The Promise That Roads Can’t Keep
It sounds intuitive: roads are full, so build more roads. For decades, this logic has underpinned billions of pounds of UK transport investment. But a UK Department for Transport (DfT) review has confirmed what traffic researchers have long argued — building new roads doesn\’t fix congestion. It makes it worse.
The reason comes down to a well-documented phenomenon called induced demand: the idea that expanding road capacity doesn’t simply accommodate existing traffic, it actively generates new traffic that would not otherwise exist.
What Is Induced Demand?
Induced demand, in the context of transport, refers to the additional vehicle traffic created by increasing road capacity. Put simply: build more roads, and more people will drive on them. The effect is so consistent that researchers have begun calling it the “fundamental law of road congestion.”
When a new road or extra lane opens, several things happen simultaneously. Drivers who previously avoided a route due to congestion begin using it. People who relied on public transport, cycling, or walking switch to cars. And entirely new journeys — ones that were previously deemed not worth making — suddenly become viable.
As one transport analysis puts it, new road capacity attracts drivers who previously avoided it due to congestion, and also encourages those who used other modes of transport to opt for the car instead. The initial improvement in traffic flow is usually temporary — congestion eventually returns to, or exceeds, its original level.
What the DfT Review Found
WSP and RAND Europe were commissioned by the Department for Transport to undertake a literature review of the evidence on induced demand, informing the development of the Road Investment Strategy (RIS). The findings reinforced earlier landmark research: induced demand is real, significant in many circumstances, and cannot be ignored in road appraisal.
This echoes the findings of the government’s own SACTRA (Standing Advisory Committee on Trunk Road Assessment) report from 1994 — one of the most important studies in UK transport history — which established definitively that building new road capacity generates more motor vehicle trips. Despite being over 30 years old, that conclusion remains as relevant as ever.
The Climate Change Committee (CCC) has also warned that the current Road Investment Strategy “still provides considerable funding for new roads which will induce increased demand,” and has urged that induced demand and lifetime emissions impacts be adequately assessed before any new road schemes proceed.
A Cautionary Tale: The M25
Few examples illustrate induced demand more vividly than London’s M25. Despite being one of the largest ring roads in the world, it has suffered from continuous and worsening congestion — a direct consequence of induced demand. New users flooded in, some simply making short “junction-hopping” trips they wouldn’t have made otherwise, and the road’s capacity was swallowed almost as quickly as it was created.
The pattern is not unique to the UK. A UCL review of policy measures confirms that investments in road building or upgrade do not reduce congestion — a finding replicated across multiple countries and road types.
Why Do We Keep Building Roads?
If the evidence is so clear, why does road-building remain a default policy response? Part of the answer is political visibility. Ribbon-cutting on a new bypass is a tangible symbol of action. In contrast, improving bus services or investing in cycling infrastructure offers less dramatic optics.
There’s also the challenge of the “predict and provide” model — a planning approach that forecasts future traffic growth and then builds roads to match it. This model fails to account for the fact that building those roads is itself what drives the traffic growth being predicted.
What Actually Works? Alternative Approaches
A growing body of evidence points to a different set of solutions — ones that manage demand rather than expand supply:
- Investment in public transport: A study contrasting Norwegian cities found that in Oslo, where road-use charging revenues were invested in the public transport network, congestion fell. In Bergen, where revenues were spent on new roads, congestion increased.
- Congestion charging and road pricing: Directly pricing road use during peak hours reduces discretionary journeys and incentivises mode shift. London’s Congestion Charge remains one of the most studied examples globally.
- Cycling infrastructure and pedestrianisation: Protected cycle lanes in London move five times as many people as a standard carriageway while using just 30% of the road space. Expanding active travel infrastructure reduces the number of single-occupancy car journeys.
- Low Traffic Neighbourhoods (LTNs): Cities like Oxford are implementing LTNs alongside electric buses, e-bike hire schemes, and zero-emission zones to systematically reduce car dependency.
- Workplace Parking Levies (WPLs): Nottingham\’s WPL scheme has successfully incentivised businesses to promote public transport, cycling, and carpooling among employees, reducing peak-hour congestion.
- Demand management policies: Evidence confirms that measures to reduce road use are most effective when combined with improvements to public transport, providing genuine alternatives rather than simply restricting choice.
A Turning Point in Transport Policy?
Both the Scottish and Welsh devolved governments have already committed to stopping investment in road-building aimed at catering for unconstrained traffic growth — a significant policy shift that reflects the weight of evidence. England’s approach under the forthcoming Road Investment Strategy 3 (RIS3) will be closely watched.
The DfT review is a timely reminder that the old playbook — build, fill, repeat — is not a congestion solution. It is a congestion generator. As transport planners and policymakers grapple with net zero targets, air quality obligations, and tightening public budgets, the case for smarter, demand-led transport investment has never been stronger.
More roads mean more cars. The evidence is in. The question now is whether policy will follow.
