Cycling as transport

Urban cycling is clearly a topic of immense interest to the transport research community, and this Special Issue captures key themes amongst the rapidly growing body of research. Several well-cited papers have already been published in Transport Reviews (e.g. Fishman, Washington, & Haworth, 2013; Heinen, van Wee, & Maat, 2009; Pucher & Buehler, 2008), and this set of papers builds on that foundation. This growing demand for knowledge on urban transport cycling is a reflection of the realisation of the limitations of automobile-dependent transport planning, in terms of congestion, parking problems, reduced levels of amenity and liveability, air pollution, resource depletion, climate change and road traffic injury (Hickman & Banister, 2014). Indeed Peak Car, a phenomenon that was comprehensively captured in a recent Special Issue of this journal (Goodwin & Van Dender, 2013), can be seen at least in part as a consequence of a shift away from the prime position the car has held in society since the end of the Second World War.

The growing interest in transport cycling from the research community is matched by the increasing interest shown by city governments in beginning the process of making their urban environments more bicycle friendly. The early adopter cities, such as Amsterdam and Copenhagen, are enjoying the benefits of their efforts, with some 40% of trips now being completed by bike (Pucher, Dill  et al., 2010). The Netherlands deserves special mention given that it is the entire country, not just one city, that records such strong levels of cycling (Fishman, Böcker, & Helbich, 2015; Harms, Bertolini, & Brömmelstroet, 2014). There is a lot the world can learn from the Dutch experience.

The Dutch and Danish achievements provide a window into a possible future for other cities that are now beginning to appreciate the benefits of making cycling ‘irresistible', as Pucher and Buehler call it (2008). The launch of Velib's 20,000 bikes as part of Europe's largest bike share programme in the French capital in 2007 and the preceding enhancement to the bicycle infrastructure network acted as a catalyst for many of the 850 cities that have established a bike share programme. Cities without a strong history in transport cycling (e.g. London, New York City, Barcelona, and Chicago) have all recently begun to lay the foundations for a transport system that better accommodates people who choose to cycle.

Based on findings from data collected in Berlin and London, the New Urban Mobility study (Hoffmann, Kandt, Smith, & Graff, 2015) has found that transport departments consistently overestimate car use. This is caused by a combination of changing attitudes, demographics, and emerging technology that are helping urban populations lower their dependence on the private car. The crucial point is that these changes and the growing interest in cycling (e.g. see Pucher & Buehler, 2012) is not actually about cycling, but it is about what cycling brings to cities. A greater proportion of the globe's population now reside in cities, and this level is forecast to reach 70% by 2050 (World Health Organization, 2010). The new urban age, combined with a growing awareness of liveability, has led to an emerging recalibration of city life. Interaction brought people to live in cities several thousand years ago and the bicycle has found itself as a very useful tool to facilitate this most basic of human needs. The bicycle, it appears, is now increasingly recognised as offering a functional vehicle for making the short- to medium-length trips that characterise much of city travel.

The positive shift towards greater levels of urban cycling must of course be qualified by the stagnant or even declining levels of cycling in developing countries, such as China. Some cities in developed countries have also begun to record flat or even declining rates of cycling. Australia for instance, which had a goal to double cycling between 2011 and 2016, has actually seen declining number of people choosing to cycle (Austroads, 2015). Despite representing less than 1% of kilometres travelled in Australia, cycling-related crashes comprise 15% of all road hospitalisations (Bureau of Infrastructure Transport and Regional Economics, 2015). Such statistics only help to reinforce the number one barrier to more people cycling, namely their concerns over safety (Fishman, Washington, & Haworth, 2012; Gardner, 2002; Horton, Rosen, & Cox, 2007).

