The Transport Committee of the United Kingdom parliament is holding an Urban Congestion Inquiry. This article forms the basis of Push Bikes' submission to that inquiry.

Build for cars, and you will get more cars using the roads, which means more cars queuing on those roads¹. Meanwhile more and more of the finite space in cities is given over to roads, at great expense both in terms of construction and maintenance. Our roads are a potholed mess, because we cannot afford to maintain them, yet we go on building new roads in order to "ease congestion". Britain is like the gambler that believes the jackpot is just around the corner, if only he keeps gambling.

By way of example, the photo depicts the recently built A38 relief road for Selly Oak and Bournbrook high streets. It is already congested, whilst the original route (which was kept open) remains heavily trafficked. As can be seen, most of that traffic consists of private cars. The cars on the left are queuing to get into a residential road, one that the Birmingham City Council has made into a rat-run so that motorists can bypass the congestion on the main road. The residents are fed up with the heavy traffic and the speeding. Each of those cars is most likely to contain just one person. Leaving aside the bus, this sea of cars represents approximately just thirty people trying to get across the junction. So thirty people are delaying a bus which can carry around ninety people. And therein lies the clue; private cars are an incredibly inefficient means of transporting people, because a car is fundamentally much larger than the occupant, and in addition it requires a large amount of space around it when moving. It makes no difference whether the car is powered by petrol, diesel, or electricity. High-tech approaches such as driverless technology (which are not even mainstream yet) will make little difference to the equation, because they do not change the size of the vehicle, its speed whilst moving, or its braking traction. The automotive city concept (trying to move everyone by car) doesn't work, as is demonstrated by congestion.

Whilst cars can move just 4,400 people per hour on a given road, on the same road trams will move 66,000 people per hour, and buses 52,800. The problem is they do not represent door-to-door transport, so unless your journey lies on the route, and the route is direct, they do not represent an attractive solution for many journeys. Walking will move 17,600 people per hour, but it is too slow for most people. In the sweet-spot of speed vs spatial efficiency vs convenience is the humble bicycle.

Cycles will move 14,400 people per hour at a speed high enough that most city journeys can be completed in a reasonable time², and they allow door-to-door journeys. They do all this without creating any additional pollution. Many urban areas in mainland Europe now have extensive infrastructure that enables cycling that feels and is safe. Every urban area that has built such infrastructure has seen a massive modal shift towards cycling. The better the infrastructure, and the more extensive it is, the bigger the modal shift. Britain is way behind, but Waltham Forrest Council has shown that the Dutch approach to creating infrastructure for cycling works here too.

So we need a plan for catching up. Fortunately, 15-25% cycling modal share (which will have a considerable benefit given the threefold improvement in spatial efficiency) is easy to achieve. Push Bikes, the Birmingham Cycling Campaign, suggests the following approach. In any urban area:

1. Use an OS map to see where are the A and B roads. These will naturally form a mesh. Use bollards, planters, and one way systems to ensure you cannot drive directly across the holes in the mesh; you have to drive out to the nearest A or B road, and go around the outside. Cyclists and pedestrians can use any public road in any direction, and where safe, cyclists can also use paths. This is called filtered permeability.
2. Build connectors for cyclists and pedestrians between the holes in the mesh. Very often there will be obvious points of connection, such as an existing crossing (controlled or zebra), or a crossroads. Improve junctions and build segregated cycle infrastructure at these points first. Signpost the quickest routes through the mesh.
3. Build segregated infrastructure along main roads to link up the initial sections as required (this is needed for people travelling longer distances). Steps (1) and (2) unravel transport modes, and may result in a link section in Step (3) being relatively unnecessary.
4. Polish. Make roads within the mesh holes look less like roads. Add more segregated sections.
5. Extend filtered permeability out into the countryside.

Meanwhile:

• Every time something is going to be rebuilt, the design should include proper consideration for cyclists, even if right now it connects with nothing.
• Remodel junctions that cyclists will be using to tighten up the geometry (so motorists are forced to slow down).
• Implement chicanes to slow motor traffic, not pinch points and speed cushions (which cause all sorts of problems for everyone, and have no real effect). Just alternating the parking between sides along a road will create chicanes.
• We need two additional junction designs, one of which may be currently legal, one of which isn't. If legal, start implementing, if not it needs to be legalised. The first is the Dutch-style roundabout (which is single lane, has tight geometry, an adverse camber, and cycle crossings that allow the cyclist to cross one lane at a time). The second is the simultaneous green junction (which allows cyclists to travel in all directions across a junction whilst motor traffic is held by red lights). These junctions are described on the Push Bikes website³.
• The toucan crossing is a very poor compromise that is hopelessly slow and inefficient for all transport users. We need to copy the Dutch approach, where pedestrians and cyclists are provided with controlled crossings that reflect their very different speeds, controlled crossings that change as soon as the button is pressed⁴.

Additionally:

• Highways engineers need professional training, and there need to be national design standards. The simplest, fastest, most reliable method of gaining national design standards is to copy the Dutch national design standards, which are tried and trusted, and available in English⁵.
• There needs to be adequate funding. The Dutch spend around €25 per head per year, but we need to spend more to catch up. This would come from the existing highways budget, of which it would be a small percentage.

By these means one reduces private car use and creates space for the segregated cycling infrastructure. Reduced car usage means less congestion. Note that Germany and the Netherlands, for example, are both already at Step 5, which gives you an idea how far behind is the UK. The Netherlands represents global best practice, and they have achieved up to 60% cycling modal share. Denmark and Sweden are other good examples.

• ¹ Cairns et al (1998), Duranton & Turner (2011), SACTRA (1994).
• ² Birmingham Mobility Action Plan: https://www.birmingham.gov.uk/bmap.
• ³ Building Blocks: http://pushbikes.org.uk/infrastructure/guiding-principles/building-blocks
• ⁴ Dutch Crossings Change Quickly: https://vimeo.com/117470593
• ⁵ Design manual for bicycle traffic: https://www.crow.nl/publicaties/design-manual-for-bicycle-traffic