On the Buses: A mixed method evaluation of the impact of free bus travel for young people on the public health

Article history

The research reported in this issue of the journal was funded by the PHR programme as project number 09/3001/13. The contractual start date was in February 2010. The final report began editorial review in August 2012 and was accepted for publication in March 2013. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The PHR editors and production house have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the final report document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

Helen Roberts is a member of the NIHR Journals Library editorial board.

Copyright statement

© Queen’s Printer and Controller of HMSO 2014. This work was produced by Green et al. under the terms of a commissioning contract issued by the Secretary of State for Health. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

Introduction

Transport policies and systems are increasingly accepted to have the potential to be both health promoting and harmful to health,1,2 and to contribute to the generation, maintenance or mitigation of health inequalities. 3–6 However, the evidence base in this area remains relatively underdeveloped, with few evaluative studies which have examined the health and related outcomes of changes to transport policies7,8 and few studies which have identified the costs and benefits specifically of public transport use,9 although methodological work in this area is being developed. 10,11 This study aimed to contribute to this evidence base by evaluating the impact on the public health of a transport intervention in London: the introduction of free travel for young people.

The intervention: free bus travel for young people

In 2005, the Greater London Authority (GLA) granted secondary school-aged children unlimited travel on buses and trams displaying the London Buses symbol (both within and just outside London,12 replacing a reduced, 40p flat fare for each journey on the London bus network. This fare exemption was extended a year later to include 17-year-olds in full-time education (p. 7)13 and now also includes all 18-year-olds (and some 19-year-olds) in full-time education or on a work-based learning scheme (pp. 8–9). 14 To access free travel, young people apply for an electronic photo card called a ‘zip’ card, which is tapped on a reader on entering the bus. As well as granting the cardholder unlimited free travel on all buses, the zip card also acts as a conventional ‘Oyster’ card, used by most residents and visitors in London. This can be loaded with pre-pay or travelcards for the cardholders to use on other parts of the Transport for London (TfL) network [tube, Docklands Light Railway (DLR), London overground and most National Rail services operating in the capital] at a discounted rate (pp. 6–11). 14

The stated aims of the scheme were ‘to help young people to continue studying, improve employment prospects and promote the use of public transport’ (p. 7). 13 That is, it was aimed first and foremost at mitigating the potential social exclusion effects for young people of fare-based urban transport systems. 15 As stated more recently on the TfL website:

Granting young people free travel is part of the Mayor’s strategy to embed more environmentally sound travel habits from an early age while helping young people to unlock education, sport, leisure and employment opportunities. 16

By removing any need to pay, at the point of use, for travel on buses, it was argued that young people would be better (and more equally) able to access goods and services (schools, libraries, leisure facilities, etc.), thus both improving social inclusion in the short term and contributing to longer-term policies to improve the sustainability of London’s transport system.

However, the intervention is likely to have had other implications for health and well-being. Policy concerns have centred on the potential impact on young people of being at risk of, and the perpetrators of, crime;17,18 the impact on older citizens of buses with large numbers of young people; and the possible effects on ‘active transport’ at a time when health policy is geared towards encouraging walking and cycling. The latter issue is one that has explicitly been aired by both politicians (see Appendix 1) and practitioners:

I would urge TfL to scrap concessionary bus fares for children in London . . . They should be walking or cycling these trips for the sake of their own health and fitness. Yet many of them are taking the bus for just a stop or two – and getting fatter and fatter . . . It is almost impossible to get secondary school kids on their feet or on their bikes in the face of the free [bus travel]. It’s high time it was abolished. 19

To date, there is little empirical evidence that would inform debate. Given the importance to public health of both addressing issues of sustainability in transport and the threatened ‘obesity epidemic’,20–22 evidence on the health effects of interventions such as free bus travel is urgently needed.

Context: London’s unique transport infrastructure

London’s transport infrastructure is unique in the UK. First, since the deregulation of bus transport in 1985, it is the only region within Great Britain with a regulated bus system,23 with the elected Mayor of London having executive control over the transport in the Greater London area. Over the past 10 years, and since the establishment of the GLA as a strategic governing authority for London in July 2000, London’s bus network has been subject to significant operational changes (including changes to bus ‘service levels’ as well as to the ways that contractual agreements between TfL and bus operators are monitored and regulated). 24 In the transport field:

[S]ervice levels can be defined according to a number of dimensions, the key ones being the frequency of public transport services (services per hour), the hours they operate (period of operation), where they operate and the origins and destinations they serve (both related to network coverage). 25

These changes have been driven by an explicit commitment to public transport, with the GLA’s first Mayor, Ken Livingstone (2000–8), stating that to resolve the problems posed to ‘the business efficiency and quality of life of the city’ by an inadequate transport system, ‘[t]he only viable approach . . . is one where passenger travel to, from and within central London must primarily be served by public transport’ (p. 12). 26 These policies have increased the numbers of bus passengers and bus trips in London, with buses now carrying around 2.2 billion passengers each year (p. 139). 27 Public buses in London now operate according to a complex management and funding structure in which TfL’s role is to plan routes and monitor service quality, as well as manage bus stops, stations and other support services. The bus services themselves are operated largely by private sector companies under contract to London Bus Services Limited (‘London Buses’), part of TfL (see Appendix 1).

Car ownership is lower in London than in other areas of the UK, and a number of policies in addition to the expansion of public transport aim to reduce car use. An important one is the congestion charge, introduced in 2003 and currently £10.00 a day, which is levied on cars travelling into London within certain times. This has implications for young people’s travel patterns, which are unlikely to be typical of the UK as a whole. Department for Education (DfE) data, for instance, record higher levels of public transport use for school journeys, and lower levels of car, walking and cycling use for school journeys than for the rest of England (see Appendix 1, Table 29). 28

Finally, the Greater London area covers 33 boroughs, which differ in terms of transport modal share. One measure of transport accessibility is public transport accessibility level (PTAL) scores (1 is poor and 6 excellent, and 0 is no public transport accessibility within the specified parameters). As the London Travel Report29 puts it, in the calculation of PTAL scores analysts factor in:

The access time (by walking) from the point of interest to public transport service access points (SAPs, e.g. bus sops, stations) within a catchment area; the number of different services (e.g. bus routes, train services) operating at the SAPs; and levels of service (i.e. average waiting times, with an adjustment for the relative reliability of different modes). 29

As Figure 1 shows, public transport is, in general, less available the further one goes from the centre of London. With less access to public transport, and fewer disincentives for driving, car travel remains higher in the outer London boroughs than in the inner London boroughs (modal share 41% and 26% respectively). 31

FIGURE 1.

Map of pan-London PTAL scores. Reproduced with permission of GLA (map taken from p 57). 30

Although the distinction between inner and outer London is significant, there are other differences between the 33 boroughs in London in terms of typical transport modes used. As Figure 2 shows, bus density (indicated by the number of bus stops) varies across with boroughs, as well as across inner and outer London.

FIGURE 2.

Bus density in London, by borough.

Young people’s travel in London

The free bus travel scheme introduced for young people in London in 2005 was introduced not in isolation, but within a context of other explicit policies and secular changes that shape the ways that people in London travel and which potentially impact on health. 2,32,33 The policies and broader cultural shifts that provide the context for young people’s travel over the period of this study include:

• Substantial improvements to the bus network (and public transport network more broadly) since the establishment of the GLA in 2000 (see above and Appendix 1).

• The congestion charge scheme, which provided disincentives for driving in the central area and which has reduced car journeys within the zone. 34

• The expansion of road engineering interventions, such as 20 miles per hour (mph) zones, which have made local environments safer for walking and cycling33 and therefore may have encouraged active travel.

• The increasing distances that young people travel to school in the UK. National Travel Survey (NTS) suggest that between 1985 and 2004 the average journey length to school increased by about 800 m. 35 In London in January 2010, the average distance (calculated as a straight line using postcode data) travelled to school by secondary school pupils was 1.5 miles (p. 37, table 7). 28 Of these pupils, 24.8% travelled ≥ 2 miles to school and so covered distances that might not typically be considered appropriate for walking or cycling.

• Broader cultural shifts that increase the number of children being driven to school. 36,37

• A focus on cycling in London. For instance, the Evening Standard’s high-profile ‘safer cycling’ campaign included coverage of cycling deaths in London, potentially adding to anxieties that parents feel about their children cycling to school. 38 Conversely, the work of the TfL Smarter Travel unit aims to reduce car travel to/from school and the workplace, with the outer London boroughs of Richmond and Sutton, in particular, subject to extended smarter travel campaigns. 39 Although many cycling interventions have been focused on adult cyclists, such as government tax breaks on cycling to work introduced in 1999, these form part of the backdrop of changing views of transport modes in the capital (see Green and colleagues40).

What is known already?

Assessing the impact of one transport policy on young people’s health in the context of these multiple other potential influences first entails mapping the potential ways in which transport interventions might change behaviour, and how these in turn might influence the determinants of health. We first turn to the existing literature on the links between transport and health to outline what is known already about the likely causal chains between transport policy interventions and their health and other outcomes, and to outline the changes that might theoretically result from an intervention designed to change the ways that young people travel. We summarise this research below starting with the most immediate and direct effects of transport policies – injury – and ending with the more distal and difficult to measure.

What this study will add

Theoretically, then, the free fare scheme for young people may have had a range of effects on the public health of Londoners. Evidence to date suggests that effects on ‘processes’ such as travel mode choice and travel mode distribution are likely to affect health behaviours and outcomes such as active travel and injury rates. Reducing fares (in this case to ‘no cost’) is likely to increase use of bus travel in the target population74 displacing other modes of transport and/or creating additional journeys.

To summarise, the risks and benefits to health from the provision of free public transport are likely to accrue from the increased availability of transport and changes in the share of modes of transport used (e.g. switching from walking to bus travel). In the USA, where public transport is more likely to replace car use than walking and cycling, there is some evidence that increasing access to public transport can increase activity levels. Increased walking to public transport is enough to have had a public health impact on obesity, particularly for men. 55–57 In England, free bus travel has been identified as providing a benefit for the health of older people. 75,76 However, it is possible that in settings such as London, with lower private car use than other parts of the country,77 improving access to affordable public transport may reduce the amounts of active travel undertaken by replacing walking or cycling rather than car use. As the risk of RTI varies by transport mode,44 any modal shift is likely to have implications for injury rates. Young people are at particular risk of assault,78 and greater access to public transport potentially increases this risk. More tangential benefits which may be associated with young people’s increased access to public transport include increased social inclusion, and decreased future reliance on private car travel. They are also likely to have effects on broader determinants of health, such as social inclusion and independent travel. These broader health implications of well-being that arise from social inclusion or the ability to make independent choices about travel are important, but there is in general less evidence on the wider social and health effects of different transportation choices79 and real challenges in operationalising concepts such as ‘inclusion’ for research. 80 Clearly, an evaluation of the public health effects of a transport intervention needs to account for potential effects on social inclusion and well-being, but these are difficult to measure. Finally, in addition to benefits or costs for the target group (12- to 17-year-olds), there may be effects on other transport users if they are displaced from buses.

The free bus pass scheme is a ‘natural experiment’ which allows us to evaluate these potential pathways linking a large-scale transport intervention to health determinants, behaviours and outcomes. Given the range of health benefits and disbenefits that theoretically arise from the intervention; the complexity of the system in which this intervention was introduced and the difficulties in operationalising distal health outcomes which are important but (to date) under-researched, any evaluation will require a mix of methodological strategies. This study therefore aimed to evaluate the impact of free bus travel on the public health by identifying the best available strategies to explore these pathways, using a mixed-method design. Our specific aims were to:

1. provide empirical evidence for the impact of this intervention on key health behaviours and outcomes (e.g. injuries, active travel) for young people

2. explore the effects on the determinants of health (e.g. access to education and training)

3. identify the effects of increased young people’s access to bus travel on older citizens

4. develop and apply methods for economic assessment, and

5. contribute to the development of methods to strengthen causal inference in non-randomised designs.

The approach we took to meeting these aims is outlined in the next chapter.

