mall. Nissa goes to the mall when she needs to try new makeup at Sephora and that's about it. But living life around the mall like her parents still kinda do? That's weird!


< Intro: The problem of automotive inefficiency

Part 1: The trip economy

Though vehicles can move quickly, they have changed relatively slowly. Yet in recent years disruption to the industry has become quite dramatic. Ridehailing startups experienced a meteoric rise to prominence. Then Bird reached a billion-dollar valuation faster than any startup before it. And Tesla is now the most valuable carmaker in the world after an incredible stock rally. Though change in transportation takes time, valuations move faster since they attempt to price in what is to come. As William Gibson famously noted, “the future is already here — it’s just not evenly distributed.” In the case of mobility, that future includes two critical factors: trip pricing and trip unbundling. Understanding these concepts and all the factors feeding into them helps shine a light on where we're headed.

Trip pricing

An drawn image of a dog performing the various events of the decathlon

Cars are like decathletes: they aren't the best at any one thing, but they perform well in aggregate

Cars perform well across various use cases - be it commuting to work, picking up kids, going on a road trip. Like decathletes, they aren't the best at any one thing, they perform well in aggregate. And since people have a wide variety of mobility needs across their lives, cars have become the catch-all answer to transportation for most consumers.

I’ve written elsewhere about how the shift from purchasing vehicles to purchasing trips (perhaps bundled with services) is reshaping the business model of carmakers.

Understanding business model disruption in the mobility industry

This shift also has a direct impact on efficiency. Historically carmakers have had very little reason to concern themselves with efficient asset utilization since their customer relationship involves selling a vehicle every few years, usually through a dealership.

The business model of carmakers: suppliers supply parts which then then assemble into vehicles which are moved to dealerships which sell them to customers; carmakers therefore don't have a direct relationship with their customers but rather an indirect relationship intermediated by their brand

Meanwhile for the customer, once they have sunk a large sum of money into the purchase of a vehicle, their primary concerns are around convenience, not efficiency. The psychology of the large sum nature of buying a car means that though you might intend to pay $35,795 for an entry level 2016 BMW X1, on the day you make this purchase you might feel compelled to add a sunroof, leather seats and choose a special color, bumping the price up to $48,970 in the process, the cost of the fully loaded version of this vehicle. The difference between those two price points amounts to the equivalent of 42,329 pop tarts. And that's not factoring in the recurring costs of owning a vehicle which consumers don't fully consider at the time of purchase such as fuel, repairs and maintenance, insurance and the hidden cost of time spent driving. Most of the key decisions happen at a single moment and humans tend not to act fully rationally when making decisions that involve big numbers.

Drawing of two pop tarts (Strawberry and Wild Berry flavors)

pop tarts, an American delicacy

When the model shifts to selling individual trips (as Uber and Lyft do), the calculus on the part of consumers shifts dramatically. Every interaction with a vehicle is a purchasing decision for which substitutes are readily available. In this scenario, people care a lot more about marginal cost. Indeed the psychology of buying a trip runs the other way to buying a vehicle: frequent users of Uber and Lyft might switch from one service to the other because the one offered a trip for $12.89 and the other offered the same trip for $12.64 (a difference of not even a single pop tart).

Crucially this "trip price" includes all the costs of operating the vehicle along with the driver’s time.

A listing of all the costs of vehicle ownership and the varying frequencies you have to pay them (cost of the vehicle, insurance, licensing, fuel, parking, etc plus the cost of time spent driving and wasted in congestion) compared with the simplified single cost of each trip when you make use of the trip economy

Ridehailing companies are actively competing to win market share. Over time, competition over trip prices is forcing them to optimize on each input they control that is subsumed into this price. For instance:

  • Vehicles: Uber’s tremendous success has been built on the back of a simple arbitrage: generating demand for a fleet of privately owned cars whose costs have already been sunk. The fundamental value proposition of UberX is to take a car that is used 4% of the time and increase that utilization and share the benefit with the driver.

  • Seats: UberPool built on this arbitrage by increasing the average number of passengers traveling in a vehicle in a given direction thus increasing the utilization of the vehicle (in technical terms, increasing passenger miles traveled (PMT) while decreasing vehicle miles traveled (VMT)).

  • Assignment & routing: Many aspects of Uber's technology stack are about optimizing UberX and UberPool by ensuring that assets are effectively allocated and routed to maximize utilization.

Beyond this, ridehailing companies are also thinking about aspects of the cost of trips that are harder to control directly: Uber and Lyft have initiatives to increase EV penetration and are looking for ways to augment driver revenue through selling goods in vehicles (although this has proven tricky) and advertising on vehicles (which is apparently working better).

It is important to note that ridehailing (at least in developed markets) is built on the foundation of traditional car ownership. Drivers are offering a service with cars that they own and are therefore responsible for the cost of maintenance, fuel and financing and generally more aware of the true cost of operating a vehicle. It’s not an accident therefore that the super-efficient Toyota Corolla is a popular choice for Uber drivers.