This Special Issue brings together reviews of recent research within the burgeoning topic of cycling. A prominent theme within this collection of review articles is the crucial importance bicycle infrastructure plays in both determining the degree to which people choose to cycle, and subsequent levels of safety. Roger Geller, Bicycle Coordinator at the Portland Office of Transportation, has classified four types of transportation cyclists. The ‘Strong and Fearless' constitute less than 1% of the population. These are people who will cycle regardless of the quality of infrastructure. Sydney (Australia) is a good example, where the bike mode share is 0.8% (Pucher, Graves et al., 2010). The ‘Enthused and Confident' make up some 7% of the population in Geller's model and although they can ride in mixed traffic, prefer separated facilities. The largest proportion of the population are the ‘Interested but Concerned', making up 60% of the community. This is the group that require enhanced bicycle infrastructure before they will feel comfortable making trips by bicycle. Most programmes seeking to boost levels of cycling need to be focused on this 60%. The final category are known as ‘No Way No How' (33%), with little prospect of attracting this group to cycle. Whilst these categories were developed specifically for Portland, it is reasonable to assume that other populations have similar profiles, and many other cities have adopted this segmented approach when designing cycling policy.

In the first paper, Buehler and Dill (2016) identify research on bicycle infrastructure and the impact it has had on cycling, and how research has developed from analysing individual treatments through to an examination of the impact of entire networks on cycling participation. A key theme from this review is the preference for separated bicycle facilities and an aversion for cycling on roads with motorised traffic (e.g. see Garrard, Rose, & Lo, 2008; Heinen, van Wee, & Maat, 2010). This stated preference was supported by revealed preference data from Vancouver, where it was found that cyclists divert from their most direct route to ride on traffic-calmed streets (Winters & Teschke, 2010). Whilst the number of studies on the influence of bicycle infrastructure on cycling has grown rapidly since 2010, the authors identify the need for further research using a statistically representative sample of both cyclists and the general population, with an oversampling of cyclists because riding remains a minority mode of transport in so many cities and countries. Many of the available studies reviewed came from cities with incomplete bicycle networks, and there is a need for more research from the cities with mature bicycling networks (e.g. the Netherlands). Another research gap relates to the relative lack of longitudinal research, as most is cross sectional. This would help determine the causal links between cycling levels and bike networks. As individual bicycle infrastructure links are developed, there will be an increasing need to measure the network effect of a connected system of bicycle routes. As the authors identify, it may well be the case that the “effect of the network is greater than the sum of its parts” (Buehler & Dill, 2016, p. 15).

It is well established that in countries and cities with low rates of transport cycling, those that do participate occupy a very narrow demographic — typically young males (Pucher & Buehler, 2011; Pucher, Greaves  et al., 2010). Countries with high levels of cycling, such as the Netherlands, have a much greater gender balance, to the extent that in the Netherlands, mode share for cycling is in fact slightly higher for women than men (Fishman et al., 2015). A pertinent question therefore is whether growth in cycling participation leads to a wider demographic of people riding. This is the question three of the UK's most prolific cycling researchers (Aldred, Woodcock, & Goodman, 2016) have sought to answer in the second paper. The authors used English and Welsh Census data from 2001 and 2011 and found no increase in female cycling within the regions that had recorded a general increase in bicycle riding. A differential threshold effect may apply — where the point at which females feel safe cycling is different to men, potentially requiring greater separation from motor vehicles. These different infrastructure preferences need to be addressed within the policy and planning frameworks in order to address the current gender inequity that is apparent, not just in the UK but also in North America, Australia, and elsewhere. It appears that it may be necessary to create specifically targeted policies and infrastructure to appeal to specific groups currently under-represented in cycling participation and this is one of the key conclusions raised in this paper.

An increasing body of evidence has emerged showing the negative health impact of sedentary lifestyles (e.g. diabetes, hypertension), which has been shown to be amplified by car use (Warren et al., 2010). Studies on the economic impact of cycling consistently find health as having the largest economic benefit (e.g. see Fishman, Ker, Garrard, & Litman, 2011; Mulley, Tyson, McCue, Rissel, & Munro, 2013; Transport and Main Roads, 2011). Götschi, Garrard, and Giles-Corti (2016) bring together the diverse literature that exists on the health issues associated with cycling. The authors found that population levels of physical activity are falling well short of World Health Organisation recommended levels, and cycling for transport provides sufficient metabolic intensity to protect against sedentary lifestyle disease. Götschi et al. argued, “integrating cycling into daily routines provides a promising approach to increase physical activity, given the many people spend 30 minutes and more commuting daily, yet struggle to find the extra half-hour to exercise” (2016, p. 2). The authors note that a dose–response relationship is evident, meaning that there are diminishing returns for additional minutes of physical activity. A key finding is that the greatest health benefit occurs when inactive people become active through cycling. This remains a challenge from a policy perspective.