Introduction

Drawing on existing research, Chapter 1 outlined the range of health impacts free bus travel may have had on the public health, and summarised the aims of an evaluation of these impacts. Like many other policy interventions with potential impacts on the determinants of health, it is impossible to generate ‘best evidence’ such as that from a randomised controlled trial (RCT) on the effects of the intervention. The scheme has already happened, there are no obvious control groups, and no opportunities for the research team to control exposure to the intervention. The intervention could, however, be considered a ‘natural experiment’, in that although not under the control of the research team, it is amenable to research using natural variations in exposure,81 such as between the target group (young people) and others in the population. Given the urgent need to improve the public health evidence base in general,45 and for evidence on transport interventions in particular, there have been calls to exploit ‘natural experiments’ to contribute ‘good enough evidence’ to inform policy decisions. 9,82 The free bus scheme is one such opportunity. This intervention has a number of advantages which make it suitable as a potential natural experiment:81 (1) a RCT is not possible; (2) we already have (from research evidence) some reasonable expectation that health impacts will accrue from changes in transport behaviour; (3) there are available secondary data sources on some of these impacts that cover both pre- and post-intervention periods; and (4) there are potential ‘exposed’ and ‘non-exposed’ individuals. Finally, there is a policy incentive, given that there is little evidence to inform current debate, and London’s transport interventions are of wider interest in the context of policy drivers to increase the sustainability of transport systems.

This chapter describes the design and approach we took to evaluating the free bus travel scheme as a natural experiment. We discuss the overall design and aims of the study, the methodological challenges in evaluating ‘natural experiments’ of this type and how we addressed them, and the specific methods used for the quantitative and qualitative components of the study.

Study design: the logic model linking free bus travel to health outcomes for young people

To map the possible range of public health effects which could be evaluated, we first developed an initial (summary) ‘logic model’83,84 of the pathways hypothesised to link the intervention with outcomes relevant to public health. This was developed, after reviewing the literature summarised in Chapter 1, in collaboration with key stakeholders, including our steering committee, young people [participants in the London School of Hygiene and Tropical Medicine (LSHTM) Young Scientists’ programme85] and colleagues, to ensure we had captured outcomes that were important to policy-makers, the public and researchers.

As discussed in Chapter 1, the introduction of free bus travel hypothetically impacts on health through a number of pathways. Our first three aims were to assess the evidence for these pathways, and specifically to:

• provide empirical evidence for the impact of this intervention on key health behaviours and outcomes (e.g. injuries, active travel) for young people

• explore the effects on the determinants of health (e.g. access to education and training)

• identify the effects of increased young people’s access to bus travel on older citizens.

The first step was to outline these pathways and to identify and include the main outcomes of interest. Second, we identified what indicators were possible and feasible to measure for the variables on the pathway and the key outcomes of safety, active travel, social inclusion and effects on older passengers. Third, we identified the potential sources of evidence for these indicators: existing quantitative data sets which covered both pre- and post-intervention periods; primary qualitative data and reviews of the literature.

At this stage, other possible (distal) health effects of free bus travel (e.g. the impact on pollution), were eliminated from the model. Evaluations of other large-scale transport interventions, such as the congestion-charging scheme, suggested only modest changes result from transport mode shifts,86 and effects from the free bus scheme were unlikely to be measurable or distinguishable from those from other causes. To simplify the task of assessing transport mode shift, we also did not assess the impact of the scheme on underground (tube) and overground train journeys, given that these form a relatively small proportion of young people’s travel in London (see Appendix 9, Table 40).

The main pathways selected as important to evaluate are summarised in Figure 3, which also shows the main data sources used to evaluate each pathway.

FIGURE 3.

Hypothesised causal pathways.

Finally, to address our fourth aim (to develop and apply methods for economic assessment) we identified sources of data for economic values for these health outcomes, largely from existing literature, and costs of the scheme, largely from TfL.

Aims, hypotheses and objectives

The model in Figure 3 suggested a number of hypotheses related to the first three aims. Specifically, that the free bus scheme would be associated with:

1. An increase in bus use and overall ‘independent’ travel (the latter represented from available data by all non-car travel), but a reduction in active travel (walking and cycling) and car use, among the target age group. (Independent travel refers to travel young people do without guardians. With no direct measure of this, we use a proxy measure of all travel except car travel and, for those aged < 17 years, motorbike travel.)

2. A reduction in bus use and trips < 1 km made by people aged ≥ 65 years, especially during the hours when children usually travel from school. (We subsequently chose to include 60- to 64-years-olds to increase our available sample size and thereby increase statistical power to detect differences between groups. The older age group was therefore aged ≥ 60 years.)

3. A reduction in RTIs in the target age group.

4. An increase in intentional injury rates in the target age group.

   1. If these changes were associated with the free bus scheme, we also hypothesised that:

5. Changes will be more pronounced in the inner-London boroughs (with denser bus networks) than in outer-London boroughs.

6. Changes will be more pronounced in boroughs with a known higher uptake of free bus travel.

7. Changes in distance/frequency of bus travel, independent (non-car) travel, active travel and in injury incidence, are greater in households with low income.

8. Reductions in car use will be greater in households with high income.

9. Changes in distance/frequency of bus travel and active travel, and in injury incidence, will be the same across all ethnic groups.

For a number of hypothesised effects, such as decreased dependence on car travel, and reduced transport poverty, there were no available direct measures from secondary data sources. For these, we aimed to generate primary qualitative data to provide some insight into broader determinants of health and well-being. These qualitative data would also be used to provide evidence for the plausibility of causal claims made on the basis of quantitative analysis, through identifying potential mechanisms, or (for instance) generating data on young people’s accounts of their travel which could be compared with data from secondary sources. The objectives for the qualitative component of the study were therefore to:

1. identify the ways in which young people and older citizens understand the role of bus travel and other transport modes in facilitating and constraining their well-being

2. identify, from analysis of interview data, plausible pathways by which policy, access to transport and behaviour interact to impact on well-being.

Good practice in evaluating natural experiments calls for an assessment of value for money considerations. 81 We therefore also aimed to undertake an economic evaluation of the scheme which took into account evidence on the public health effects.

This study therefore aimed to evaluate the impact of free bus travel on these outcomes, using a mixed-method design, and to assess the economic costs and benefits of the scheme. Our final aim was a methodological one of contributing to the development of methods to strengthen causal inference in non-randomised designs. In Chapter 10 we discuss the methods used to do this, in particular through revisiting and revising the causal model in Figure 3 (see Chapter 10).

Challenges in assessing the public health impacts of transport interventions

Calls for ‘evidence’ on the public health effects of policy interventions45 generate challenges, given the difficulties in making causal claims about the relationships between policies and their intended and unintended outcomes. ‘Real world’ interventions are inevitably messy, often in themselves complex (with heterogeneous, often ill-defined, components)87 and implemented in poorly bounded target populations and settings, with unknown exposures for those intended to benefit. The systems in which they are implemented (cities, schools, countries) are simultaneously subject to a range of other ‘interventions’, both explicit policies and less easily delineated cultural and social changes. More challenging, methodologically, are the feedback loops – both predictable and less obvious – that might mean causal directions change over time or are modified by unknown other influences.

The intervention evaluated in this study is a typical example. The introduction of free bus travel for young people in London had wide-ranging intended consequences, including addressing social inclusion, through increasing access to education and training, and reducing future car dependence among young Londoners. Given concern about obesity in young people, there has also been political interest in the unintended consequences of potentially reducing the amount of ‘active’ transport. This is a classic ‘messy’ intervention. Without evidence from randomised trials (hardly logistically possible), or even plausible control settings (there are no obvious comparators to London, given its size and unique transport infrastructure), designing a study with high internal validity (i.e. likely to make credible claims about the causal effect of the intervention on public health in London) is challenging. Also of concern is the question of external validity: how can we make credible claims on the more general causal question of whether public transport concessions are likely to benefit or harm the public health?

The ‘complexity’ of both the intervention and the setting are immediately apparent. Free bus travel was introduced in two stages, in 2005 (for those aged < 16 years) and in 2006 (for those aged < 18 years). Furthermore, ‘the intervention’ is not neatly delineated in time, as it replaced a range of other concessionary fares for young people, and, as young people had to apply for a photocard (now called a ‘zip’ card), uptake was gradual and unevenly distributed across the study area (Greater London). Given London’s unique transport infrastructure, there are no obvious external comparison populations of young people in other cities. Potential confounders in any before-and-after design range from the theoretically knowable (e.g. other interventions potentially influencing the behavioural outputs in which we are interested) to the more general and difficult to measure cultural shifts, such as rising concern about children’s independent mobility or changes in the ways in which urban environments are conceptualised.

Challenges to making credible claims also arise from the relatively complex causal pathways that connect the intervention (providing free bus travel) and the outcomes (various health behaviours, determinants of health and health outcomes) summarised in Figure 3. Providing free bus travel might increase or decrease the amount of walking or cycling young people do, which in turn may positively or negatively influence health. More walking and cycling may have implications for obesity or future CV health,52,88,89 for instance, but also increase exposure to pedestrian or cycling injury risk. 44 It is theoretically possible to quantify these different effects, and calculate (however imprecisely) likely aggregate benefits. 79 However, the evidence base on links between active transport and health outcomes is currently weak90 (see Appendix 8), and is particularly weak for young people, for whom putative health benefits may be too far in the future to measure. What is even more challenging is accounting for (possibly unknown) feedback loops which potentially change these effects within the system. As Shiell and colleagues91 note, the challenges of evaluating interventions in complex systems require new ways of thinking to deal with self-organising systems which may be sensitive to initial conditions and in which components are tightly interconnected.

One potential example of a feedback loop is that the effect of walking on health may be modified by both known and unknown factors, such as the cultural meaning of walking. Bostock60 for instance documents the negative impacts on health of walking for those who have no choice but to walk. Such factors might not only change the likely benefits of walking for different sectors of the population, but also change differentially as a result of the intervention, thus modifying the equity effects of an intervention over time.

Notwithstanding these challenges, development of the evidence base for public health in this area is important, and ‘natural experiments’, despite their inherent weaknesses, may offer the best way forward for evaluation. 9,82 Ogilvie and colleagues9 also suggest that single studies of transport interventions are unlikely, on their own, to provide credible evidence for causal claims, and we need to begin to build the evidence base to generate ‘good enough’ evidence for policy and for potential future integrative reviews. This study aimed to provide one such contribution, using a pragmatic, mixed-method and iterative approach to addressing the challenges above.

The general approach: a mixed-method study

This evaluation treated the introduction of the free bus scheme as a ‘natural experiment’ to which young people in London were exposed after 2005/6, and other populations (young people before 2005, adults in London, the population outside London) were not. This therefore permitted a number of comparisons, including some ‘change-on-change’ comparisons using before-and-after and comparative populations. We used a mix of secondary quantitative data, primary qualitative data and reviews of the literature to generate evidence for the links in the causal pathways hypothesised in Figure 3.

The design was pragmatic, in that our aim was not to assess whether receiving a free bus pass had a health benefit or loss for an individual, but rather to evaluate the broader impact on the public health of the scheme as a whole. The analysis is therefore conducted at a population level. Quantitative data, largely from routine data sets [including police records of road injuries, Hospital Episode Statistics (HES) and travel diaries], were used to assess links between the intervention and bus use, mode change (active travel/car use), displacement of older passengers, and injury. Drawing on best practice guidelines for natural experiments81 we published a protocol for the quantitative component. 92 This specified hypotheses and main subgroup analyses93 (see Appendix 7). To minimise the threats from confounding, we used change-on-change analyses with adult passengers as a comparator where appropriate. To strengthen the credibility of causal claims, we employed a range of sensitivity analyses to test the credibility of inferences and (where possible from available data sources) time series or ‘dose–response’ analyses.

To explore outcomes for which there are no routine data sources, and which are more difficult to quantify, we generated qualitative data from interviews with young people and older citizens. These data were also used to offset some of the limitations in using secondary data and to explore mechanisms for putative causal relationships. Our design was iterative in that these data were also drawn on throughout the study in order to refine the analyses of routine data sets, and to refine our understanding of the logic model.

On the final link in the causal chain, the impact of active transport changes on health, we had no direct measure. We therefore conducted a systematic literature review to assess the strength of evidence for the credibility of the link90 (see Appendix 8).

The economic component aimed to represent the health consequences of the intervention and compare this with the cost of implementing the scheme. This utilised outputs from the quantitative analysis on ‘change attributable to the intervention’ to populate the benefits of the scheme, with data from TfL, Home Office and Department for Transport (DfT) to estimate costs for crime on the transport system, and additional data from TfL for costs of the scheme and operating costs. The economic evaluation involved a comparative analysis of alternative scenarios in terms of their costs and consequences.

The key alternative scenarios compared were:

• do something – the bus network is free for 12- to 17-year-olds

• do nothing – the bus network is not free for 12- to 17-year-olds.