The key here is that trip pricing rewards the reduction of inefficiency in car utilization. This shift is primarily tied to business model: because consumers are evaluating prices every time they make a trip (rather than once every few years when they buy a car) and because the trip price encapsulates the overall cost of supplying that trip, competitive dynamics between trip operators force them to find ways to optimize the use of vehicle assets.

Unbundling trips

Uber and Lyft leveraged smartphones to make a car appear at the tap of a button. The ability to digitally hail a car also brought with it the possibility of using other vehicles for other purposes. In many cases, a car is not the best solution. Depending on the circumstances, walking, a bicycle or a scooter might be a more effective way to complete a particular trip. Especially as cities become congested, smaller vehicles may be faster and more effective. Meanwhile a growing amount of micromobility lane infrastructure has been set up, supporting these kinds of trips.


A drawing of cats doing the sprint and the pole vault

The first kind of trip alternative to cars came in the form of docked bicycle sharing networks such as Paris' Vélib or New York's Citi Bike, but bicycles really started to scale with the rise of Chinese free-floating bike sharing providers like Mobike and Ofo. These solutions proved there was significant demand for smaller (and cheaper) form factors, but given that these vehicles weren't optimized for many markets and the rollout was too aggressive even in markets where they were a fit, both these companies ultimately flopped. Then came scooters, a stronger product-market fit and a lot of capital to supercharge the global expansion.

A key aspect of micromobility, implied in the name, is the idea of a vehicle specialized or "right-sized" for the required trip. Scooter and bike sharing operators have rapidly iterated on the vehicles they deploy, evolving from light frames that lasted only a few weeks to sturdy vehicles that are easy to maintain and include swappable batteries. Rather than decathletes, these vehicles are more like sprinters and pole vaulters. Like with ridehailing, these services are subject to the same forces of competitive pricing that drive towards the maximization of efficiency. Right-sizing and the rapid evolution of micromobility form factors are a consequence of this effect. In contrast to ridehailing companies, micromobility fleet operators have far greater control over their vehicles and therefore the forces pushing them to improve efficiency are stronger and the evolution has been faster.

As consumers increasingly look to apps to complete trips, there are new form factors still emerging, including speedier and more comfortable vehicles that include storage space better suited to longer trips, such as electric bikes (Bond, Jump), electric mopeds (Revel) and strange creatures such as Wheelz and Bird's Cruiser which fall somewhere in between. While the car ownership bundle means that there is often a mismatch between what consumers need and the vehicle platform they are using to complete a trip, micromobility corrects for this.


A drawing of cats doing the sprint and the pole vault

Trips don't only involve getting people to places, they also involve getting things to people. Fulfillment for logistics has existed for a long time, but the desire for rapid fulfillment is what substitutes most obviously for a car trip where you'd otherwise drive somewhere to pick something up. Food delivery services such as DoorDash, Postmates or JET and grocery delivery through Instacart, Amazon Fresh or Ocado are good examples of these kinds of trips.

As these services mature and the subsidies used to drive growth ahead of profitability fade, the same pressures to squeeze out greater efficiency powered by the trip pricing model dominate. Bundling orders and increasing delivery rates are the best way to improve efficiency, but over time the vehicles that couriers use are also a key factor in the equation. Much like ridehailing, these businesses are built on a model of gig work so it's harder for the operators of these services to directly manage their courier fleet, but their incentives remain strongly aligned in order to find ways to do so. Companies such as Zoomo, which partner with delivery services to lease professional grade e-bikes to couriers, are directly addressing this challenge. Meanwhile, ground delivery drones such as the ones Postmates is testing are also a way to drive down the unit economics of delivery. These vehicles are, so to speak, the hurdlers and high jumpers in the unbundling mobility equation. But they form a part of a broader trip economy.

A Lexicon of Trips

So micromobility and delivery are two ways in which the trip economy is becoming a part of our lives, changing how consumers move (or don't move) and how service providers repurpose vehicles to these specific needs.

Let's meet a few interesting characters and consider the different kinds of trips they might take with the goal of building out a more comprehensive framework. (All names are selected from this wonderful Disney list of Scandinavian names, which I happen to be partial to.)

Click to expand each card!

Bo the IndependentKari the NurturerEspen the StoicNissa the Magnetic

Trip purpose

Trips connect the pieces of our lives together. These stories help outline how individuals navigate a very wide variety of trip options in a world of increasing choice. There isn't always a single "right" solution, but individuals respond to choices. The options available to them in turn are a consequence of those choices playing out over time. From these stories, each representing a major category of trips we take, we can build a framework for better understanding the rapidly emerging trip economy (retail trips span all characters).

