Will the coming mobility revolution make urban traffic congestion better or worse? McKinsey investigates. 

The age of modern transit began in 1863 when the first underground train began rolling in central London. The line was short and smoky, and nothing like it had ever been seen before. But it worked, and cities around the world began to follow London’s lead. Over time, city authorities came to see providing transportation as one of their core responsibilities; governments often owned and ran transit systems themselves.

Despite their best efforts, however, traffic is getting worse in many cities, and urban mobility has become increasingly complex. Between 2010 and 2016, congestion in London rose by 14%, in Los Angeles by 36%, in New York by 30%, and in Beijing and Paris by 9%.

Congestion carries health consequences, in the form of accidents and air pollution. Demographic trends — more people living in urban areas—will exacerbate the strain, which today isn’t solely about the movement of people. E-commerce is also growing fast, adding to the demand for urban commercial transport.

The technological changes associated with mobility’s “second great inflection point” create myriad opportunities for cities to address these challenges. As ride-sharing platforms grow, digital vehicle connectivity deepens, electric vehicles (EVs) become mainstream, and autonomous vehicles (AVs) take hold (see “The trends transforming mobility’s future,” forthcoming on McKinsey.com), it’s possible to envision a future of “seamless mobility.”

In such an environment, the boundaries among private, shared, and public transport would be blurred, and travelers would have a variety of clean, cheap, and flexible ways to get from point A to point B. McKinsey’s analysis suggests that seamless mobility could be cleaner, more convenient, and more efficient than the status quo, accommodating up to 30% more traffic while cutting travel time by 10%.

For now, however, today’s reality is far from seamless. Vehicles that are fully autonomous do not yet exist in meaningful numbers, EVs still make up only a small percentage of the global vehicle fleet, and traditional internal-combustion-engine (ICE) cars represent about 40% of passenger-kilometers—often more than rail and bus services combined.

Starting from this baseline, urban-mobility systems in dense, developed cities, such as London, New York, and Seoul, could evolve in a variety of ways over the next dozen years.

Scenario 1: Business-as-usual urbanization

Imagine a world in which population growth continued, but large cities managed their transport systems largely as they do today, with little innovation in pricing or policy.

Transport demand would increase in line with population growth (about 15% by 2030), and greenhouse gas (GHG) emissions would rise proportionally. City dwellers would travel in mostly the same ways as they do now, and private cars would continue to account for about 35% of passenger-kilometers. Average travel times would increase by 15% because capacity would be strained. In the United States, without substantial change, this is the likely scenario: few cities, even the largest, have comprehensive plans that take into consideration new and forthcoming technologies.

Scenario 2: Unconstrained autonomy

What if autonomous transportation options follow in the footsteps of bike-sharing programs and e-hailing, which hit the road faster than the regulatory policies needed to guide them did?

It’s certainly plausible. By 2030, shared AVs—or robo-taxis—could navigate to, from, and within a central business district. McKinsey’s analysis suggests that, by that point, they could become an attractive alternative to private vehicle ownership, with the cost per mile of riding in a robo-taxi running about the same as the cost of owning a moderately priced private vehicle, and that travelers could adopt them for individual or shared use for about 35% of their travel by 2030.

Scenario 3: Seamless mobility

Now for a third possibility: What if cities encourage the use of shared AVs through regulation and incentives? That would make it possible for residents to “mix and match” rail transit and low-cost, point-to-point autonomous travel in robo-taxis, autonomous shuttles, and autonomous buses easily.

McKinsey’s analysis suggests that pooled AV shuttles could grab 25% of the market (twice as much as in the unconstrained autonomy scenario), and private cars and privately used robo-taxis could provide about 30% of passenger-kilometers in 2030, compared with 35% for private cars today.