A pragmatic look at cycling in Amsterdam by a traffic engineer | Oli Davey

Along with some like-minded friends, the Urban Movement Team recently returned from a 3-day trip to the Netherlands, spending a couple of days in Amsterdam and a further day in Utrecht. 

In total we cycled approximately 50 miles. We covered everything from a weekday rush-hour to a Sunday afternoon jaunt; the city core to the sleepy suburbs; and 1960s tower blocks to modern canal-side developments. There’s always more to see and learn, of course, but I felt that we covered quite a bit in the time we had (particularly with the assistance of some well-informed, local ‘guides’), providing us with a pretty rounded experience.

This trip followed hot on the heels of a similar visit to Copenhagen by the team a couple of months prior, making for some interesting comparisons. So what did I learn?

The Dutch have obviously been doing something right – their cycling mode share is fantastic. I’m always a bit dubious about exact numbers (it’s often very hard to ascertain exactly what area and times the quoted figures cover, and to ensure that they are comparable with numbers for other locations), but I’ve heard numbers somewhere around the 40% mark used in the past and this certainly feels credible.

What I was most interested to get out of this trip was a better understanding not just about what the Dutch have been doing to get people on their bikes, but (taking inspiration from Simon Sinek’s book ‘Start With Why’) also try to get to the bottom of how they have done it, and why. For me, this understanding is critical, because I don’t believe that we can simply copy and paste pieces of Dutch infrastructure into contemporary UK streets. We need, rather, to apply Dutch principles to our own context in order to achieve similar outcomes (hopefully I’ll elaborate on this a bit later).

Sometimes it’s disheartening to look at the most modern elements of Dutch cycling infrastructure. It all too often illustrates just how far we need to go in this country, politically, culturally and socially, to create places that make cycling a comparatively easy and attractive choice. This is why it was particularly helpful to be given a guided tour of a number of suburban developments in Utrecht whose construction ranged from the 1960s right up to the present day. The obvious evolution of their infrastructure provides the back story to their current position and provides some clues as to their philosophy.

What was most encouraging from a UK perspective were the similarities that we seem to share with the Dutch. The streets in both Amsterdam and Utrecht share a utilitarian, engineered approach. By comparison, based on our Copenhagen trip, the Danes seem very design-led, with a sense that many of their streets had a significant architectural element. Now, I’m not saying that we shouldn’t aspire to a higher design quality for our streets, but this clearly isn’t a prerequisite for making them more civilised and cycle friendly. There was a familiarity and pragmatism about Dutch street design that I feel the majority of UK engineers would recognise, even if the actual outcome looks different (a Dutch DMRB if you like). Why am I making such a big deal out of this? Well, it hints at an ability for our existing highway authority structures to deliver this type of infrastructure without the need for wholesale changes (a feeling that my trip to Copenhagen did not instil).

Something else that became apparent after three days of pedalling around was that Dutch drivers and cyclists like to make ‘progress’ almost as much as their UK counterparts (again something that I didn’t find in Copenhagen). It’s not aggressive, but there was a noticeable amount of red-light running, nosing out into traffic and general bolshie-ness that any UK road user would be familiar with. This was in stark contrast to Copenhagen where everyone seems to follow the rules to the letter. Again, although this may not be something that we want to encourage, it’s another sign that we may not be as far away from delivering better cycling infrastructure as we might think.

So, in at least a couple of ways, Amsterdam feels more applicable than Copenhagen to the UK. If their street design techniques and driver behaviour share some commonalities with the UK, then where are the main areas where the Dutch seem to be doing things differently?

Firstly, they have a very strong two-tier network. Much has been made about ‘going Dutch’ and the strict segregation between modes that the Dutch employ, but this only tells half the story and I’ve heard comparatively little about how this segregation works as part of a wider strategy. The Dutch employ a classic cellular model, whereby a grid of main streets takes all through traffic and where each mode is provided with its own, segregated space (this type of infrastructure appears to form the basis of the ‘Go Dutch’ campaign). However, once inside this grid vehicular traffic is generally still allowed but through routes are either cut off or circuitous, and appropriate street design is used to keep vehicle speeds low. Within these same areas cyclists share the carriageway with motor vehicles.