A visualization of the five primary trip types from the home: Work, Education, Retail, Medical and Social

Here's a comprehensive (although of course necessarily reductionist) list of trip types:

  • Work: Includes commutes to the office as well as business travel
  • Education: Taking children to school or daycare, attending university or other educational and extracurricular activities (including those related to religion)
  • Retail: Trips with the goal of purchasing goods or services (including food and groceries)
  • Medical: Trips to medical care including medical checkups, hospital visits and dental appointments
  • Social: Visiting friends, family, going on vacation and other social or recreational activities
  • Other: This is a catch-all category for any other kinds of long tail trips people might take as well as trips taken for their own sake (e.g. wandering, joyriding, road trips)

The US National Household Travel Survey gives some insight into the breakdown of these trips amongst American consumers.

A visualization of the five primary trip types from the home: Work, Education, Retail, Medical and Social

Source: 2017 National Household Travel Survey

These categories are aligned and colored according to the framework above. Trips home are the return leg from most trips or combination of trips, and therefore the predominating category. Transporting someone else is a tricky category since these trips can fit any of the other purposes for the person being transported, although many of them will likely be parents transporting their children on educational or social trips.

Demand side factors

In America, the vast majority of trips are made in a car. When considering how the trip economy might substitute for car ownership, it is helpful to consider the factors that define a particular trip's needs, many of which were highlighted in the preceding set of stories. Here's a list of the requirements impacting the demand side of any particular trip.

Utility requirements

Distance (how far?)

Trips vary in length, but most are quite short. In fact, the vast majority (about 77%) are less than 10 miles and about 60% of these trips are just 3 miles or less. Cars, being decathletes, can cover a significant range yet most trips don't require this.
A bar chart showing the distribution of trips according to trip length, with the majority of trips being less than five miles in length

Source: 2017 National Household Travel Survey

Urgency (how fast?)

Mostly people would prefer to travel faster than slower, but not all trips are equally urgent. Espen's trip to the hospital after falling down would be a lot more urgent than Kari driving with the boys to the beach. This is especially true when it comes to delivery trips: hot food requires a much shorter delivery time than groceries or packages. Speed may also matter within a band of reliability: you might not need to be there immediately, but you can't under any circumstances afford to be horribly late. The degree of urgency in turn will affect which mode is the most effective for the trip.

Regularity (how often?)

Some trips, like going to a wedding, happen only once. Others, like commuting to work, happen repeatedly with very little variation in timing or route. People like Bo who make a trip repeatedly have a stronger incentive to optimize that trip.

Cargo (what is being moved?)

Different trips require different things to be moved. Plumbers for instance need to bring with them an array of wrenches, plungers, augers and hydro-jetting equipment in order to do their job. Large families require a lot more space than individuals and also special seats for young children. And trips to the store usually involve bringing things back home. Logistics trips meanwhile have very specific cargo requirements with some variability (e.g. hot food vs radioactive uranium waste vs horses).

Comfort (what is expected?)

Most people want their vehicle to not smell horrible but users have varying expectations and tolerances for different aspects of their trip comfort. Some like Espen may want to exert as little effort and have as much space as possible while others like Bo may prefer to turn their commute into a workout or might not mind a squeeze. Temperature and weather conditions affect trip comfort as does the smoothness of the ride. Safety might also be a factor, including when the cargo is valuable (e.g. cash-in-transit) or vulnerable (e.g. children).

Inconvenience & other cost factors

Operator (who's driving?)

A paradigm shift of the trip economy is that people become a kind of freight: something that needs to be moved. But most people have the ability also to operate vehicles themselves, which raises the question of who is operating the vehicle: the user or someone else? Cars are usually driven by their owners, but Ubers and taxis have drivers. Micromobility assumes a user will be driving, but for food or grocery delivery someone else brings things to you. From the user's perspective (even though this is often overlooked) there is an opportunity cost to operating a vehicle: you could be doing something else during the time you spend driving. This is especially true for the almost 10% of trips that involve transporting someone else. Conversely, for users that can't drive and are dependent on others for mobility (including those with disabilities, the elderly or children), having an operated service is an appealing way to increase independence.

Outbound vs Inbound (coming or going?)

Connected to the question of the opportunity cost of making a trip is an important distinction between outbound trips where the primary objective is for the user to go from their home to somewhere else (car trips, micromobility, ridehailing) vs inbound trips where things come to the user, usually at home (food and grocery delivery along with other kinds of logistics). This is particularly relevant to trips in the retail category, such as Kari ordering headbands on Etsy or Espen getting his medication delivered.

Stores have always served as a middle ground in this equation. Retail trips are on average shorter than any other trip category (see chart below). Merchandising is a process for bringing goods closer, even though these supply chain operations are largely invisible to customers. Consolidation in retail (larger, but fewer stores) and the rise of e-commerce have likely been the driver for retail trips becoming longer over time.
A line chart showing the change in average vehicle trip length over time, with trips on average and especially Work and Retail trips becoming progressively longer over time (between 1970 and 2010)

Source: 2017 National Household Travel Survey

There are many reasons you'd still want to go to a store (try something on, get advice, in-person services such as a hairdresser, etc) but one of the most important consequences of covid has been to dramatically shift the decision calculus away from physical retail stores and towards things coming to us through the trip economy.

