Performance of Municipal Cycling Policies in Medium-Sized Cities in the Netherlands since 2000

Abstract

With its high cycling mode share, the Netherlands is often seen as a best practice for cycling policies. However, there is little insight into the drivers behind this phenomenon, specifically which policy interventions increased cycling rates and which did not. The knowledge gap on the effectiveness of cycling policies seriously limits the potential for learning from the Dutch experience. This paper will address this gap, by exploring the performance of Dutch cycling policies in 22 medium-sized cities since 2000. First, the existing ideas regarding the effectiveness of cycling policy are reviewed. These insights structure the exploration of data from Statistics Netherlands and the Dutch Cyclists' Union, complemented with a survey of local policy-makers by means of an explorative data-mining methodology called rough set analysis. Our findings support the following hypotheses regarding the performance of cycling policy in Dutch cities: first of all, the way cycling policy is implemented seems important: setting measurable and verifiable goals, following through with most of the proposed policy interventions, allowing for experimental measures to be explored and showing strong leadership. Second, providing adequate cycling infrastructure and decreasing the attractiveness of car use (e.g. by increasing parking tariffs and increasing the area of paid on-street car parking) seem to be key drivers. Finally, we found that external circumstances, such as demographic trends, seem to influence cycling policy outcomes. Future research is needed to test these hypotheses.

Keywords:

• cycling
• cycling policy
• effectiveness
• rough set analysis
• success factors

Acknowledgements

We would like to thank three anonymous reviewers, the journal Editor, our colleagues at the University of Amsterdam for their input and useful comments on earlier drafts, and Quinn Kelly, who contributed to the review of literature during his visit to Amsterdam as part of his Summer Internship from Cornell University. We are also grateful to all of the participants of our web survey who provided a rich insight into the effectiveness of cycling policy. Finally, we would like to thank the participants of our community of research and practice (the so-called Fietscommunity), which was facilitated by the CROW Fietsberaad and the VERDUS research funding programme, and which brought together the knowledge from practitioners and researchers in multiple sessions with approximately 70 participants.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This research was supported by The Netherlands Organisation for Scientific Research (NWO) as part of the Sustainable Accessibility of the Randstad programme.

Notes

1. Note that low bicycle use according to Dutch standards is still substantially higher than almost all other cities and towns in other parts of the world.

2. Although the exposition measure ‘fatalities against distance travelled’ is most used in the literature and in national statistics, the alternative measure ‘fatalities against time spent in traffic’ or ‘fatalities against number of trips’ is often more appropriate when comparing different modes of transport.

3. Note that in the international context even the largest Dutch cities are rather modest in size, but they are considerably larger and therefore not comparable to the Dutch cities in our sample, which are (in alphabetical order): Alkmaar, Almere, Amersfoort, Arnhem, Breda, Delft, Den Bosch, Deventer, Dordrecht, Eindhoven, Emmen, Enschede, Groningen, Haarlem, Leeuwarden, Leiden, Maastricht, Nijmegen, Tilburg, Venlo, Zaanstad and Zwolle.

4. Although this demarcation between below and above average is extremely high according to international standards (below-average values are still substantially higher than almost all other cities and towns in other parts of the world), 35% is the average value for the sample of Dutch cities and we only used it to further distinguish between cities with overall increasing cycling levels. Note that in our sample most cities with above-average cycling levels, the cycling share is 40% or higher, whereas the lowest cycling levels are around 20%.