In this paper, the health benefits of cycling were found to outweigh the negative impacts from crashes (e.g. see de Hartog, Boogaard, Nijland, & Hoek, 2010). However, the safety risks associated with cycling “play a greater role for the individual, as they affect crash victims immediately and deter potential cyclists from riding” (Götschi et al., 2016, p. 19). This issue provides a common theme throughout all the articles, and Götschi et al. (2016, p. 19) identify the Netherlands and Denmark as the safest countries in which to cycle, with five times less fatalities per 100 million kilometres cycled than US cyclists, citing research by Pucher and Buehler (2012). This is consistent with the findings in Buehler and Dill (2016), where riding on separated bicycle paths reduced crash probability almost tenfold (see also Teschke et al., 2012). The authors conclude by making the point that the best approach to maximising the population health rewards offered by cycling is to make it safe and convenient, and this is a central theme threaded through all these papers.

The idea for this Special Issue was conceived whilst working in the Netherlands, in an office that overlooked an incredibly busy bicycle path. A surprisingly large proportion of these cyclists were riding electric-assist bicycles (e-bikes). Originating in Japan in the early 1980s (Rose, 2012), the Netherlands now has the highest proportion of e-bike sales per capita of any country in Europe, accounting for one in six bicycle sales. In the fourth paper, Fishman and Cherry (2016) capture pertinent themes that have emerged in the last decade of research on e-bikes. The article documents consumption patterns in different parts of the world, which reveal that China is the world leader in e-bike use, though they typically resemble a ‘scooter', with either non-existent or non-functional pedals. The major motivating factor for e-bike use is the ability to maintain speed with less effort. Importantly, e-bike use has been found to provide the necessary levels of physical activity to protect against sedentary lifestyle disease, although the intensity is found to be less than a conventional bike. Interestingly, some studies have found that e-bike users ride more than they would have otherwise, potentially providing additional benefit, not just in terms of physical activity, but also in terms of transport system efficiency, provided that they are used as a replacement for motor vehicle trips. As battery and motor technology continue to improve, it is likely e-bike sales and use will continue their sharp rise.

Another rapidly growing sub-topic within research on transport cycling is bike share, which now represents a very active area of empirical research. This is perhaps no surprise, given that there are now over 850 cities with bike share programmes, up from just a handful in the late 1990s. The next paper (Fishman, 2016) provides a synthesis of the literature published in the last two years and builds on his earlier review (Fishman et al., 2013). The rapid growth in bike share can be explained by the confluence of the need to refocus urban transport policy towards sustainable transport, coupled with increasing availability and affordability of the technology that underpins modern bike share programmes. Convenience is the key motivating factor explaining bike share membership, which encompasses the distance between home and work and the closest docking station (where bicycles are picked up and returned), as well as the ease with which people can spontaneously use the system. Although it appears that the majority of cities with bike share programmes are satisfied with their level of use, there are lessons learnt from cities with underperforming bike share programmes. Melbourne and Brisbane (Australia) have amongst the lowest levels of use (between 0.3 and 1 trip per day, per bike), due to a combination of deterrent factors which include mandatory helmet legislation, poor bicycle infrastructure, and, in the case of Melbourne, a very small number of bicycles relative to the size of the city (e.g. see Fishman, Washington, Haworth, & Mazzei, 2014). The growth in the number of cities establishing bike share shows little sign of abatement, and many of these new systems are likely to employ innovative technology, including in-built, solar-powered Global Positioning System (GPS), Near Field Communication and electric motors. Many of these features will enhance the convenience with which people can use bike share, which is likely to boost usage rates. In the near future, it is expected that public transport smartcard integration and access via tapping one's mobile phone will become the standard method by which people access bike share.