Trip Price (how much?)

The final dimension of the demand side equation is the willingness of a consumer to pay for a particular trip. Consumers will be willing to pay relatively more for solutions that fit a larger set of their demand requirements and relatively less when they have to compromise. Meanwhile trip providers have to offer solutions that meet customer needs and must do so within a cost band acceptable to customers.

Supply side challenges are critical in mobility since, in contrast to technology where scaling costs are low, providing a mobility solution involves significant marginal costs and must be functional in real world conditions.

Status quo and convenience bias (why change?)

Alongside all of other considerations, including price, consumers avoid forming new habits and prefer things that are easy over things that are hard or complicated. Tied in with this is the fear of unreliability in a new service and the risk of getting stranded or something else going wrong. For instance, having a dedicated vehicle with space for storage is particularly appealing for parents like Kari since they don't have to share the vehicle with anyone else and can leave the booster seats in the vehicle along with anything else they might need. In addition, owned vehicles are already bought and paid for. Given these sunk costs, the idea of paying for trips feels expensive.

This bias towards convenience and the status quo is particularly important when considering new options rather than relying on more familiar solutions, although some consumers have a stronger affinity to trying new things.

Supply side factors

Meanwhile, suppliers are offering vehicles or trip services that are designed to meet specific customer needs at a certain price point. The factors that they optimize for map directly onto demand side requirements.

Utility properties

  • Range: How far can the vehicle reasonably travel?
  • Speed: How fast can the vehicle go? How fast can it move through traffic or other specific conditions?
  • Routing: What route do vehicles travel? Is there a way to combine several trips in order to increase the effective PMT (passenger miles traveled) or GMT (goods miles traveled) for a fixed amount of VMT (vehicle miles traveled)?
  • Capacity: What is the capacity of the vehicle for carrying people? How much cargo space does it have to store goods?
  • Condition: Does the vehicle have weatherproofing and air conditioning? Is it comfortable and spacious, or cramped? Does it drive smoothly? How safe is it if there is a collision? Is the customer driving the vehicle or is someone else driving?

Cost inputs

‍Besides the utility properties that allow a vehicle to meet consumer needs, there are various factors that are required to make trips viable but add to the cost of each trip:
  • Labor: Cost of maintaining the vehicle fleet and driving the vehicle. When users aren't doing this work, labor could be contractors / on-demand workers or full-time employees.
  • Energy: What it costs to refuel or recharge the vehicle
  • Cost of assets, depreciation and insurance: Mobility involves assets that need to be financed, insured and periodically replaced
  • Vehicle storage costs: Cost of parking the vehicles and other real estate assets required to maintain a service

Price bridges supply and demand

Consumers are more willing to pay for services that meet their mobility needs, but suppliers are limited by various costs they have to cover in order to offer a service. Price acts as a bridge between the two sides of the marketplace detailed above, mediating options depending on how effective or efficient each one is.

The trip decision matrix shows how consumers make trip purchasing decisions: the requirements for a trip (distance, urgency, regularity, cargo and comfort) less the opportunity and inconvenience costs of making the trip combine to determine what consumer needs and how much they will pay for it; the consumer weighs these requirements against the available options based on how their corresponding utility properties stack up against the trip requirements and the cost of the trip

On the demand side, consumer willingness to pay for a trip is determined by their particular needs. Detracting from this is the inconvenience incurred in the process of taking the trip. Examples of inconvenience are having to drive, finding and paying for parking, waiting for a ride or a delivery to arrive or having to exert effort while riding a bicycle or walking (how these are weighted vary from person to person). Overcoming the status quo bias in favor of car ownership (including that most of the cost of cars are paid up front) is also an aspect of this inconvenience cost.

On the supply side, trip price represents the cost of offering the trip with the operator's profit margin added in. So long as the trip price does not exceed the perceived value of the trip, consumers will be willing to use that service.

More concretely: Ridehailing has a high overall trip utility score (since it's fast, comfortable and reliable) and since you don't have to exert yourself much to use the service the inconvenience cost is low. However, the price of the trip is quite high, primarily driven by the cost of the driver, making it too expensive for many users. However, when you're late for an important meeting, you'll be willing to pay a fair bit more if it's the quickest way to get there.

In contrast, micromobility services might be cheaper and might also be valuable to users because they are fast and fun, but the inconvenience of trying a new type of vehicle, the uncertainty of finding a scooter and the perceived lower safety, especially when there isn't good micromobility infrastructure, will detract from this and mean that many users will choose not to use this service.

Consumers try to minimize the amount they pay even when services fit their needs. Therefore competition on price amongst trip providers pushes them towards greater system level efficiency while also seeking ways to improve the overall utility of trips and minimize customer inconvenience. Reframed in this way, it should be clearer how the trip economy market dynamics operate to improve mobility outcomes across many dimensions.

Environmental factors

Since the trip economy is a marketplace for different solutions, it is affected by several environmental factors that determine the quality of the choices available to consumers.