The methods for the quantitative and qualitative components are detailed below. Details of the specific methods used for the economic evaluation can be found in Chapter 9, and for the literature review in Appendix 8.

Methods for the quantitative components

London’s transport system, comprising dense private and public transport networks and a focus on national road and railway networks, is unique in the UK. There are therefore no obvious comparator cities, or settings, which could be used to account for national trends in use of alternative transport modes, or to account for national trends in levels of safety. London has also experienced the introduction of other transport policies over the study period (e.g. London congestion charge was introduced in 2003), which may have altered choices of travel mode within the population.

Design

We estimated health impacts of introducing free bus travel for young people by analysing routine data sets on travel and safety. To control for secular trends and the effects of other transport policies, we used a change-on-change analysis to estimate any changes in travel and safety in the target population of young people that were associated with the introduction of free bus travel. This change-on-change analysis compared pre–post intervention change in amount of travel (or safety) in the target age group, with the corresponding pre–post change in amount of travel (or safety) in an older age control group (adults aged 24–59 years) (Figure 4). The age range of the control group was chosen to exclude younger adults who may also have experienced free bus travel, and to exclude older adults who may also qualify for an older citizens’ bus pass.

FIGURE 4.

Timeline showing travel surveys and introduction of free bus travel in London.

As a sensitivity analysis, we also examined pre–post intervention change in safety in the target age group in London, with the corresponding pre–post change in the target age group outside London, using national data for the rest of England.

Change-on-change analysis

To test each hypothesis (listed in Aims, hypotheses and objectives) we estimated the pre-intervention and post-intervention mean of each travel and safety indicator and calculated 95% confidence intervals (CIs) using bootstrap methods implemented in Stata statistical software (StataCorp, College Station, TX, USA). These results are presented graphically for each indicator; Figure 5 gives an example. [Line joining solid squares shows change in target age group (12- to 17-year-olds); line joining solid circles shows change in control age group (25- to 59-year-olds). Diamond (right-hand side of figure) shows the change in the target age group relative to the change in the control age group. 95% CIs are shown as vertical lines.]

FIGURE 5.

Example of graphical presentation of results showing change in travel and safety indicators.

Methods for the qualitative components

To generate data for exploring the pathways linking the intervention to social inclusion, future car dependence and the effects on the well-being of older citizens we conducted a qualitative study of young people aged 12–18 years (a few participants with disabilities were aged > 18 years at the time of interview), and older citizens aged ≥ 60 years in London, based primarily on individual and small group interviews. These data were also used to explore the mechanisms for relationships measured using quantitative methods and to further our understanding of the intervention in context.

Ethical issues

The study was approved by LSHTM’s Ethics Committee (Application no. 5635). The quantitative component relied on secondary data analysis and did not raise particular ethical issues. For the qualitative interview component, the main ethical considerations were those of ensuring adequate informed consent and maintaining confidentiality. As our young participants were all of secondary school age, we considered they were competent to make an informed decision about participation without parental consent. Our consent forms included a space for parental permission in case gatekeepers (such as schools) required this: none did. Our consent procedures were discussed with young people in the pilot phase, who commented on our information sheet and whether or not they considered parental permission appropriate. Of perhaps more significance is the potential for studies such as this to exclude particular groups from the population. Given that our primary recruitment strategy was to work with community groups to identify participants to invite, we were mindful of the possibility of excluding those who may be most marginal in other ways (such as young people excluded from school, or relatively isolated older citizens). We therefore sought deliberately to include participants through alternative approaches such as pupil referral units and personal contacts where appropriate. Observations conducted on the buses were of public behaviour and were used to inform our topic guides and interpretations only: fieldwork notes did not identify individuals, and are not quoted in this report.

Introduction

Bus fare subsidies are used as a policy instrument to achieve a number of goals, including those of equity (through addressing transport exclusion), reducing traffic congestion and sustainability (through encouraging mode shift from private car use). 23 Achieving these assumes ‘demand elasticity’: that a reduction in fares will increase the use of services. Although it is reasonable to assume that this does hold true, there is considerable debate around how to calculate demand elasticity, given that this depends very much on context: the availability and efficiency of bus services; the attractiveness of alternatives; local congestion; population ability to pay. There is evidence that increased demand from fare reductions take many years to accrue, for a number of reasons, including the difficulties for most people of making short-term changes in travel behaviour, and the cohort effects of changing habits as children, for instance, become habituated as ‘bus users’. 99

This project evaluated the effects of an intervention which reduced fares to £0. We therefore hypothesised that the first step in the causal chain would be an increase in bus use by young people attributable to the intervention. This could not be assumed, given that fares for young people were discounted before the intervention, and bus use in London was rising for the wider population (see Appendix 1) in the context of historical innovations such as integrated travelcards which facilitate ancillary bus use, and more recent policies such as the congestion charge for private cars. This chapter first reports evidence from the analysis of travel diary data to identify changes in bus use that could plausibly be attributed to the intervention. We then draw on the qualitative data to explore the role of ‘free bus travel’ specifically in young people’s travel choices in order to assess the credibility of this attribution and to provide some context for the kinds of journeys made by bus.

Quantitative evidence on impact on bus use

As described in the previous chapter we used a change-on-change analysis to estimate any changes in travel in the target population of young people that were associated with the introduction of free bus travel, rather than general changes in bus provision. Here we compare the pre–post change in weekday term-time bus travel in young people with the corresponding pre–post change in bus travel in adults aged 25–59 years. The changes to bus trips, distance and percentage of short-distance trips made by bus are shown graphically in Figure 6 (and in Appendix 9, Table 34). [Line joining solid squares shows change in target age group (12–17 years); line joining solid circles shows change in control age group (25–59 years). Diamond (right-hand side of figure) shows the change in the target age group relative to the change in the control age group. 95% CIs are shown as vertical lines. Horizontal dotted line indicates a ratio of 1 (i.e. no relative change).]

FIGURE 6.

Changes in bus use, pre- to post-introduction of free bus travel (a) Bus trips per day; (b) bus distance per day (km); and (c) per cent of short trips by bus.

Free travel enables a range of journeys

The travel diary data includes only those trips made during term-time weekdays. In qualitative interviews, young people from all ages and parts of London reported using buses for a range of journeys, including getting to school, visiting friends, getting to local amenities and for more extensive days out, across the week. In all boroughs, young people emphasised the ease of getting around, and indeed the range of sites that might be visited by bus. The fact that travel was free was reported as a key reason that buses were used for a large number of discretionary trips, including those at weekends, as well as what could be considered essential journeys, such as those to school:

I take the bus every day . . . [for] going to school, going to dancing, going to see my friends, maybe going to church . . . because it’s free . . . I can go to different places, so anywhere I want to go

Hav, 15 years

Mostly every Saturday we’ll probably just jump on a bus because we have a free [pass] and go anywhere, and get another bus from there, and another one. And we just travel, we don’t know where we’re going . . . [once] we ended up near Hammersmith, and near the West End

Isl, 16 years

Indeed, free travel was reported to have opened up the range of places that young people could go in London, as well as the frequency with which they chose to visit these places. It made possible wide-ranging exploratory trips to the city centre and neighbourhoods at a distance, and it enabled young people to maintain (geographically) wide friendship groups.

Free travel enables the bus to be a site for socialising

Bus travel was frequently an end in itself: an activity undertaken to enjoy the journey rather than to get somewhere. That bus travel was free facilitated the treatment of London buses as a site for socialising by young people, with buses becoming key public spaces in the city for young people to convene and socialise both as part of their school journeys but also in the evenings, during school holidays and at weekends. This is not to say that prior to the intervention buses had not been treated by young Londoners as a space in which to socialise with their peers. Rather, by rendering bus use free for young people on an unlimited basis in London, the intervention dramatically shifted the degree to which buses could be used in this way. The bus network became a part of the freely accessible geography of London for young people, not only as a way of getting to and from destinations but also a vital set of destinations in itself. By contrast, if fares were reinstated one participant suggested that she did not ‘think that anyone would really go out as much to be sociable’ (Sut, 15–16 years). The bus was therefore not just a vehicle for a peer outing, but a site in itself which facilitated social interaction, primarily with peers:

You see friends, when you go on the bus you can see friends on there sometimes.

Yeah, yeah, yeah.

You can talk about people as well . . . reasonable amount when you’re on the bus.

Yeah, exactly, when you get on the bus you can just see friends. If you’re on your own and then you see someone, you just see them get on the bus . . .

Sut, 15–16 years

. . . It’s one of the main things you do on the bus, if you go out with someone you sit down and talk about things

Sut, 15–18 years

Given that interacting was an important benefit of travel time, the bus was only preferable if it maximised these opportunities. Thus, when a more private conversation was wanted, or if friends were not able to take the bus, walking might be a better choice:

Sometimes I prefer walking because sometimes, when I didn’t have my Oyster card and a friend used to take the bus but then he would stop and get off and we’d walk and that would give us more time to say what we want, because you know on the bus you can’t really talk much.

Hav, 14 years

For the majority of young people, travel and sociability with peers was emphasised in accounts, particularly those in group interviews. However, many also talked of ways in which the scheme had fostered travel with family members, and enabled family outings:

Before when I didn’t have my zip card I didn’t really go out much. I didn’t because I had to basically, I had to pay and then I managed to get my zip card. I could go different places to socialise with friends and especially travelling around London because I actually quite like exploring it here with my mum . . . it’s easier for [mum and me] because my dad doesn’t have to drop us off all the time.

Hav, 15 years

Short journeys: using the bus ‘because it’s there, and because it’s free’

For short journeys, when there is no cost to the user, and buses are relatively available and accessible, there are no disincentives to using the bus, and many young people talked about using the bus to go ‘short distances, literally three stops’ (Sut, 14–18 years). Indeed, the distances travelled by bus in preference to walking could be extremely short:

My dad takes me a couple of metres down the road, it’s only about 200 m down the road. And then from then I go and get the bus to school. And then there’s only a few metres from where I get off the bus to go to school. [. . .] I’m on the bus for roughly about less than a minute.

Hav, 14 years

In young people’s accounts, this preference for bus travel over even short distances was clearly the default choice: it was, implicitly, ‘just what you did’. Indeed this assumption that the bus would be taken as the obvious choice was recognised by many young people, who noted that on reflection, they would ‘jump on a bus’ even when this was an inefficient travel strategy in terms of getting to your destination quickly:

Sometimes, because the 18 that goes past the [local community centre], that comes into Sutton, my mate and I stood there waiting for one, this is going to be completely strange to you, but by the time we got on it, we got there and thought, you know, it would have been much quicker if we just walked there.

Sut, 14–18 years

The zip intervention, which removed any economic disincentive to using the bus, was also acknowledged as the key reason for making ‘jumping on’ for a couple of stops the default option:

I think that the fact that knowing the bus is free helps me want to get on the bus, so I’m saying I might jump on the bus, if I had to pay I would not get on the bus.

Isl, 16 years

Yeah, if it was a lot, if the price was higher I definitely would not be taking buses as much as I do.

Hav, 14 years

Maximising sociability: the importance of free travel for all

If free bus travel removed disincentives to ‘just jumping on’, it did not explain why this was preferable even for journeys that might be more efficiently made by other modes that were also free, such as walking. One important contributor to making bus travel the default choice for most journeys, rather than the choice for certain journeys, was the ways in which the scheme maximised possibilities for sociability.

In young people’s accounts of travelling around London, it was clear that socialising with peers was a pervasive aspect of, and influence on, young people’s mobility, affecting decisions about whether, when, where and how to travel, ranging from the routes they chose to where they sat in the bus. Travel with friends was valued as being a defence against boredom (a pervasive theme in many accounts); more fun; and as reducing the chances of getting lost, or as at least providing someone else to be lost with. Furthermore, travelling was a site for demonstrating important aspects of friendship, such as loyalty. Young people described situations in which travel decisions were made not because they resulted in a more efficient or enjoyable journey, but because to have done otherwise might be construed as a breach of friendship, as illustrated by this story:

I got on the bus and everyone else was just left there. And then he [the driver] just, he, I was like can you open the door because you’re not letting my friends on? I was going to come off. Drove off, I had to walk all the way [back to join my friends] and that’s actually quite a long walk

Sut, 14–16 years

The participants in this focus group then went on to highlight the possible consequences of failing to show this form of loyalty, with one participant recalling ‘we’ve had arguments over buses’ and another describing having got ‘really angry’ when she was left by herself during a bus trip. As she concluded ‘Your friends just like leave you on the bus. I said, like, I’ll just sit by myself, thanks a lot. It’s kind of loyalty to get on the bus with your friends’ (Sut, 14–16 years; emphasis in original). Similarly in a different interview, one boy explained:

[If a friend loses their zip card and isn’t allowed on the bus then] you’d have to get off or something. [. . .] If you leave them, if you, if they can’t go on the bus or, they see it as a kind of betrayal.

