Fixing congestion on the M4 around Newport requires cutting around a quarter of peak traffic, but no road-based solution currently costed hits that target. The Welsh Government has launched an 18-month study to assess options, with engineering consultancy Arup’s latest review showing costs from £14m to £1.8bn.
Study benchmarks congestion cuts and costs
Arup’s review, completed between June and August 2026, priced various measures to ease traffic through the Brynglas Tunnels, where the motorway runs at 114% of its theoretical capacity by 2042. The consultancy says a 25% reduction in peak traffic would be needed to avoid recurring congestion, but the best option evaluated delivers only up to 17%.
The report does not feature new traffic modelling and treats its figures as indicative. It also notes that including an option does not guarantee feasibility.
£14m junction closures cut 3% of traffic
The cheapest measure is fully closing junction 27 and the east-facing slips at junctions 25 and 26. Arup says this would cut M4 traffic by about 3% and could be delivered in two years. It would reduce local vehicles merging near the tunnels, but would shift traffic onto nearby roads such as Malpas, High Cross and Ridgeway.
Arup recommends trialling these closures under an experimental traffic regulation order first. Officials warn faster journeys may attract extra road users, which could reduce the net benefit.
Public transport and road improvements add up
Other options include increasing rail services on the South Wales Main Line for £30m, which would cut 1% of M4 traffic over five years. Introducing signs and HGV rest facilities to reroute freight to the A465 costs around £49m, cutting 1% of traffic.
Combining junction closures with A465 freight measures forms Arup’s fastest and cheapest package, costing £63m and cutting up to 4% of traffic in under three years.
Adding a Newport–Cardiff bus rapid transit line at £335m could further reduce traffic by 2%. Building six new stations between Cardiff and Severn Tunnel Junction costs £800m and cuts 6% of traffic over six years.
Applying all three public transport options together with the £63m highway package totals around £1.23bn and cuts up to 11% of M4 traffic.
Major scheme costs hit £1.8bn with modest cuts
Widening the Brynglas Tunnels is one of the riskiest but most capacity-adding projects at an estimated £1.7bn. It involves converting the hard shoulder into an extra lane, building a new westbound three-lane tunnel, and keeping the existing tunnels for eastbound traffic. The speed limit would remain 50mph, and major disruption could last about seven years.
Targeted improvements to the A48 Southern Distributor Road would cost £1.7bn and reduce M4 traffic by around 5% over eight years. Dualing both the Southern Distributor Road and the A4810 Steelworks Access Road would cost £1.8bn and cut traffic by 6% in eight years.
Best combination falls short of target
Arup’s best combined package, priced at around £3bn, pairs road upgrades on the A48 and A4810 with junction closures, A465 measures and all public transport options. It could achieve around 17% traffic reduction after eight years, still below the 25% target needed to remove congestion.
Excluded options and political context
The study excludes a new motorway and a Magor–Castleton tunnel due to affordability, environmental and planning challenges. It also omits tolls and road user charging since the Welsh Government is not pursuing these currently.
Political views diverge. First Minister Rhun ap Iorwerth has ruled out the “black route” relief road but supports road-based solutions. Newport Council’s Labour leader criticised junction closures, while Conservatives and Reform call for building the relief road.
Supporting measures underway
A five-year programme will run alongside the study, focusing on motorway renewal and maintenance, improved traffic management, local targeted upgrades, better freight use of the A465 and support for Cardiff Parkway and Magor and Undy stations.
Deputy Minister for Transport Mark Hooper said the intention is to cut peak-hour traffic by around 10%. No cost has been given for the 18-month study itself.