The most important determinant of the viability of a mobility marketplace is population density. If there are more potential customers to create trip demand, then there are more opportunities for solutions to meet this demand. Tied closely to the number of consumers is their ability or willingness to pay for services. If purchasing power is lower, even if there are many potential consumers, then the variety and quality of services will likely be lower.

Due to the overarching importance of density and ability to pay, the trip economy is primarily an urban phenomenon. Micromobility, food delivery and ridehailing service marketplaces all tend to be significantly stronger in large, dense, wealthy cities.

Climate, weather and topography

A drawing of Nissa walking in the rain with an umbrella

Baseline weather conditions as well as variability of the climate significantly affects what kinds of trips people are likely to take. Places where it snows more are less likely to see widespread micromobility adoption compared to places where the temperature is moderate year-round. Rain also usually puts a dampener on travel as does extreme heat.

In addition, topography and other geographic factors like altitude affect how people get around. For instance, in a city that is extremely hilly, it's less likely that people will be excited to use bicycles than in a city that is flat.

A way to think about the effect of these climate factors is through the lens of density of demand spread over the course of a calendar year.

Cultural factors

Even if there are many consumers with the ability to use new mobility services and these services can be offered at a reasonable price, there are a slew of cultural factors that tie into willingness to try new services and consistently use them over time. These range from openness to new experiences, fitness levels, perceptions of safety, and what takes on status and becomes fashionable within particular social groups. Political opinions about the environment and labor also tie into this.

Supply side cost factors

Markets vary not just in the density of consumers and their willingness to pay, but also in the baseline costs of offering mobility services. A primary input to offering any kind of mobility service is the cost of labor. This is particularly important when labor is required to operate vehicles, but even in cases where a fleet is available to consumers to operate, that fleet still needs to be serviced, recharged and repositioned.

Markets might also have other input costs that affect mobility services such as tariffs on importing hardware. This is particularly relevant in the context of the US-China trade war and European anti-dumping restrictions.

Infrastructure and the cost of real estate also factors into how easy it is to offer a service in a given area. Access to a warehouse with the ability to recharge and repair vehicles is essential to operating a micromobility service. In some places these kinds of facilities are easier to find and cheaper to rent than in others.


Much of the persistent appeal of car ownership comes not only from the fact that cars on average solve most trip needs (good range, speed, comfort and storage capacity) but also that our urban environment is largely built around cars, not least suburbs and all they contain.

Let's consider mobility through the lens of public interest including infrastructure and other externalities.


Infrastructure is the context for mobility, the place where it happens. Governments have a strong incentive to build out infrastructure since mobility is the lifeblood of other economic activity, allowing people to meet and be productive. To this extent, it makes sense to subsidize infrastructure to some extent.

What is built determines what makes the most sense to use. Regulators have to decide how to allocate finite urban space, which involves tradeoffs in land allocation between transportation modes and against alternate uses for real estate.

A visual representation of the four types of infrastructure networks: Vehicle space (purple), pedestrian space (green), transit space (blue) and micromobility space (yellow)

There are four primary types of transportation network infrastructure:

  • Vehicle space: Space for vehicles such as cars, trucks and buses, including roads, highways, on-street and off-street parking spaces; in most places, infrastructure spending is skewed heavily in favor of this category
  • Transit space: Space for public transportation including railway for trains, subways and light rail as well as bus stops and dedicated right of way for buses or taxis
  • Pedestrian space: Mainly sidewalks, crosswalks and bridges as well as spaces such as parks, shopping areas and buildings
  • Micromobility space: Bike and scooter lanes, sometimes protected, sometimes shared with other vehicles as well as parking space dedicated to these types of vehicles

Infrastructure not only needs to be constructed, but must also to be repaired, cleaned, policed and otherwise maintained. The weight of vehicles and how infrastructure is used affects these maintenance costs.

Other externalities

Congestion & throughput

The challenge with mobility infrastructure is that it gets congested.

Let's start with throughput. Throughput is the number of people able to move through a particular road or other transportation corridor within a given time. As trip demand increases on a road, the number of vehicles traveling along it increases too and with it, so does throughput. That is, up until a critical point when everything slows down dramatically. This cascade happens because traffic flow is limited by the weakest point in the network and one vehicle slowing down affects all those behind it when they are packed closely together.

Congestion is bad because it dramatically reduces throughput at just the moment when the need is greatest. Much like an energy grid which cuts out during periods of peak demand, congestion stops offices and factories from being productive twice a day in most major cities. And much like with optimizing energy production, the challenge is to find ways to flatten the peak periods of demand. One way to do this is to price road utilization. This is tricky to do for technological and political reasons but is effective in reducing demand and recovering costs for infrastructure.

To some extent congestion is self-regulating: people will avoid making trips at times when congestion is worst. But many trips are unavoidable, in spite of the aggravation of congestion: getting to work in the morning, picking kids up from school, getting to a scheduled appointment.