Hav, 15 years

Travel choices were therefore not simply passive reflections of existing relationships. Rather they could also be constitutive of relationships by providing an active opportunity to demonstrate friendship (show ‘loyalty’) or fail to do so (‘betray’ a friend).

Given the importance of sociability and loyalty within friendship groups, the fact that everyone is entitled to free bus travel and could therefore all travel together is a key factor in choosing buses, at least for some journeys. Several participants, from all age groups, were explicit in giving this as a rationale for taking buses rather than other modes (such as the London underground or train, which has to be paid for), so as not to exclude those who could not pay for transport:

[We sometimes go by bus] because it’s free as well so if people run out of money on their Oyster then we’ll all go with them because we don’t want anyone to go by themselves.

[. . .] I usually travel everywhere by tube [London underground] if we can. But like you said, if some of us have got no money on our Oyster then we’ll just take the bus.

YS, 17 years

[How I go] depends if I have friends with me, ‘cos they don’t like to go on the train, so if I have friends with me then go on the bus . . . ‘cos they don’t have to pay.

YS, 16 years

That the intervention itself, rather than the general increase in bus transport, facilitated ‘travelling together’ was also evident in (rare) accounts of limitations in mobility experienced by groups in which one member did not have free travel:

One of my friends he’ll actually, that’s the only bus he can take [a particular bus route that was easy to board without paying] because he never applied for an Oyster. And I keep on moaning at him for it. But then so if we ever go anywhere it has to be on [that bus].

Isl, 15 years

The removal of financial barriers for all also increased young Londoners’ options within the context of a particular bus journey. For instance, when it came to demonstrating friendship, free bus travel removed any economic disincentive to getting off a bus that your friends could not board. A few young people explicitly stated that this had an effect on their likelihood of showing this form of ‘loyalty’:

[We get off the bus if] a couple of people get on and then some of them are left. If we didn’t, if we had to pay, I would stay on the bus. I would stay on the bus because I’m not paying no more money.

Hav, 14–16 years

More often, this absence of a disincentive was implicit in the way in which young people automatically accommodated friends without thinking, or else focused on non-economic factors when relating stories of the costs imposed by accommodating friends (e.g. waiting for a long time, waiting in bad weather, ‘having to walk’ to rejoin friends).

Bus travel, therefore, had become the default mode because it was in part a ‘lowest common denominator’ activity for groups, not only for getting to destinations, but also an activity in itself that had no direct financial costs, and to which all members of a peer group therefore typically had financial access. Most obviously these considerations applied to ‘bus hopping’ journeys, which were undertaken primarily in order to spend time with friends and only incidentally to ‘end up somewhere’. The same was generally true of recreational group journeys to destinations that were predetermined but not ‘necessary’. Bus hopping and discretionary journeys also had in common the fact that the free nature of bus travel was often a key consideration: as one girl said ‘You wouldn’t be willing to pay just to get on a bus for no reason at all’ (Hav, 17 years). A final common element across these journey types was their frequent spontaneity, such that several participants described always taking their zip card ‘just in case’ whenever going out to meet friends.

Conclusion

The increase in the number of bus trips and kilometres travelled by young people after the introduction of free travel mirrors that of adults in London, suggesting that other factors (such as improvements in the network) may be responsible for the increase in the number of journeys made by bus as main mode during term-time weekdays. However, the qualitative data suggested that the intervention had a role in making bus travel the ‘default’ mode of transport for young people in London, particularly for short journeys. Offering ‘free’ fares was an important element of this, and for some young people, cost was explicitly a factor, at least for discretionary journeys. For most, free fares facilitated a rise in short journeys, as there was no financial incentive to forgo a journey. Additionally, the bus was an important site for socialising in itself, as well as a mode of transport for instrumental reasons. Given the importance of sociability to young people’s mobility, the universality of the benefit was important in making bus travel available to all within a peer group.

In summary, this suggests that first, the intervention is best conceptualised as ‘universal free bus travel’, as it was this ‘universal’ element that determined some of the effects reported, and that second, there is weak evidence that this intervention had an important role in generating discretionary bus journeys that might not have been taken otherwise.

Introduction

That increased levels of physical activity have benefits for physical and mental health outcomes has long been recognised. In the 1950s, Morris and colleagues100 identified health benefits from bus conductors’ relatively more active jobs, compared with those of bus drivers. Given the challenges of increasing physical activity from sports or leisure participation,101,102 policy interest has focused on encouraging ‘active travel’ as one way to increase levels of physical activity in the population,1,45,89 with evidence that obesity rates are increasing in countries and settings in which ‘active travel’ (walking and cycling) is declining. 103,104 It is therefore plausible to hypothesise that interventions which increase the amount of active travel within a population are likely to have a positive impact on health, and those which reduce it may have a negative effect. Appendix 8 summarises the evidence from a systematic review of the evidence that active travel confers health benefits. Although there is little prospective research addressing this question, and no studies to date that show an association between active travel interventions and obesity, there is some accumulating evidence that active travel can have benefits for other health outcomes.

Chapter 3 identified an increase in proportion of short journeys by bus by young people, which could plausibly be attributed to the intervention, in the context of general increases in bus use made by the population of London, and a suggestion from qualitative evidence that young people’s discretionary journeys by bus had increased as a result of the intervention. This chapter addresses whether or not this increase is at the expense of ‘active travel’, that is travel by modes such as walking and cycling. We first report on changes in young people’s active modes compared with adults, as reported in travel diary data. To contextualise these changes, and assess how far the intervention may explain them, we then turn to the qualitative data.

Quantitative evidence on impact on walking and cycling

Here we compare the pre–post change in walking and cycling in young people with the corresponding pre–post change in walking and cycling by adults aged 25–59 years. Again, the changes to walking and cycling trips, distance and percentage of short trips are shown graphically in Figures 7 and 8 (and in Appendix 9, Table 34). [Line joining solid squares shows change in target age group (12–17 years); line joining solid circles shows change in control age group (25–59 years). Diamond (right-hand side of figure) shows the change in the target age group relative to the change in the control age group. 95% CIs are shown as vertical lines. Horizontal dotted line indicates a ratio of 1 (i.e. no relative change).]

FIGURE 7.

Changes in walking, pre- to post-introduction of free bus travel (a) Walking trips per day; (b) walking distance per day (km); and (c) per cent short trips by walking.

FIGURE 8.

Changes in cycling, pre- to post-introduction of free bus travel (a) Cycle trips per day; (b) cycle distance per day (km); and (c) per cent of short trips by cycle.

Bus use replaces walking for short trips

In the last chapter, we identified a preference for using buses because they were free, accessible and (crucially) because all members of a peer group could travel together. Although walking in theory would also offer these benefits, in general, young Londoners do not consider walking a viable mode of transport for any but the shortest of journeys. In part, as outlined in the last chapter, bus travel had become the default mode of transport for most young people, as suggested by these accounts of bus travel for journeys which, young people themselves suggest, could in theory be walking journeys:

Sometimes if I’m really lazy or just can’t be bothered. If I’m walking and I’m past a, I’m next to the bus stop and the bus is making its way up I just jump on. But if not, I just keep walking, I can’t be bothered to wait.

Isl, 16 years

[I]f I’m like going to meet a friend or whatever and I’m, or I’m going to Romford, but I could walk to Romford in about ten minutes. But if there’s a bus I’ll get on it because it’s quicker and easier.

Hav, 15 years

With buses generally accessible, available and free, walking, as a main mode of transport, had become something of a ‘last resort’ for many situations, something done because you ‘had to’, implying that there were no other more attractive options available:

My school is awkward . . . we have to walk

you’re at the top of the hill . . .

no buses actually go right to the school . . . and then you’ve got to walk through the parks to get to the other bus stop

Sut, 16 years (emphasis added)

By extension, when asked directly, many participants suggested that if the free travel concession were to be removed it is likely that many short journeys would be done wholly on foot once more:

And you said earlier if you had to pay the fare you might change your journey? What do you think you would do different, differently?

Well it’s pretty simple for me because I have only like on the bus is only one minute, it’s pretty simple, it’s not really a long walk at all, only it’s for me personally I find that literally almost no effect at all, however now thinking about other people and their situations that’s going to cause some effect but.

So you would just walk it if you had to pay it?

Yeah I’d walk it, yeah.

Hav, 14 years

[I]f you didn’t have the free bus travel, how would you get to school do you think?

I’d have to walk, I’d probably walk. It’s, it’s about, but I’d have to leave much earlier because it’s about a half an hour walk, 5 minute bus journey. I’ll take the bus any day.

Hav, 15 years

This is not to say that the zip card has eliminated all walking trips. In some instances, despite free bus travel, young people reported that they would opt to walk rather than take the bus:

[I] walk [to school], because I live nearby so I feel a bit, I’d feel a bit stupid getting the bus . . .

I used to quite often, during the summer I’d walk home from school. Even though it is a good 50 minute, hour long walk, but sometimes it’s just easier than waiting for the bus and then getting all crammed on it.

Sut, 15–16 years

As these instances suggest, however, opting to walk would usually be in response to perceived conditions (the crowdedness of the bus or the ‘stupidity’ of getting a bus short distance) rather than a proactive decision in light of the health (and other) benefits of travelling by foot. In this respect, it would appear that for the most part secondary school pupils are inclined to persevere with trying conditions before opting to walk:

And my friend who lives in between Epsom and Cheam said it got to a point where for two weeks, every day, the bus . . . was too full to just stop for her. So in the end she just had to leave her house half an hour earlier and walk.

Sut, 14–18 years

The absence of cycling in accounts of transport

Cycling is notably absent from accounts of travelling and travel preferences, in particular as a candidate mode of transport. When asked directly whether or not they would consider cycling to school, most said no, offering a number of reasons why this was an inappropriate choice: main roads were too dangerous (several recounted injuries from cycling), or at least considered by parents as too dangerous; cycling was unpleasant in poor weather; bicycles were at risk of theft from schools and bicycles that were owned were in any case (in most accounts) broken. More significantly perhaps, cycling rarely figured as a possibility for consideration as a method of transport, being (for many) inherently ‘ridiculous’. As one young woman said of those who did cycle to school: ‘I just find it funny. I don’t know why I just, I find that I do find cycling funny’ (H&F, 12–17 years).

The exceptions were telling. Given the focus on sociability in young people’s accounts of transport, cycles were notable for offering more individual, even confidential, travel. One young man, one of the few to cycle to school, also preferred to use his bike on occasions when he did not want to encounter peers: one was going shopping for his mother ‘Because I don’t like getting on the bus with shopping . . . [unless] I have one of them expensive bags’ (YS, 14–15 years). In general cycling was associated with childhood:

I used to cycle every Sunday with my dad and younger brother, but now we don’t anymore. Yeah, just, we kind of stopped, yeah.

H&F, 15 years

It continued to be attractive to some as a leisure activity, and boys in particular talked of riding BMX bikes, ‘for, like, just going to the skate park and using it there’ (Hav, 12 years). Here, cycling was a sociable activity for peer groups which, like opportunistic bus travel, was a possibility for offsetting boredom:

We’re like sitting in the estate and talking and stuff, and it just gets boring after a while, so sometimes we just ride, ride our bikes and cruise

H&F, 15 years

Free bus travel does not, then, seem to play a major role in decisions about choosing to cycle, as cycling is not primarily considered a candidate mode of transport for most journeys.

Free bus travel generates additional activity

The replacement of short walking trips (one to three bus stops) by bus trips did not necessarily reduce the amount of physical exercise young people were getting, or even the amount of ‘active transport’. For some, having the pass had clearly generated additional walking stages of a journey or whole trips that would either have not been conducted without the free bus pass or would have been carried out as a car passenger. There was first some evidence (if at the margins) that a free bus pass enabled access to journeys that would not otherwise have been taken, providing possibilities for exercise:

If I didn’t have free travel . . . I wouldn’t be going places I would be probably staying quite local and through using free travel it means I can go places that I’ve always wanted to go, maybe heard of from friends and family because they’ve gone on trips.

