Where are all the cyclists?

Download the article as a PDF The opening lines of the Manifesto produced for the Pedal on Parliament cycle ride in April of this year summed up the ‘why’ question of cycling promotion. Cycling should be the obvious solution to many of Scotland’s ills. It is cheap, healthy, democratic and convivial, benefits local economies and makes the streets a safer place for all. Cyclist benefit themselves – physiologically their bodies are, on average, many years ‘younger’ than non-cyclists’, and they suffer less from the ‘western’ diseases that beset Scotland so – and they benefit others, cutting congestion and improving air quality. OK, it’s not as catchy as Marx and Engels’ opening to the Communist Manifesto but then encouraging cycling should be more of an opportunity to be grasped than spectre haunting the bourgeoisie. On the face of it, all the push factors are there – the rising cost of car ownership, the time spent commuting, the need and desire for both adults and children to be more active, slimmer and healthier and the need to save money on local services. But since the Scottish Household Survey started in 1999, the percentage of adults usually cycling to work or education has increased from just 1.8% to 2.5%. You can think of that as a 39% increase over the 10 years but it’s still miserably low, even when you compare the best Scottish cities like Edinburgh or Aberdeen, where 7% and 5% of adults cycle to work or education, with the best European cities like Copenhagen (30%) or Groningen (55%). Even including fair weather commuters and leisure cyclists the total percentage of adults who cycle is 9.9%. Scotland has around 4.3 million adults so if 10% are cyclists then this implies that Scotland currently has around 430,000 cyclists. We know where the cyclists are by local authority (Fig 1) and by modelling the characteristics of cyclists we can estimate the distribution of cyclists within local authorities (Fig 2).

Fig 1 – Cycling by adults in each local authority in Scotland Source: Scottish Household Survey 2009/2010 Click the image above to get a larger version Fig 2: Estimated distribution of adult cycling in Edinburgh Derived from 2009/2010 Scottish Household Survey, green shading shows areas where estimated adult cycling is above the Scottish average Click the image above to get a larger version But how many should there be? Beyond the simplistic comparison that in Edinburgh, for example, 36% of households have an adult bicycle but only 15% of adults cycle, this is a much more difficult question. Answering it involves working out how many people, who are not currently cycling, could become cyclists if conditions were different or they made different decisions. One way to estimate the potential number is to develop an explanatory model of existing cyclists, change the variables and see what happens to the estimated number – a sort of statistical experiment.

The probability of a person being a cyclist is influenced by a complex set of variables that operate at different levels. It matters what local authority you live in, whether your neighbourhood is urban or rural, affluent or deprived and at a personal level, whether you are male or female, young or old, have children or don’t and whether you rent or own your home. Each on its own is quite weakly related to cycling but in combination, each additional variable ‘explains’ an increasing amount of the variation in cycling. From the data in the SHS we can build a statistical model of the variables that influence cycling. Changing which variables are put into the model changes the probability that someone will be a cyclist and therefore changes the estimated number of cyclists. For instance, the characteristics of an area can increase or decrease the probability of being a cyclist. If you live in Edinburgh or Moray then, all other things being equal, you have a higher probability of being a cyclist. If you live in a deprived area, you are less likely to be a cyclist. On balance, area effects tend to suppress the likelihood of someone being a cyclist so removing the influence of these effects increases the estimated number of cyclists to 496,000, potentially increasing the number of cyclists by 17%. However, removing area effects removes both positive and negative influences. What if we only remove those which have a negative influence? In effect, we’re saying ‘without changing anything else, what would be the impact on cycling if we made everywhere at least as good as the better places?’ The estimate increases to 541,000 or a 28% increase in the number of cyclists. This is only a model of cycling but it suggests some practical consequences. It suggests that local characteristics suppress demand for cycling; that across the country people with the same characteristics as people who are currently cycling might also cycle if their area levelled up and copied the types of provision found in areas with higher rates of participation. Of course, ask cyclists and they’ll tell you that even the best areas in Scotland have much room for improvement. Therefore, all areas could improve on what’s already there and aim for Groningen levels of cycling. Check out Our Latest Thinking on current Scottish political and social issues here