A line graph showing the distribution of vehicle trips by trip purpose and start time split out according to trip type; retail trips tend to be in the middle of the day and the other types of trips tend to happen in the morning and late afternoon

Source: 2017 National Household Travel Survey

It is tempting to think that the ultimate goal is to reduce the overall number of trips. But trips are life. Like energy powering factories and homes, if people can't make trips when they seek to, something valuable is lost. The goal is not to reduce congestion by limiting demand but rather to improve the ability of the supply side of the marketplace to meet demand by maximizing the total number of completed trips in a given time period.

A chart showing passenger capacity of various transport modes based on maximum throughput ranges standardized against 3.5m wide lanes: transit modes have the highest max throughput starting with subway (60,000-90,000 passengers per hour) followed by heavy rail (40,000-60,000), bus rapid transit (BRT) (11,000-38,000), light rail (18,000
20,000); pedestrian space allows 10,000
15,000, cycling similarly 8,000
12,500; roads in contrast only allow 1,500
2,400 passengers/hour and have the lowest throughput

Source: Environmentally Sustainable Transport - Manfred Breithaupt; supplementary sources: road capacity, cycling, pedestrians, BRT, rail. Estimates vary, NACTO has somewhat different but directionally similar estimates. Maximum throughput is important because network capacity is constrained by arterial chokepoints; average throughput is significantly lower than maximum throughput.

Therefore what we should focus on is roadway throughput: the number of people able to get where they want to through a transportation corridor at a given time. People getting where they want to means they can do more of the things they need to do. As density increases, the relatively low throughput of roads becomes a greater challenge. This is why transit, which has significantly higher throughput than roads, is a fundamental part of the transportation system in most major cities. But smaller vehicles as well as vehicles carrying more people are also effective for increasing throughput. Cars, meanwhile, work best at lower density.

Safety & wellbeing

Drawing of Kari cycling with the boys

Moving requires velocity making safety a key consideration. Reducing dangerous behaviors is a first step: preventing speeding, bad driving, not stopping at intersections, intoxicated driving and other violations of road rules. In addition, encouraging safety measures such as airbags or helmet use is important. These steps are important for vehicles interacting with one another in vehicle space, but it's even more important when pedestrian space and micromobility lanes intersect with road space.

Beyond just preventing bad things from happening within public thoroughfare, there are positive benefits that come from constructing public space in a way that encourages healthier living. Green spaces lead people to be healthier while exercise is one of the key determinants of wellbeing, which are factors that influence how one might choose to allocate public space across transportation modes.


A drawing of Espen with one of his granddaughters coughing due to air pollution

Most of the energy that vehicles use produces pollution, both in the form of particulates that damage human lungs and greenhouse gases that affect the climate. Both have been an increasing focus of regulation which forces vehicle manufacturers to improve fuel efficiency and reduce the quantity of pollutants they emit. Besides these regulatory "sticks", there has also been a wave of subsidies applied to low- or zero-emission vehicles. Most of these standards are focused on passenger and commercial vehicles since they are the primary source of emissions. Noise can also be a kind of pollution, and vehicles produce a fair amount of it. Micromobility rollouts have for the most part not been considered through this lens even though these vehicles have significantly lower energy needs, they're much quieter and they have significant potential to substitute trips.

Equity & opportunity

Not having access to a car is a significant barrier to success, particularly in the US. Car ownership, like a good credit rating, is a prerequisite to cost savings from stores such as Walmart and IKEA or access to better employment, childcare and education as well as healthcare (even drive-thru covid testing has required a car!). Transit fails to fill this gap since it is largely positioned as a mobility solution of last resort across America's relatively low-density cities. The correlation between vehicle ownership and economic access is likely a significant factor explaining why the US now lags other developed nations in terms of social mobility.

Measuring and paying for externalities

How can these externalities be factored into the decisions consumers make? The current tools policymakers have for paying for infrastructure and disincentivizing certain behavior are quite clumsy. In many cases, governments are subsidizing exactly the wrong thing. Regulation usually has unintended consequences and the broad inefficiencies of car ownership as they currently stand are strong evidence for this. These are the tools that governments currently have for addressing externalities.

Transfer externalities into the price of something else

This is a horrible way to align incentives but sometimes an effective way to make things happen.

Parking minimums are an example of these dynamics in action: as part of the cost of any new piece of real estate, the cost of parking is included. That includes houses, schools, churches, hospitals and a long list of other places you might be interested in going to - which all have been built with abundant parking as a result. Besides the questionable value of all this parking, for those who do not own cars there is an unfortunate misalignment since in paying for groceries, rent, school fees, and the like, you are subsidizing parking that you may not use. People who expect parking everywhere don't understand what the true costs are and how they are being paid. Given this sense of entitlement and the challenges of unbundling costs which are so deeply tied together, the effects of such policies are very hard to reverse.