Sut, 15–18 years

By the same token, other journeys might be undertaken less often if free bus travel was not available. As one focus group participant put it when asked how journeys would change without free bus travel, ‘I don’t think anyone would really go out as much to be sociable’ (Sut, 15–16 years). She goes on:

[S]ometimes when I go out with my friends I get three buses there and three buses back, depending on where I’m going, and I wouldn’t pay that much to spend three hours out, because you think about it, if they started making us pay, that’s like, at the moment it’s what, £2.00? £2.20 for [a] single bus fare. You times that by six that’s, you’re going to end up paying a lot of money for just going out with your mates for three hours. You’re already trying to save money doing stuff that doesn’t cost us. [You don’t want to be paying for] getting there as well.

Sut, 15–16 years

Free travel, in that it facilitated travel with friends, particularly to new places, therefore encouraged extra trips, and trips further afield than would otherwise have been made. Such ‘exploratory’ trips inevitably entailed some physical exercise (even if minimal) either as an end in itself or as a by-product:

Me and my friend tend to just get on the bus and go somewhere and then just get off and get the bus back . . . We saw a park once on a bus and we were like, that’s nice, and got off there for a while.

YS, 12 years

If I didn’t have free travel I would have to be doing everything through walking and stuff but I wouldn’t, that means that I wouldn’t be going places I would be probably staying quite local and through using free travel it means I can go places that I’ve always wanted to go [ . . . ] and then you can explore places on your own, you can see what you want to see, not with the guidance of someone who’s obliging you to go and see a museum.

Sut, 15 years

Preferences for active travel

A preference for using buses because they are convenient, sociable, free and everyone can therefore use them was not necessarily a preference for bus travel per se. Indeed, inherent dislikes of being on buses, particularly crowded buses, were common, with concerns about crowds, dirt, germs and the lack of hygiene commonly articulated:

And the thought of the germs travelling in the air. And the fact that that swine flu has just hit as well, so it puts you off public transport

Isl, 15–16 years

It was very packed on the bus, it was quite hot and sweaty and yeah, there were a lot of people crowded into little spaces and buggies and crying babies and everything, it was horrible. [W]hen I’m going to school it’s normally quite quieter because there aren’t as many people, but today because I was going into Romford, I had all the work people and everyone and it was oh, horrible. . . . I feel claustrophobic.

Hav, 15 years

These dislikes meant that more active modes such as walking were widely seen as ‘healthier’ in principle. Young people often spoke of their own preferences for bus travel as being ‘lazy’, for instance, and, when reflecting on the zip card scheme, many recognised the potential health limiting effects of possibly replacing walking with bus travel:

What do you think . . . about why you guys get free bus travel?

I think it could be because some people are lazy, tired, if they’re tired they won’t go to school. So then the government try and encourage them to go in, and they’ve got free travel. You don’t have to use your legs that much to get in there, so.

But then wouldn’t that be against the whole, defeat the point of the whole government thing, the government fitness thing? Because if they’re trying to encourage people to get fit, why encourage them to take the bus then?

True.

Isl, 15–16 years

Although, as detailed above, these considerations did not generally translate into personal preferences for what were perceived as ‘healthier’ modes of travel, some individuals did report preferences for walking because of the intrinsic pleasure of exercise:

[W]hen I didn’t have my Oyster card when I walked to school I felt, because like I was in the air, fresh air, I had fresh air and I could exercise a bit, walk my legs, when I was at school I felt much, much more concentrated and much more alive. When I’m on the bus it’s like, 50% of the time I would slouch on the table, sleeping, even if I’m standing up on the bus, whether I’m sitting down or standing up it just like this musty atmosphere on the bus, it smells a bit and it’s just dull really. So outside, when you’re outside walking it just makes you feel more alive and it does, it makes you feel a bit healthier even though it’s such a short walk.

Hav, 14 years

The exceptions, those who did explicitly choose to walk where possible rather than get the bus, attributed this preference to health aims:

I want to be, like, fit and everything, so exercise wise, I make sure I don’t get the bus too much, so it helps

Hav, 17 years

For the majority, though, ‘fitness’ as a health goal did not feature as a factor in decision-making around transport decisions, and was not something prioritised in stories of travelling around the city. Rather, the well-being considerations of young people were more closely aligned to social well-being: ensuring a journey that maximised social inclusion (in travelling with peers, as described in Chapter 3) and minimised certain risks (which we turn to in Chapter 6).

Conclusion

There is credible evidence that although the intervention increased the use of bus travel for short trips, in the context of rising general use of bus travel in the population, that this did not necessarily mean that young people were significantly reducing their amount of active travel overall. The apparently paradoxical finding from travel diary data that, although the number of trips by walking as a main mode reduced, there was no significant reduction in distance walked, was explained to some extent by the qualitative data. This suggested that the bus pass facilitated more journeys and, given that all bus journeys require some walking, this entailed undertaking some active travel as part of a trip. This additional activity may well have served to offset the loss of trips by walking as a main mode. Although the bus pass may have had only a marginal role in generating exercise as an end in itself (e.g. to access sports facilities), exercise as a by-product of public transport was more common. ‘Bus hopping’, although generating a large number of very short trips which may not be recorded in travel diaries, also generates activity, and simply being on the bus was not necessarily sedentary. The oppositional, and at times explicit assumption (e.g. Mota and colleagues105 and Van Dyck and colleagues106) that urban bus travel is a purely ‘passive’ mode in contrast to the ‘active’ modes of walking and cycling may underestimate activity.

On cycling, levels in young people declined, from a very low base. The use of adults as a comparator group here is perhaps less helpful for attributing change to the intervention, given the rise in adult cycling for other reasons (including a number of schemes to increase cycling levels). Although there is little evidence from the qualitative data that cycling is considered a candidate mode of transport for many young people, we do not know how far this has changed since the introduction of free bus travel. It is clear that young Londoners have not benefited from the increases in cycling seen in adults.

What is encouraging from a public health perspective is that health is a consideration for many young people in thinking about their transport behaviour, and one which means that at times ‘active’ modes are preferred. However, the health outcomes that are implicitly prioritised in the majority of young people’s accounts of travel are those of social well-being. The scheme may (as young people themselves note) reduce the incentives for ‘healthier’ active travel, but it also provides the conditions for travel that is both sociable and inclusive: travel that is inherently ‘healthier’ from the perspective of young people themselves.

In summary, there was no evidence from the travel diary data that the intervention had reduced, overall, the amount of active travel that young people did. The qualitative data suggested that treating bus travel as a non-active mode of transport may be inappropriate if the aim is to use active travel as a proxy for physical exercise, as bus travel generates both additional trips (particularly discretionary ones) and some activity within the transport system itself.

Introduction

The sustainability of transport policies is a key issue for public health. 20 The Organisation for Economic Co-operation and Development (OECD) Vancouver conference107 identified the growth in motorised private transport as a severe challenge for developing sustainable transport systems. Mass transit systems, in general, are less environmentally damaging, and subsidising bus travel has long been seen as a promising policy for reducing private car use, with evaluations of concessionary schemes for older passengers suggesting some displacement of car trips. 108 One explicit aim of the intervention was to foster sustainable transport, both by displacing private car journeys in the short term, and by inculcating more sustainable travel habits in the longer term. This chapter first assesses the evidence for short-term change using travel diary estimates for change in young people’s car travel. Evidence for longer-term change is more difficult to assess, given the relatively short time frame of this evaluation. We therefore use the qualitative data to explore young people’s views of driving and public transport, to provide some insight into these longer-term implications.

Quantitative evidence on impact on use of private motor cars

The changes to trips by car as a main mode, distance travelled by car and percentage of short trips by car are shown graphically in Figure 9. [Line joining solid squares shows change in target age group (12–17 years); line joining solid circles shows change in control age group (25–59 years). Diamond (right-hand side of figure) shows the change in the target age group relative to the change in the control age group. 95% CIs are shown as vertical lines. Horizontal dotted line indicates a ratio of 1 (i.e. no relative change).]

FIGURE 9.

Changes in car travel, pre- to post-introduction of free bus travel (a) Car trips per day; (b) car distance per day (km); and (c) per cent of short trips by car.

Qualitative evidence on impact on current car travel

The free bus pass had displaced some walking trips (see Chapter 4), making the bus the default option for what might have been short walking trips. For longer trips in the city, however, walking was rarely considered to be a viable option even if there were cost or other disincentives to alternative modes. In the absence of free bus travel, hypothetically, young people thought they would typically forgo the journey, pay for the journey, or persuade parents to provide lifts, depending on the need. As described in Chapter 4, young people in outer London in particular said that if they had to pay for the bus, some trips would be made by asking parents for lifts:

If I was casually going out somewhere I probably wouldn’t have got the bus, but if I needed to go . . . I’d probably pay for it. But usually I’d catch cadge a lift . . . parents always seem to be conveniently free . . .

Sut, 14–18 years

Again, the importance of universal free travel was key to explaining the displacement of some car journeys, given the importance of sociability to travel planning. Unlike bus travel, parental lifts might not be easy to co-ordinate with others for planning travel and socialising:

[Free bus travel] makes things easier because if it was, for example if I was getting a lift with parents [ . . . ] they’re going to be, ‘oh I can only give you a lift at this time’

Sut, 15 years

My mum or dad would drive me if I want them to but it’s like I said you meet friends on the bus and things like that.

Sut, 13–16 years

To some extent, free bus use had therefore replaced some private car use, particularly for the kinds of discretionary journeys for which lifts from parents would have been a potential option for longer trips, and a first port of call for many. As these discretionary and longer trips may be more likely to be outside term time and weekdays, the travel diary data may have underestimated the impact of the scheme on displacement of car travel.

Driving remains seductive in prospect, if not reality

Current preferences for bus travel with friends over some lifts from parents did not, though, necessarily translate into imagined future preferences. Indeed, the majority of young people expressed a desire to drive when older, with initial comments about driving apparently reflective of beliefs about the status and value of driving widely reported in other research. 109 Future car use – for almost all young people – was still either a normal expectation or an aspiration, and older friends with cars were already a valued resource:

. . . I’ll get a car.

Yeah.

Loads of people now, you know some of them that’s old and that, I’ve got friends now though that are 17 they’ve got cars already, and they’ve actually got a license, so you’d probably go with them or whatever.

Yeah . . .

Get lifts with them?

Yeah.

If I haven’t got my own car or something, I’ll call my friend up . . .

Sut, 15–16 years

Car driving was for many (particularly young men) a rite of passage that was eagerly awaited as a route to enhanced status with peers and potential partners, in contrast to still using an ‘Oyster’, which would mark you out as somehow deficient, as these boys jokingly agree:

As soon as I’m old enough to drive I’m going to drive

I’m waiting for the day that I get my hazards

I’m going to get it before all of you, ha ha! . . .

. . . when I come home from parties I’m cold . . . so cold, so I always imagine I have a car just come . . .

Yeah, a car you could drive!

. . . also it’s more impressive to girls if you’ve got a car, if you’re 20 and you haven’t got a car and you

just do public transport

can’t go out to meet girls and that . . .

If you’re still rocking an Oyster!

[laughter]

YS, 14–15 years

However, to some extent stated desires to drive were reflective not just of the anticipated status attached to car travel, but also of its significance for achieving independence. Here, for instance, a disagreement between two friends reflects their differences in how ‘independence’ is constituted in relation to learning to drive:

What is it that makes you want to learn to drive?

I want to be free, I want to be independent and anyway, driving’s fun isn’t it?

Yeah, so what will be more independent about it?

I’ll be able to do things myself, be able to go where I want. [ . . . ] I’d just be more independent because you see a lot of working people always in their cars.

How about you, are you planning to drive or?

Well we all want to be start driving now but I don’t want to. I want to do it at a time when I’m able to afford lessons for myself, so I can buy myself a car because I don’t really, again I think it’s independence, I want to be able to pay for myself rather than depend on my mum.

YS, 17 years

Some young people in inner London did challenge their peers’ accounts of the future benefits of driving in London. One young man noted the practical problems of ‘long’ traffic, and being stuck ‘in the High Road for 40 minutes in a hot car’, before countering his friend’s rationale for wanting a car when older:

Do you know how many girls you can get with cars bruv? You just honk at them . . .

I think driving in London is stupid, I don’t see the point . . . I like cars, but in London, no.