Continuing the theme addressing hot topics within transport cycling research, Romanillos, Zaltz Austwick, Ettema, and De Kruijf (2016) review the techniques and research outcomes made available through the automated collection of ‘Big Data' associated with cycling. Relative to other forms of vehicular transport, cycling has remained largely deficient in routine data collection. GPS data, live point data, and journey data now present important opportunities for researchers to gather information on the route selection, speed, and volumes of cyclists, helping to fill the data gap that has existed on the journeys made by bicycle. The ubiquity of the smartphone, in combination with commercially developed Apps (e.g. Strava), enables individuals to collect data on their ride history, which can be aggregated into population datasets (e.g. Strava Metro), for use by government transport planners. These data can be used to identify route preferences, helping to inform the future development of a city's cycle network. As identified by the authors, the key limitation of commercially available data of this type is that individual journey data are not available, due to privacy concerns. Modern bike share systems automatically record when a bike leaves and returns to a docking station, and this has been used to provide useful insights into usage patterns (e.g. see O'Brien, Cheshire, & Batty, 2014). The application of Big Data for bicycle planning is still very much in its infancy, but is likely to provide agencies with a very important source of information from which to make evidence-based decisions regarding the development of bicycle infrastructure networks. It is important to recognise whilst Big Data clearly offers great knowledge-building potential, the veracity of the data (possible biases, noise, or abnormality in the data) presents ongoing difficulties to the users of these data. Additionally, we may know little about why a population exhibits certain travel patterns, and this supports one of the conclusions of Romanillos et al. (2016), namely that we need to know more about the socio-demographics, attitudes, and motivation for travel behaviour. Much of these contextual data cannot be gleaned from a very large spreadsheet and instead rely on additional data collection techniques. When used in combination, it is likely that these data will provide important insights for both the research community and transport practitioners.

As previously stated, no country matches the Netherlands, either in the proportion of the population who cycle, or in cycling safety (Schepers, Twisk, Fishman, Fyhri, & Jensen, in press). It is commonly said that the Dutch are ‘blind to cycling'; meaning that it is such an ordinary activity, undertaken by such a broad section of the community (13% of trips by those 75 years and over are by bike),¹ that it has simply not warranted much attention, until recently. In typical Dutch fashion, they spent the last 40–50 years quietly  building what is unarguably the most extensive bicycle infrastructure network in the world. Thankfully, a growing body of Dutch researchers and practitioners are beginning to tell what amounts to one of the most successful stories in sustainable transport planning, namely the bicycle friendly transition of the Netherlands. Harms, Bertolini, and Brömmelstroet (2016) document the policy interventions responsible for high levels of Dutch cycling. The authors use data from Statistics Netherlands, as well as the Dutch Cyclists' Union, in combination with the results of a survey of local policy-makers. The findings from this analysis suggest that some of the most important factors leading to successful bicycle participation outcomes include setting measurable and verifiable goals and implementing the policy interventions proposed in strategic plans. Several other interlinked factors were also found to impact on cycling success, including the willingness to apply experimental measures, innovative bicycle infrastructure, and decreasing the attractiveness of car use (Harms et al., 2016). As cities around the world struggle to manage the competing demands on finite road space, this latter point is especially pertinent. It suggests that a reallocation of road space may be necessary in order to overcome the perceived and real safety concerns that hold back would be bicycle riders in countries with low cycling levels. This is a recurring theme amongst all the papers.

The number of cycling conferences held in the UK, Australia, and North America seems to have increased dramatically over recent years. The final contribution is a Comment piece from the former Chairman of Cycling England, Phillip Darnton OBE, who takes aim at the phenomenon of cycling conferences. Ironically, I met Phillip at a cycling conference, in which he delivered the Keynote address titled Are cycling conferences a waste of time? (Darnton, 2013). Obviously the organisers did no't think so … but he raised an interesting and seldom asked question: Why do cycling advocates talk to themselves? Darnton argues that rather than preaching to the converted, a transition needs to occur in which it no longer becomes “about cycling” … it's about a sense of Place” (Darnton, 2016). Moving the discussion beyond ‘cycling' appears particularly plausible given that for most countries and cities, the bicycle remains a marginal mode of transport, carrying a very small minority of all trips. This Special Issue contributes to transport research and policy development and has gone some way to show that it is not ‘about cycling', but it is about sustainable, productive, and prosperous cities.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1. According to the Dutch National Travel Survey (Central Bureau of Statistics, 2014).