Do New Roads Reduce Traffic Congestion? (Spoiler: No)
If you’ve spent much time watching the news in Australia, you’ll have a pretty decent idea of how roads are normally built. It typically goes:
Research says city traffic is up X%…
Government slips X points in new poll…
Minister announces $X Billion for new roads…
It happens every other year, so our cities become a never-ending sea of new road construction. So why is there still traffic? For that matter, do new roads reduce traffic congestion?
The short answer is no – new roads don’t reduce traffic congestion. That’s primarily thanks to Induced Demand.
Australian traffic congestion can’t be solved by simple building more roads (Image: Michael Coghlan)
In case you’re in any doubt over that claim, here are some examples of cities that learned it the hard way:
A 2014 study listed Los Angeles, San Francisco, and Honolulu as the most congested cities in the US, each with exclusively road-based transport. With no suburban trains, new roads are the only option these cities have to reduce congestion… but it hasn’t helped.
Between 2000 and 2013, China’s expressway network went from 16,300km to 70,000km. Despite that, Beijing’s 2013 average commute was 1 hour 55 minutes, 25 minutes more than a year prior. Beijing’s traffic was estimated to cost the city around $11.3 Billion annually.
Despite a huge road network, Los Angeles is still heavily congested – new roads don’t reduce traffic.
The reason why building more roads doesn’t reduce traffic congestion is due to the phenomenon of Induced Demand. It’s driven by a simple principal: people don’t want to be stuck in traffic.
So, when congestion is high people are more likely to use driving alternatives like public or active transport. When new roads are opened it increases the supply of road space, and traffic decreases temporarily. Because traffic is now slightly better, people get back in their cars, resulting in congestion returning to normal.
Induced demand works just as well in Australia as anywhere else. Take our most famous roadway, the Sydney Harbour Bridge. Between 1986 and 1991, 180,000 cars crossed the bridge each day. The harbour tunnel opened in 1992, and by 1995 the almost 250,000 cars crossed the harbour daily. That’s an increase of 38%, when population only increased by 4%.
More road capacity increases demand for itself, until there’s just as little supply as before, if not less. If the aim of government is to truly reduce traffic congestion, there are better ways to do it than building more roads.
The Sydney Harbour Tunnel didn’t reduce traffic, it cause more congestion thanks to Induced Demand in Australia
Unfortunately, there’s no magic bullet to reduce congestion across Australia. The reasons for congestion vary wildly, from a lack of public transport, development problems, peak demand periods, and even geography. That being said, there are some approaches that generally reduce congestion:
Increasing public transport capacity typically increases patronage, taking cars off the road. The priority should be on routes currently at (or over) capacity.
Reducing private car ownership is a reliable way to reduce the amount people drive. For example, each GoGet car takes 10 privately owned cars off the street. Support for services that help people live without a car will help people drive less.
Increasing density near new or underused transit hubs is a long term approach, but a very effective one. If a particular train or bus line is underused, increasing housing supply nearby will result in a smaller proportion of those residents using a private car.
Congestion pricing has had a long history of success in reducing traffic, especially in cities like London and Singapore. Even New York City will introduce congestion pricing in 2021.
These are just a few examples of what can be done across the public and private sectors to reduce congestion, but there are many more levers to be pulled, and each city will require a different mix of these to successfully reduce traffic.
Public transport and GoGet car share are proven ways to reduce traffic congestion in Australia
So should we keep building new roads? Well, yes and no. Transport by car and truck is now an unavoidable part of our modern transport mix, so we need to continue to have good roads that go to places people need to get to. There’s also still a major role for road-based public transport, so on a very basic level, roads will still need to exist for some time.
There are also situations where new connecting roads may have a positive effect on traffic in other areas. Sydney’s NorthConnex aims to give trucks a route between Sydney and Newcastle that doesn’t use local roads. However, it may induce demand to the point that local roads see no benefit after it opens in 2020.
As far as the roads we already have, there may be better ways to use them. The attitude of most modern planners is street space should be designed for ‘people first’, cars second. That means a variety of things, including:
Reserving lanes for buses, which transport more people than cars,
Pedestrianisation, which turns lanes into footpaths to increase walkability,
Turning car lanes into bike lanes, ideally separated from other traffic,
Reducing traffic capacity to build more green space and community infrastructure.
A combination of these approaches, where appropriate, can actually increase the transportation capacity of a street, while reducing the space needed for cars.
Other benefits can come from street optimisation too. For example, increasing walkability and building bike lanes is strongly linked to increased retail footfall. In New York City, a series of cycling and pedestrian projects led to a 49% increase in nearby retail trade!
So where does this leave us? Well, when someone tells you new roads will improve traffic, put your skeptic hat on!
More broadly, we need to accept that good transport planning is hard to sum up in a ten second news grab. There are lots of little things that will inch us closer to a better functioning city, but Transport planning is a marathon, not a sprint, and we’ll only win it inch by inch.