Isl, 15 years

Such practical objections to the rewards anticipated from driving were rare, but were unsurprisingly more common for the few participants who were already driving, for whom the realities of parking, congestion and the implications of not being able to drink had already tarnished any romanticised views of driving bringing independent travel or freedom. One, who now regularly drove to school since passing a test recently, noted that parking was expensive, and buses (as well as still being free) could be more convenient for many journeys (Sut, 15–18 years). Additionally, she and a friend note that buses are intrinsically more interesting:

Yeah I do like getting the bus because you can meet new people on the bus and because everyone’s doing the same thing and it’s every day it’s nice to see the same people and it’s quite a nice commute so it gives you a lot of independence . . . every single bus journey is always interesting because something would always go on and you could listen to other people’s conversations if you want to and brighten up your day a bit.

I love the bus, I like journeys with my friends, obviously bus is good because if you want to have a drink then you don’t, you can use the bus to get home instead of driving. And always, you can always have funny conversations with people on the bus when you’re on the way home from a club or something

Sut, 15–18 years

What is notable about this exchange is not that these young women, at an age where they are just learning to drive, prefer the bus (indeed they note that driving would be preferable to waiting in the rain for a bus) but that the bus is not only ‘normalised’ as a mode of transport, but even attractive. There is no sense in this account that using a bus is stigmatised as a young person’s mode of transport, one primarily for poor people or only for those unable to drive. For those old enough to have actually experienced the advantages and disadvantages of different travel modes in the city, driving is (they admit) still a novelty attached to adult identities, but not intrinsically more socially valued than the bus.

The normalisation of bus travel?

Demonstrating an absence is difficult, but what was striking across the data was perhaps the lack of any comments about the inherent low status of bus travel. Although buses were often unpopular for being dirty, crowded or potentially risky (see Chapter 6), because all young people used them, there was no stigma attached to using the bus. Neither was there any sense that they were, as young people, relegated to public transport. Indeed, bus travel was, in the accounts of many young Londoners, simply the way in which many Londoners got to work and other destinations. This exchange gives a flavour of the accounts of bus journeys, with the range of other passengers who might be encountered on a typical journey:

I’ve had many a conversation with older people, not so much like 30 to 50 year olds they don’t, they keep to themselves . . . You can see mothers chatting to other mothers from their primary schools and stuff

Those are the workers who are so miserable that they just stand there and then especially when a bus is packed they say like, so rude and they get in your face and they’re just like, why are you standing in my way? . . .

And then the school kids

And then school, well children yeah

And then, but the good thing, sometimes the good thing about having old people in the bus is that you get that moral side out of you because when they come on the bus and you’re sitting down in the seat you feel like oh because they’re old you should give them your seat. So you feel good when they seat down because they normally say thank you

Sut, 15–18 years

It is not that free bus travel in itself created this normalisation. Two other conditions were necessary. First, the fact that universal free travel meant that bus travel had become a default, taken for granted mode of transport for all young people, meant that bus travel was ‘normal’ for all their peers, whatever other modes of transport were financially or otherwise available to them (see Chapter 3).

Second, was the context of enhanced bus provision, which had increased the modal share of bus travel in general in the capital (see Appendix 1). This generated routinely experienced encounters with a range of other Londoners, and visitors, on the buses (including older citizens, commuters, mothers with young children), making bus travel very visibly a ‘normal’ way for all kinds of people to travel. Without the relative accessibility and availability of buses in London, free travel, even if universal, would not become a preferred mode for so many journeys. A few, particularly those who had experience of other places and who lived more centrally, where bus services were in general more frequent, noted this explicitly:

The good thing about London is the amount of buses we have because when you go to other places you’re waiting half an hour for a bus scheduled to come at this time. And here you can just wait two minutes, you’re on a bus, you know what I mean

Isl, 15 years

For most, however, the taken for granted accessibility, and ubiquity, of bus travel as a normal way for a range of travellers to get around was merely an unremarkable backdrop to their accounts of travel in London.

Conclusion

In the short term, it is difficult to assess how far this intervention has changed the levels of private car use in London. A range of other policy interventions (notably the congestion charge) have reduced the advantages of car travel, and the number of trips by car has declined for all Londoners. Additionally, as the majority of young people’s car journeys are those made by adults as well, the reduction in young people’s car travel is included in that of adults.

In outer London, the qualitative data suggest that the free bus pass has displaced some car journeys, reflecting young people’s preferences for travel independent of parents and which allows sociable travel with peers. In the longer term, it is difficult to assess how far the intervention may have shifted perspective on the desirability of driving. However, although most young people still express positive views of learning to drive, what is notable is that bus travel has become normalised for this group. The broader context of London’s changing transport system is an important condition of the effects of free travel. In a transport system in which bus travel is in general more available and more reliable than in other parts of the country, it carries little of the stigma attached to bus travel in other parts of the UK, where bus use is disproportionately a mode used by the young, older citizens and poorer households. 110,111

In summary, a range of policies contributed to the reduction in car travel by young people in London in term-time weekdays, and to the normalisation of bus travel as a non-stigmatised alternative to driving.

Introduction

Transport policies have the potential to impact on safety and inequalities in safety through a number of pathways. First, different modes of transport incur different risks of RTI for users and others,44 so policies that shift mode distributions are likely to change the numbers of RTIs. However, mode share may also affect relative risks (as in the example of ‘critical mass’ effects for walking and cycling, see Jacobsen112), so there may be feedback loops which reduce these effects over time. In Chapters 3–5, we noted a rise in bus use and a decline in other (currently more risky) modes, which could reduce the number of injuries young people experience on the road. Second, perceptions of risk may change as transport modes become more or less acceptable or well used. Third, as the overall number of trips or distance travelled goes up, young people may be more exposed to both road injury, and other risks, such as assault.

This chapter assesses the impact of the intervention on the risk of RTI and risk of assaults by comparing changes over time in young people with those in adults. Given that a range of other factors also contributes to declines in road injury, and in assaults, we also compare what happened in London with the rest of England. We then draw on the qualitative data to provide evidence on young people’s views of the risks of transport mode use in the context of free bus travel.

Quantitative evidence on road safety

As outlined in Chapter 2, we had hypothesised that the introduction of free bus travel for young people in London would result in an overall reduction in RTIs in young people relative to the control age group. Trends in RTIs and assaults by year are shown in Figures 10–13.

FIGURE 10.

Annual rates of RTI in London.

FIGURE 11.

Annual rates of pedestrian RTI in London.

FIGURE 12.

Annual rates of cyclist RTI in London.

FIGURE 13.

Annual rates of car occupant RTI in London.

Quantitative evidence on assaults

We hypothesised that the introduction of free bus travel would result in an increase in injuries to young people due to assaults as they travel further and more frequently.

Assaults in London

Using HES we found that the number of young people hospitalised with injuries inflicted by assaults increased from 2321 in the pre-intervention period to 3322 in the post-intervention period. The number of adults hospitalised with injuries inflicted by assaults also increased over the study period, from 11,829 to 14,641 admissions.

The rate of hospitalisation for injuries in young people inflicted by assaults increased over the study period, from 1.13 hospital admissions per 1000 person-years to 1.61 admissions per 1000 person-years. This change represents an increase of 43% (95% CI 35% to 51%) in the admission rate. In the control age group, there was a smaller increase from 0.77 admissions per 1000 person-years to 0.91 admissions per 1000 person-years, a relative increase of 19% (95% CI 16% to 22%).

Our measure of relative change therefore suggests that free bus travel was associated with a greater increase in rates of assault in young people (relative change 1.20, 95% CI 1.13 to 1.27). However, when examining trends in hospital admissions for assaults by year, as shown in Figure 14, it appears that the largest increase in assaults among young people occurred between 2002 and 2005 (i.e. predominantly before the introduction of the free bus travel scheme).

FIGURE 14.

Annual rates of hospital admission due to assaults in London.

Assaults nationally

Hospitalisations from assaults increased among young people aged 12–17 years living outside London over the study period from 1.04 per 1000 person-years to 1.12 admissions per 1000 person-years, an increase of 8% (95% CI 5% to 10%) as shown in Figure 15. Compared with young people living outside London, the free bus travel intervention was associated with a greater increase in rates of assault in young people in London (relative change 1.32, 95% CI 1.25 to 1.40).

FIGURE 15.

Annual rates of hospital admission due to assaults in London compared with outside London.

Free travel as a facilitator of risk management

Free bus travel was reported to have had an important role in allowing young people to ‘practise’ skills necessary to negotiate London’s bus network. Such skills were sometimes acquired by making ‘rehearsal’ journeys: journeys which, importantly, could be undertaken without any economic disincentive. For example, one girl described how when she first got her zip card she tried out ‘getting on the bus three times just to see how it worked – I just went round the corner three times and got off’ (YS, 12 years). Other young people described learning by initially making sure they travelled with someone who already knew the route:

Yeah, I know [my trip today to central London is quite far], but I know how to go somewhere when someone’s like shown me. I brought my dad up here and I followed Annie and my brother this morning.

Hav, 14 years

I feel so much more confident going by myself everywhere now . . . [but before] if I wanted to go to say Oxford Street to return something I’d be like to my friend, I’d be, ‘oh do you want to come shopping with me’ rather than just go and return it. But now I just go by myself, and I’ll be fine with it.

YS, 17 years

Thus, for many young people, the company of friends or family enhanced their sense of security on novel journeys that involved travelling further afield than their familiar, local area. Being able to do this without financial cost facilitated these rehearsals. For many, the fact that buses were free was also to some extent a safety net, preventing one being ‘stranded’ and providing a contingency plan if things went wrong:

When I came here to London I didn’t have free bus travel by that time [ . . . and] it actually limited me and didn’t allow me to go to places that I would actually go to when I had the free bus travel. For example, when you go out because since I still knew I can get lost easily, you know? If you have to pay for the bus it’s going to limit you from getting back.

YS, 14 years

Say if I go out, and it’s getting late, or if my original journey, say if the train’s cancelled, I know I can just get a bus. I’ve got it free, I can go a different route. Whereas if, say if I’d already paid for a train ticket, and then I was halfway through the journey and then the train stopped, and then I didn’t have any money, then I couldn’t go another route. So it’s like really important, I think, yeah, so security.

Hav, 17 years

The bus as a site of risk

If travelling for free helped young people develop their skills in independent travel, the buses themselves were to some extent also sites of risk. There were gender differences in how risk was reported. For boys, in particular, the most potentially harmful threat on buses is older teenagers: ‘like 18 years old, 16 years old’ (Hav, 14 years). Many talked in general about the types of incident which could ‘kick off’ on a bus, with arguments generated if disrespectful looks were perceived, or you were on a bus in the ‘wrong’ part of town:

Also buses are really easy place to get attacked, in our area there’s some gangs, and some of them live in different parts, and one day when I was on the bus one of the boys who lives in another part of town was on the bus, but in the wrong part, and a guy came on the bus with a bottle and started hitting him saying, why are you here?

YS, 14–15 years

Yeah, it’s one of those . . . things because say if you’re a girl and then you’re actually pretty good looking, the guys [will be] catcalling and stuff. Or if you’re a guy, yeah, and then you’re one of those alpha male guys, and then they’ll end up, you’re going to end up having a stare off or something. They’ll be, and it’s usually called screwing or whatever, looking at them funky, like neh, and then they’ll just be like, oh I’m watching you and then just start stuff. And then it’s just not really safe . . .

Isl, 15–16 years

For all secondary school-aged young people, whatever gender and wherever they lived in London, the most commonly reported risks were those of encountering those from other schools or neighbourhoods on the buses:

I’ve got people from other schools that stare at you because of what uniform you wear . . . and if you get on the bus that ain’t your school

It just gets awkward . . .

I don’t know, you just get glared at. And it’s more likely to trigger an argument, rather than getting on a bus with your school

Isl, 15–16 years

Typically, direct experiences recounted were of routine aggressive interactions between other teenagers. These were risky in that they had the potential for generating more serious incidents, but assaults were rarely reported as directly witnessed or experienced. Very serious assault incidents were usually only recounted as ‘moral tales’ that happened to (often unknown) others, and used in group discussions to illustrate the kinds of risks that one should be aware of, and which were ‘common knowledge’:

It’s actually been on the news that people were actually glaring at each other, a couple of people were actually glaring at each other and a fight broke out and someone actually died from it . . .