Interestingly, the description above could just as easily have come from the 1963 publications ‘Traffic in Towns’, also known as the ‘Buchanan Report’. This Government study is often cited as the genesis of modern traffic engineering in the UK and many people feel that we have only recently begun to unpick the damage that has resulted from the application of the principles contained within it (even if some of these principles have often been unfairly taken out of context). So my initial reaction to this approach, as someone who likes to think of themselves as a relatively forward thinking engineer, was instinctively one of concern. However, the Dutch reality is a network that clearly indicates the behaviour expected of road users in different locations, with none of the apparent negative aspects often associated with cellular networks.

This ‘whole network’ approach, rather than just focusing on strategic corridors, which tends to dominate in the UK (such as London’s Cycle Super Highway), is an important one, especially if we want to move on from a bunch of middle-aged, white males in lycra only using their bikes for the commute. Providing appropriate conditions for cycling straight from your front door appears to be critical in making cycling attractive not just for a keen few for a specific journey, but for everyone most of the time.

At a more detailed level, the ability for traffic engineers to design junctions that allow crossing pedestrians to have priority over vehicular traffic is, in my mind, an imperative. I suggest that if Dutch engineers were restricted by the same regulations that govern UK signal design then many of their junctions simply wouldn’t work in traffic capacity terms. Conversely, it would be almost impossible to design an urban signalised junction in the UK with comparable levels of cycle infrastructure as the Dutch whilst maintaining ‘realistic’ levels of traffic capacity. This is an area that the DfT really need to look into.

The final observation that I will leave you with relates to critical mass. As an example, roundabouts with on-carriageway, annular cycle lanes have been tried in the UK but with limited success. Some have even been removed following safety concerns. They have been suggested as part of projects that I have been involved with, but have always been dropped for this reason. And yet in Utrecht we stumbled across just such a roundabout that appeared to be working well. The point that I am trying to make is that much of the infrastructure that we saw, both in Utrecht and particularly in the centre of Amsterdam was in many cases no better than what we occasionally have in the UK and particularly in London. The difference seemed to me to be a far greater number of cyclists that meant that drivers were always aware of their presence. This is something that has been raised many times before, but cannot be over-stressed. And is another important reason why we can’t simply copy Dutch infrastructure and always expect to mirror their outcomes.

One of the most interesting things to do when looking at a ‘foreign’ idea is to work-out how you might apply it to a local example. My commute into work means that I regularly cycle through Brixton along the A23, passing the Tube station and associated bus stops. And the question that I kept coming back to was ‘What would the Dutch do?’ Nowhere on our visit did I see any street that came close to replicating the scale of the different demands that this street faces. And this is something that isn’t just applicable to Brixton or even the A23, but so many of the town centres on the radial routes heading into the City Centre. London is so much larger than Amsterdam that if you were to travel a comparable distance from the centre of Amsterdam to an equivalent Brixton you would be in the middle of a field or on the side of an inter-urban motorway. Just to resolve the conflict between cyclists and buses in Brixton seems impossible without a significant mode shift away from the bus or significant restrictions on other users – a problem of such scale that I’m not sure that the Dutch have ever had to resolve. Again, bespoke solutions will be required.

As I said at the top of this piece, I think that it’s important that we move on from simply understanding what the Dutch have been and are doing and start to consider how they’ve achieved it and why. Blindly trying to apply Dutch infrastructure in the UK is unlikely to convince anyone that it is a workable solution in all but the simplest of situations or the cycle infrastructure will have to be so compromised that it’s of little value. We need to find workable solutions for achieving a 30 or 40% mode share for cycling that is workable in a UK context and be pragmatic about where this mode shift will come from and who and how other modes (but not necessarily people) will lose out.

A different result but the same engineering-led approach

Simple, hardworking streets are the norm

The other half of Dutch cycling infrastructure

Different terms of engagement help to simplify junction designs

An annullar roundabout but in a different context

How do we apply this approach to the London context