Regulate externalities into the price of the good itself

Another way to deal with the costs of externalities is to include them into the good that is being regulated. Vehicles have to meet very specific homologation standards and the cost of developing and testing vehicles according to these standards is carried by consumers in the cost of the vehicle. One thing that is increasingly regulated into the price of a vehicle is its emissions through various global fuel efficiency standards (e.g. European Regulation (EC) 443/2009 and US CAFE standards - see chart below). Another example is safety technology rated according to the Euro NCAP safety standard.

Line chart showing how governments around the world (China, Japan, Europe, US) are setting increasingly stringent targets for vehicle CO2 emissions

Source: ICCT

Governments can also subsidize the cost of certain vehicles either directly or in the form of tax incentives. There are better and worse ways to do this. Norway taxes vehicles progressively based on a combination of weight, CO2 and NOx emissions and uses revenues to subsidize electric vehicles, which has driven EV sales to over 50% of new vehicles. In contrast, a few US states including California use a system call ZEV credits, which disproportionately rewards carmakers that produce EVs with greater ranges (a Tesla Mode Y earns about 3x the ZEV credits of a Nissan Leaf) and arguably mostly helps rich people buy luxury vehicles.

Taxes & tax incentives

Governments at various levels use tax money to pay for infrastructure or create other incentives (such as the ZEV credits just mentioned). Regulators have some accountability to voters when spending public funds, but this spending is hard to track and the relationship between the payment and the benefits to citizens is loose, making for a weak feedback loop. The gas tax tries to more directly tie infrastructure spending to revenue generated by utilization, but the improving fuel efficiency of vehicles has decoupled these revenues from the true costs of road wear and tear. This will only get worse with heavy electric vehicles.


Fines in the form of speeding or parking tickets are another way to factor in externalities on a per incident basis (these costs can extend beyond financial penalties to include driving restrictions). Mostly fines work to regulate behavior through fear: when you speed, there's a good chance you won't get caught, but the cost of the punishment is high to create an appropriate disincentive. This approach works but is fairly patchy.

Tolls & surcharges
A drawing of Bo driving a car through a toll booth

Tolls and congestion fees (like those in London or Singapore) are direct ways of charging users for utilization or externalities like congestion since they are charged on a per trip basis and may even be specific to route and time of day. However, these measures are hard to implement since consumers have a baseline assumption that road utilization should be free. Practically, implementing these systems requires expensive infrastructure such as tollbooths that need to be manned or gantries that need to be installed in large numbers. Furthermore, charging for road use can create delays, fees might not be clear and users may be charged unexpected fines causing frustration. On the whole, directly linking payments to road utilization has been challenging and implementation has been limited.

Trip pricing and externalities

Current tools for pricing in transportation externalities are clumsy and hard to adapt to a changing reality. In some cases, such as massive subsidies to parking and road infrastructure, they lack visibility and are not clearly aligned with desired outcomes. In other cases, such as tolls and surcharges, they are difficult and expensive to implement effectively.

Much of the challenge stems from the decathlete bundle of car ownership. Since it involves a very large upfront purchase with very low usage costs, pricing in externalities creates significant friction. In contrast, the trip economy offers a much simpler way forward.

Characteristics of trip pricing

Trip pricing means that externalities can easily be factored into a trip price and consumers are more likely to accept these charges since they are blended into the complete trip price. Consequently, trip pricing is more effective in shifting consumer behavior since each trip price includes all price factors for consumers to make informed spending decisions. In addition, trip pricing offers a feedback loop since it is easier to collect data and measure the effects of externality pricing.

The importance of data is worth emphasizing. Currently, we have relatively little information about road behavior and few tools for shaping it. However, since mobility apps exist on smartphones and smartphones are really good at collecting data, everything that is hard to measure in car trips is automatically measured for mobility trips: distance, speed, start and end point, vehicle type, etc which then feeds into the trip price. How to manage this data is a complicated question, but its value is clear.

Demand side factoring

Trip pricing allows very granular control of various inputs into the trip cost to shift consumers behavior. Let's take congestion pricing as an example. As mentioned, it is hard to institute congestion pricing in the current framework of car ownership since it requires cumbersome infrastructure and creates friction. But considered within the trip pricing framework of ridehailing, it is quite easy to factor a congestion charge into any trip according to a range of parameters. Specific charges could be added to specific routes at specific times to shift trips away from the most congested parts of the road network. Or at periods of peak demand, single person trips (UberX) could incur a higher fee while shared trips (UberPool) could receive a subsidy. Similarly food delivery orders that are not batched could be charged extra at peak hours. Targeted price incentives could even be used to encourage users to shift a ridehailing trip to a micromobility trip. Overall, these nudges would allow the overall number of vehicles on the road at peak times to be optimized and throughput to be increased.

Considerations such as equity can also be factored into the price consumers pay for trips. Just as governments subsidize the cost of transit, they can similarly subsidize the cost of trips for communities or individuals where they see a positive benefit in doing so.

Overall, trip pricing gives a very precise way to shift consumer demand using price signals and incentives while measuring the outcomes through data to improve outcomes over time.