Isl, 15–16 years

Extreme stories relating to gun and knife crime, much covered in the media, were told as ‘the kind of thing that happens’ and a perceived risk, but were rarely part of the direct experiences of the young Londoners in this study. It is perhaps telling that in group interviews, such stories often had particular functions, such as to illustrate (often jokingly, as in the extract below) the dangers of neighbouring areas. Not quite urban myth, but certainly not a routine expectation of most bus travel, the limited direct visibility of knife and gun crime for most young people was illustrated by the ways in which stories were challenged by peers as atypical of bus travel in general, as in this brief exchange from a group discussion, which moved quickly from disbelief from two participants to joking aside about the neighbourhood of one participant:

I’ve seen a lot of dodgy goings-on on buses though.

Yeah.

Yes.

People get out guns.

What, whoa, sorry?

Qu’est-ce que c’est?

That was around Meadwell Bridge though.

What are you trying to say about Croydon?

It’s a sprawling urban mass of despair!

Sut, 14–18 years

For girls, fear of ‘perverts’ (H&F, 15 years) was a commonly mentioned concern, with stories of being followed by men from the bus reported across the boroughs. Again, despite the common knowledge of this as a risk of bus travel, few actual incidents of assault were reported. One exception was a girl who recounted being sexually assaulted (by other school children) during a bus journey, an assault which was reported to the police [(borough deleted), 15–16 years]. However, for many girls, buses were recounted as more secure than other modes of travel, particularly walking, after dark:

If it’s late at night time, probably around five or six . . . then depending on how it is around the estate and how I’m feeling I might jump on a bus, just to get past the estates. Yeah, just for the sake of safety.

Isl, 15–16 years

Boys also discussed avoiding certain areas, particularly after dark, but were more likely (particularly in one-to-one interviews) to suggest that buses were a riskier form of transport at night:

. . . but at night around [area], the main town is quite, yeah you, because you get all the people going to the clubs and stuff and it’s, it goes mental . . . because you get club people and gangs and stuff . . . I try [t]o be at home before it gets really dark too . . . [A]nd when you do get on the buses at night, because I’ve been to competitions . . . got back about 9, 10 o’clock at night and had to get on the buses you do find that you find some weirdos on the buses, which isn’t nice.

Hav, 15 years

Yeah, I would go on the bus but I wouldn’t go upstairs on a bus at night because I’ve had a few situations where there’s quite like dangerous people upstairs like . . . So I wouldn’t, I would definitely not go upstairs when it’s dark and I would want to get, say if I want, it is dark and I want to, and I need to get to a place, I would want to get there faster, so I wouldn’t take my time going upstairs and take my time getting downstairs, I’d get off bus straight home, finish.

Hav, 13 years

As these accounts suggest, although concern about potential assaults was common, in the main young people discussed their ability to manage these risks, rather than their vulnerability to them. Strategies for avoiding or mitigating the risks that were routinely encountered in travelling around London included avoiding certain places, or being outside alone after dark, but also adopting a ‘street wise’ deportment, which was essential for avoiding confrontations, especially with other young people. This entailed avoiding eye contact, appearing confident and knowing when to back down: ‘keep yourself to yourself, head down, headphones in’ (Hav, 14 years).

In terms of avoiding potential conflicts, many timed journeys to avoid other school’s leaving times, or covered identifying school uniforms if they were travelling alone. If you had to travel alone, personal stereos were useful for creating a ‘private space’ in which interaction was less likely to happen, and mobile phones provided some widely described risk management strategies: allowing the creation of the ‘social’ through real or pretend phone calls ‘so they do not target you’ (Sut, 15–16 years) or the use of a ‘fake call’ app to make the phone ring. Other strategies included speaking in non-English languages to arrange ‘escapes’ with friends (Isl, 15–16 years) and staying on the lower deck of the bus near the driver. The zip card, on occasion, was mentioned as reducing the level of risk associated with travelling – not just in ensuring that one could afford to get home (as above) but providing a free ‘escape plan’ of ‘just being able to jump on a bus’ to avoid risky situations. In short, although a few participants preferred to avoid buses altogether (or at particular times) because of the risks of aggressive incidents, and most considered the bus as a potential site of such incidents, in general the risk of assault was not a major factor in accounts of bus travel.

Managing risk: a source of pride and entertainment

One reason that these risks, despite causing some anxiety, were not reported as a deterrent to bus travel is that they are also a source of spectacle and entertainment. Dramatic incidents were often discussed humorously in group discussions, and in individual interviews (as above) young people stressed their skills in managing the risks. Encounters that could be unpleasant or risky were simultaneously an attraction of the bus journey, providing a source of potential entertainment and a fund of stories that served to create shared narratives of ‘us and them’. ‘Weirdos’, for instance, were often mentioned as a risk of a bus journey but rather than being necessarily threatening, these were sometimes known eccentric locals, who were part of the community, and, as noted here (following a discussion of the risks of bus use) they were explicitly part of the ‘fun’ of travel:

you do meet a few weirdos on the bus

Yeah

Yeah, that’s part of the fun

Like the dancing guy . . .

. . . He’s a local celebrity

. . . He’s like the wizard man! . . . singing to people on the bus

There’s so many people like that

Sut, 14–18 years

Indeed, in focus groups in particular, telling stories about interesting or unusual encounters on the bus, which may have been unpleasant or even frightening at the time, was related as part of the excitement of travelling, particularly with friends. The key issue of whether incidents were risky or entertaining was whether travelling was alone or with friends. Although choosing a particular mode or route was on occasion presented as a risk assessment strategy (for instance avoiding a park, or an area after dark), a major contributor to whether or not the choice was risky was the availability of peers to travel with, as we outlined in Chapter 3.

Conclusion

Road traffic injuries declined among young people relative to adults after the introduction of the free bus travel scheme. The major contributions to this decline were reductions in car occupant injuries and cyclist injuries. Pedestrian injuries in young people declined at a similar rate to adults. These findings are consistent with the changes in travel patterns reported in Chapters 4 and 5, which suggested a relative decline in distance travelled by car (see Chapter 5), a relative decline in distance cycled (see Chapter 4) and no change in distance walked among young people (see Chapter 4).

Quantitative evidence indicates a relative increase in assaults among young people in London compared with both adults in London and young people living outside London. However, there is some suggestion that much of the increase in assaults in young people in London occurred before the introduction of free bus travel in 2005.

For young people, the bus was a frequent site of conflict between different groups (schools, gangs, older teenagers), and a place where (reportedly, but rarely in their own experience) more serious incidents happened. However, bus travel was also recognised (particularly by girls) as safer than other modes of transport (especially walking) and most young people had a range of strategies at their disposal for managing the routine risks encountered. At the margins, free travel enabled ‘escape’ strategies, from both the worries of being lost, and (on occasion) the ability to just jump on a bus to avoid a troubling situation.

In summary, there is no strong evidence that the free bus scheme contributed to an increase in assaults in young people, and some evidence that the scheme is associated with a decrease in RTIs, reflecting the transport mode changes following its introduction.

Introduction

Social inclusion is fundamental to well-being, and transport access is increasingly recognised as important to social inclusion in settings such as London. 63,80,113 Differential access to transport is one of the ways in which health inequalities between people and places are generated,5 and age is one social factor that influences the risk of ‘transport exclusion’. In the UK, for instance, the Social Exclusion Unit (p. 2)114 cited transport-related problems as restricting young people’s capacity to take up education or training opportunities. Young people’s exclusion from participation has been variously conceptualised as arising from immobility,115,116 disempowerment98,117 or dependency on adults for transport. 67,118,119 An explicit policy aim of the free bus travel scheme was to increase access to education, training and other opportunities for young people – aims which have significant potential to enhance future public health and address one contributor to inequalities in health.

However, measures of ‘social inclusion’ are problematic, in that this is a multidimensional concept that is inherently difficult to operationalise. 80 We had no direct measures of changes in dimensions of ‘inclusion’ which might be important, such as ‘participation’ or ‘access to social networks’. Furthermore, outcomes which may be causally related to inclusion, such as changes in participation in training or education, are likely to be casually distant from an intervention such as free travel. We therefore used the travel diary data to provide some evidence on one proxy measure: how the number of trips made by young people to work or school changed relative to adults after the intervention. Additionally, we looked specifically at distances travelled by level of deprivation after the intervention to see if there was any evidence that impact of the scheme changed social gradients in this measure of participation.

The qualitative data analysis provided evidence on the salience of transport exclusion in young people’s current travel decisions, and the ways in which the free bus travel scheme was used in fostering independent travel and access to a range of goods and services that are essential to social inclusion.

Quantitative evidence on participation and inequalities in participation

In the absence of direct markers of social inclusion, we analysed data on the frequency of journeys by all modes during the week, and on the socioeconomic gradient in distance travelled by all modes by purpose of journey.

Distances travelled by level of deprivation

We also examined whether or not there had been any change in the distances travelled by young people by level of deprivation and purpose of journey before and after the intervention. Figure 16 shows trends in distances travelled by young people for education, entertainment, personal business and shopping pre- and post-intervention by decile of IMD. The figures generally show no clear evidence of diminution in socioeconomic gradients following the introduction of the free bus travel scheme for young people. Nevertheless, the patterns were suggestive of a flattening in the gradient for travel for shopping (at an overall slightly lower level post-intervention; p-value for change of slope with IMD was p = 0.024).

FIGURE 16.

Distances travelled by young people according to journey purpose by decile of IMD, pre- and post-introduction of the free bus travel scheme (a) Education; (b) entertainment; (c) personal business; and (d) shopping.

Zip and social inclusion

Qualitative evidence suggested that universal free bus travel for young people removes one barrier to social inclusion: that of transport costs. The intervention was reported to have facilitated participation through making some journeys financially accessible, at least at the margins. Although most young people were not explicit about the ways in which this happened, there were a few comments on how free bus travel offered opportunities for access to sport and leisure:

The local sports centre near me is, we’ve got to get a bus to get to it. So my brothers do that, and my mum takes my sister because they have like that little baby club thing there. So if a bus, the price went up, my mum wouldn’t take my sister to the little clubs where she can meet other little kids. And my brothers probably wouldn’t go to the gym at all.

Sut, 15–16 years

For some, free travel did make a reported difference to family budgets, not only for those in the poorest groups, but also for those whose families nevertheless struggled:

My mum’s lost her job and stuff, so it’s difficult for her. [For me to take the bus] doesn’t cost my mum anything, and it just helps her out.

H&F, 12 years

I don’t think that’s fair because personally in my family, which you guys wont repeat, my dad’s only bringing in one income and my mum’s not working because she’s looking after [siblings] at home. And she’s got me and Callum to look after as well. So because my dad’s bringing in probably, I don’t know, probably about 1000 over the limit that it’s supposed to be, or however much over the limit, I’m not entitled to get EMA, free school meals, and if it was, free travel. But we’re still struggling so it’s not fair how it is.

Isl, 15–16 years

Notably, the instances of increased opportunity of access recounted were often group-based activities, with the intervention enabling families to more easily afford to go on outings:

When I was younger because my mum was pregnant at the time . . . me and my dad used to go up London because it was free for me . . . We used to go the Science Museum and things like that . . . so it was quite fun.

Sut, 13–16 years

To some extent our data generation method (with most young people interviewed in small groups) perhaps discouraged disclosures of the financial impact of free travel for less well-off young people. The young person above, for instance, prefaced her account of the difficulties their family would have without the fare concession with a plea to other interview participants not to repeat her circumstances outside the group. However, the fundamental impact of free bus travel on social inclusion was evident in the taken-for-granted nature of inclusion implicit in young people’s responses. Across the data set, at every age, in outer and inner London, young people’s accounts suggested their independent access to both local and more distant destinations by bus was an absolutely routine expectation: a normal and unquestioned entitlement.

I just get two buses to school, and on the weekends same, I just get the bus anywhere. Like sometimes it can be far like the West End, or not, it could just be like [local high street] or something.

Sut, 15–16 years (emphasis added)

In this respect, transport-related exclusion due to financial factors was notably absent as a salient concept for the participants in this research. Indeed the taken-for-granted nature of being able to afford to get anywhere is perhaps indicated by the rather extreme response of one participant (echoed less succinctly by others), who told us that if she could not use the buses for free she ‘wouldn’t come to school’ (H&F, 15 years). Across the boroughs, young people emphasised the ease of getting around, and indeed the range of sites that might be visited by bus:

I normally go [by bus] just to go to school, go see family.

Probably the same thing, [I use the bus to] go to school, see family. Probably get the bus to football match, Regent’s Park, Camden.

I get the bus to school and my cousin’s house.

I get the bus to school and other activities. I go to music classes . . .