Supply side factoring

Price signals can be used not only to shift demand in trip marketplaces, but also to influence the behavior of trip suppliers. Regulators can also influence any of the supply side inputs through price signals.

Rather than dictating the best way to do something, regulators can set a target and allow the market to determine the best way to optimize outcomes. For instance, regulators could set limits on the size, weight and parking footprint of vehicles. They can also charge fleet operators according to how much time their vehicles spend parked and fine them for parking violations by their users. In micromobility, this already happens and suppliers such as Bird encourage their users to report these kinds of violations through its Community Mode. Regulators can also manage fleet size not only through tenders (which have become popular) but also by setting a price to deploy additional scooters that add costs if vehicles are underutilized.

On the whole, for almost any mobility externality, there are multiple regulatory levers to pull once mobility services have fleet operators that are responsible to regulators and that charge users a fee on a trip basis. The challenge is figuring out which are the right ones while minimizing unintended consequences and not getting drunk on this new power.

Ensuring a level playing field

Because trip pricing is so responsive to every priced-in externality, it creates a significant challenge when car ownership is a competing choice given how relatively hard it is to do the same for car trips (and this is not considering significant existing subsidies to car ownership). As a result, it is hard to create a level playing field, especially since regulators are vulnerable to a status quo bias that makes them forget about car externalities but want to aggressively regulate new modes such as micromobility.

Take Portland for example. The city decided to levy a 25¢ charge on each scooter trip (close to 10% of the average trip cost), but added nothing to the cost of using a car. Portland has further complicated matters by setting an overall cap of 2,500 scooters and requiring at least 20% of the fleet be deployed in lower-income East Portland.

Currently there is no city in the US that directly subsidizes micromobility. Given that the negative externalities of cars are far greater than those of scooters (more pollution, more road wear & tear, congestion, more dangerous to other road users, less accessible to lower income residents, etc) cities should be seeking to shift the equation in the exact opposite direction to the one in which regulatory allowances and money currently flow.


Uber's high-profile fight with the city of LA over trip data collected through the Mobility Data Specification (MDS) highlights that privacy is in tension with trip pricing. The upshot of granular data collection is that it makes it significantly easier to factor in externalities. The challenge is to do it in a way that still protects consumer privacy, although a stable state should emerge that balances these concerns as it has in other industries.

New tools

Before the trip economy, regulators have had relatively few tools for optimizing transportation outcomes. Mobility modes have been effectively binary: public and private. The trip economy collapses the gap between these two, offering the efficiency and convenience of a private market with the ability to easily factor public benefit considerations. Trip pricing therefore is a gift to regulators that will continue to give as priced trip volume grows, especially in dense urban environments.

So long as costs are transparent and clearly communicated so that people understand what they're paying for and why, and charges and incentives are aligned with the right goals, trip price signals offer a way to dramatically reduce the inefficiencies that are so widespread in our current transportation system.

The trip triangle

In summary, the following overarching consequences result from a shift towards a trip economy:

  1. Demand: In cases where there is sufficient density, demand expands from a default of car ownership to an array of trip types fitting more precisely to specific trip requirements
  2. Supply: Competition to meet demand amongst various trip suppliers means that underlying costs are streamlined in order to reduce the trip cost or otherwise improve other aspects of the trip to meet demand requirements. Over time, the overall range of trip supply expands to meet a broader range of needs

  3. Externalities: Since price signals actively and continually shape both the supply and the demand side of the mobility equation, pricing becomes an effective tool for regulators to factor in externalities, ideally to improve societal outcomes

This results in what I call the trip triangle, a framework for considering how price signals, including input from externalities, are balancing supply and demand much like a seesaw:

A visual representation of the trip triangle, where supply and demand dynamically balance one another out like a seesaw, but externalities can relatively easily be priced in by regulators to tilt the seesaw one way or the other

The trip triangle: supply and demand are balanced through price signals and regulators can relatively easily price in externalities by factoring them into trip prices

It's worth highlighting why this is transformative in contrast to the prevailing paradigm: carmakers have never had to consider the systematic utilization of the vehicles they produce. They simply sell vehicles and so long as they meet regulatory standards and only customer interactions are when the consumer services their vehicle or returns to purchase their next vehicle. In contrast, the DNA of mobility disruptors that sell trips is fundamentally different, with efficiency considerations baked into their business model along with direct accountability to local regulators. (This is also why the current changes present such a challenging cultural shift for incumbents since their business processes are structured in a fundamentally different way).

This shift therefore not only helps address the rampant inefficiencies of our prevailing transportation system, but it also increases consumer choice and gives more effective tools to regulators for improving societal outcomes.

This potential is most dramatic in well managed dense cities which will have the greatest selection of mobility options and the least systemic inefficiency. But the implications are broad and will grow as technology improves and the range of trip offerings expands to encompass more trips.

We consider this technological improvement next.

Share this chapter with others:

Part 2: Technological building blocks