I normally get the bus everywhere,

Isl, 12–13 years

[T]hat’s [fare exemption] really helpful, whenever I really need to go anywhere it’s just, it’s no hassle

Sut, 14–18 years

The importance of the intervention in facilitating the ability to travel without restriction (‘no hassle’) is perhaps evident in the fact that, for most young people, the only accounts of limitations in mobility came from reports about those who had no zip card to prove their entitlement:

[T]he day I was robbed I lost my Oyster. I had a missing [glasses] lens . . . buttons ripped off my shirt and a bruise on my face. And then I tell him [the bus driver] I don’t have my Oyster, I got robbed, and he’s like ‘I’ve heard all these excuses . . .’ and he was actually swearing at me . . . and then he kicked me off

Isl, 15–16 years

Another participant had had his right to free travel rescinded by TfL (as part of the ‘Behaviour Code’ penalties introduced in relation to some travel concessions for young people120):

[W]hen I didn’t have [free bus travel] I did struggle in terms of not getting everything done because I didn’t have that freedom to get on a bus

H&F,12–17 years

Similarly, one young man noted of four friends who had their passes taken away: ‘It puts a strain on their social activities because they can’t go out as much’ (Hav, 15 years).

Social inclusion: the importance of transport access for discretionary journeys

As one young man put it: ‘with my free travel I always explore places that I didn’t even know were real’ (Sut, 15 years). He gave an example of a trip to a distant part of London with friends to a vintage clothes market they had heard about in a popular TV programme. As several participants explained, this role of free bus travel was particularly important for such social or recreational trips because these were not ‘necessary’. Such trips were therefore most likely to be abandoned if bus travel were not free:

[Without free bus travel] I think that under-sixteens would only go to places that they thought were really necessary like school [ . . . but] I think they would stop going to places like the park. [ . . . ] If it were a park that was far away I would go there less often if I had to pay.

YS, 15 years

Such discretionary trips were highly valued, providing opportunities for young people to travel together, and to participate in social life, and to travel independently of parents. This ‘independence’ was not framed solely in terms of travelling without parents, but also as travel that did not entail reliance on parents to either transport or the costs for transport:

[Free bus travel] is good, it’s really useful. It gives, at this age especially, it gives us more independence to do what we want, especially on buses [ . . . because] if it was too expensive we’d probably end up getting our parents to drive us everywhere which would be a real problem, so.

I think at this age it’s really important to have that because we need to learn about the world or London now sort of thing, and how to travel by ourselves.

YS, 12–13 years

Social inclusion and independence

Linked to this sense of independence is the equality of opportunity that free travel had provided all young people in London to experience the city, and develop their growing sense of the world and their place in it, providing, as one put it, ‘A sense of freedom’ (Sut, 16 years). New journeys provided opportunities for young people to test their skills in planning and managing the unexpected. For the work experience students included in the study, the trip to the LSHTM was a good example. When asked, they reported a variety of methods for preparing for what was a ‘new’ journey: using Google Maps, TfL journey planner and ‘common sense’ to find new routes.

For all participants, access to free travel meant young people were better able to explore London, and develop an understanding of the city’s geography without fear of ending up somewhere with limited means to return to known parts of the city:

When we [my friends and I] was in London we just saw a bus that was going to, we was by Trafalgar Square . . . and we saw the buses that, is going towards Oxford Street, didn’t know exactly where it was going . . . and, and we get on it, we’re lost, see, see where we end up . . .

[ . . . I]f you didn’t have the bus pass, like would you, would you have gone up to Trafalgar Square if you didn’t have it, for example?

Yeah. We, we would’ve walked, but it’d probably taken us like an hour so we might not have got round to Oxford Street probably [ . . . ] The amount of time it takes.

Hav, 14 years

This is not to say that prior to the introduction of free bus travel young Londoners only made ‘necessary’ trips or did not experience themselves as becoming more independent. Indeed, several participants said that it would not affect their travel behaviour if they had to pay. Others, however, described the zip card as making a real difference at an age when you may have a parental permission to travel by yourself, but prefer the company of peers than travel ‘alone’:

The good thing about free travel is that when you’re old enough to be able to get places your own, your parents don’t take you, [ . . . you’re still] not going to be employed because you’re in fulltime education. And you might actually be at that stage where your parents give you a certain amount of money or no money and you have to use that money to get, to do what you want to do. And then you’re old enough to want to go to the cinema every weekend and see your friends and just get out of the house and if you can’t afford that then it is a little, it’s annoying.

Sut, 18 years

These benefits for independent mobility did not just apply to those with limited family finances. Rather it was raised by participants from a range of social backgrounds, as well as by both genders and by participants from across London. The benefit arose not simply from the removal of financial barriers, but from access that did not require parents to fund it. Importantly, even families with better resources might be unwilling to subsidise discretionary journeys: ‘That’s the only trouble, you don’t, it’s not essential that you get travel and you get to where you want’ (YS, 14 years).

Fostering ‘belonging’: zip cards and citizenship

Through its fostering of wider-ranging travel, the intervention contributed, for some, to a sense of ‘belonging’ to London as a community and of ‘being a Londoner’. For young people, often aware that their concession was unusual to their city, this sense of belonging was at times explicitly framed as having an effect on well-being through fostering pride:

It [the zip card scheme] . . . makes you feel proud [to be a Londoner] because you’re at the front of everyone, because you’re the ones who have brought in these new schemes that are working and making your life easier . . .

And also you have this mutual understanding of [being . . .] a Londoner, you’re the same as me now . . . And there’s . . . this sense of community in this huge, huge [city.]

Sut, 15–18 years

There was an awareness that this was a benefit not enjoyed by peers outside the capital, with one focus group participant describing how her ‘cousin [who] lives really far away . . . just wishes she could have more buses and the free travel . . . to get around more’ (Hav, 14–15 years). Beyond the pride in having a scheme that was innovative, and marked their unique city, the zip card enabled participants to ‘know’ a larger geographical area than would otherwise be the case. We have already outlined in earlier chapters the ways in which young people made excursions to the city centre and other parts of the capital. Many of the young study participants, in particular from the outer London boroughs, would recount exploratory bus journeys they had conducted ‘up London’ or to ‘the West End’, and those from all boroughs talked about the ways in which bus travel had opened up other parts of the city:

Mostly every Saturday we’ll probably just jump on a bus because we have a free [pass] and go anywhere, and get another bus from there, and another one. And we just travel, we don’t know where we’re going . . . [once] we ended up near Hammersmith, and near the West End

Isl, 16 years

I think if you get the bus a lot you can try and, like, vary it [the routes you take] up so you get to know London

Isl, 12–13 years

Concessionary bus travel, that is, affords young people a topographical engagement with their urban surroundings which enhances their familiarity with the city by rendering them ‘more aware of where you’re going, how to get to places’ (F, YS, 14–15 years), and a number of young people explicitly welcome the opportunities they had to not only ‘get to know’ or ‘learn about’ London by travelling widely in it, but also feel that they ‘belonged’ as citizens. As one young person put it:

I like it [having the zip card] because you feel kind of unique . . . and it’s only in London. [Y]ou can travel around London because you’re a kind of a Londoner, but other people can’t.

Sut, 17 years

Social exclusion: young people with disabilities

Although financial barriers did not exclude young Londoners from travel in the city, this does not mean that the transport network was universally accessible. Young people with disabilities faced a number of problems in using buses, and transport exclusion was a key issue for them. For those using wheelchairs, limited space on the buses meant that only one could be accommodated at a time. This made it impossible to travel with friends if they were also using wheelchairs. Given the importance of travelling together, and for the journey to be part of the outing, outlined in earlier chapters, this is a real limit on social inclusion for young people.

Across three interviews with young people with disabilities, including six young people, only one person said they preferred the bus, because of the challenges of finding somewhere to park (even with a blue badge). The young people with disabilities used Freedom Passes, rather than zip cards, to access public transport (see Appendix 1). Notably, in contrast to the zip cardholders, they did not report any bus journeys as entertaining outings in themselves. Also in contrast with their more able bodied peers, who talked of bus travel as ‘no hassle’, and could recount a number of positive aspects of bus journeys, for young people with disabilities, buses were generally experienced specifically as ‘hassle’, and a barrier to participation, rather than a route for enacting participation. Journeys themselves were liable to be inefficient, as the wait for a bus with room to board might be long, and at times embarrassing, given the frequent problems with the wheelchair ramp:

Some ramps don’t tend to work, so that’s a bit of a hassle . . . sometimes it’s dangerous with an electric chair, it’s heavy . . . I just think the bus driver should check the ramp is working

O, > 16 years

In addition to the practical problems encountered (e.g. difficulties in seeing the visual destination display from the wheelchair space; not being able to see outside on older buses, with their high windows), some had experienced a lack of consideration from other passengers and drivers. One example came from a young man with epilepsy, who recounted how when he had had a seizure on the bus:

. . . people completely ignored my pleas for help. The driver drove on as if nothing was happening

O, > 16 years

Free bus travel was not, then, a sufficient condition for transport inclusion in the absence of accessible transport.

Conclusion

The quantitative evidence suggests some increase in travel for work or education among young people in London relative to adults. The scheme also appears to have influenced the travel of young people for education, entertainment and personal business in a relatively similar way across areas of deprivation.

The qualitative analysis suggests that financial barriers to transport were not an issue for young people in London. As we have no historical data to compare, it is impossible to attribute this specifically to free bus travel, but it seems likely that free and – crucially – universal free bus travel, in the context of a reasonably accessible transport system, meant that for most young people, lack of ability to pay did not restrict their mobility or their opportunities to be independently mobile in their neighbourhoods and in the wider city. The exceptions (those with no zip card; those with disabilities) support this interpretation.

Although social inclusion was explicitly tied in the policy aims of this scheme to access to education and training, our qualitative data suggests that a more salient aspect of the scheme for many was the enhanced access it provided to discretionary journeys. These were important journeys in terms of enabling participation in social life for young people, but parents may be less willing to pay for discretionary travel. In summary, there is some evidence that the free travel scheme promoted access to education and training for young people, and good evidence from the qualitative data that it fostered social inclusion and independent travel for those young Londoners able to access the bus network.

Introduction

One policy concern was that older citizens, in particular, would be displaced from the bus network if numbers of young people using the buses increased, either because of overcrowding or through ‘fear based exclusion’. 17,18 Older citizens also receive free bus travel, and in London a ‘Freedom Pass’, paid for by the London boroughs. Until April 2010, older people with their principal residence in London were eligible for a pass at the age of 60 years entitling them to free travel on bus, London underground and tram services within the capital, without time restrictions, and rail services at off-peak times. Since then the qualifying age has been increasing in a graduated way, and will become 66 years by 6 October 2020. As older citizens are also vulnerable to transport exclusion, with a consequent impact on their well-being,121,122 one potential negative effect of young people’s increased access to bus travel might be a reduction in their ability to use buses.

We used the travel diary data to explore whether there was any evidence that older people were in general using the bus less, or were specifically displaced from the network at times when young people were more likely to be using it (i.e. during peak school journeys times). We also explored older citizens’ accounts of using the bus to explore how far the presence of young passengers affected their uptake and experiences of bus services.

Quantitative evidence on impact on older people’s bus use

Older citizens’ accounts of sharing the bus network with young people

Some older people voiced a degree of disapproval of free travel for young people. This was based on a perceived ‘unfairness’: the Freedom Pass benefit was associated with a ‘reward’ for working hard all their lives, something young people had not yet done. 123 There was also a sense of temporal unfairness, with older citizens noting that they had not benefited from free travel when they were young themselves, or when their children were young. Some also expressed a dislike of young people ‘abusing’ the privilege, by jumping on and off the bus needlessly:

Look how long we waited to get our Freedom Pass. Then these youngsters got it straight-away haven’t they?

Sut, 80–84 years

Well I used to have to walk to school. They’re not now, they get on for two bus stops.

Yeah.

And what are two stops? . . .

Should use their legs.

H&F, 75–84 years

However, on reflection, many considered that free travel for young people was a socially useful benefit, with advantages to young people and their families:

But what if they’re young people who want to maybe go and see their old granny at the weekend or go and do her shopping in the evening?

Well

Yeah, it’s a difficult one really.

Debatable, isn’t it?

Yeah.

But basically it’s supposed to be in education, isn’t it?

Isl, < 65–80 years

It helps the parents especially now that things are a bit tight. I do believe they should.

H&F, 70–74 years

Older people were, then, not particularly positive about the free fare concession for younger people in general terms. What was notable, however, was not the range of opinions on whether or not young people should have free bus use, but that this disapproval of the zip card scheme was never couched in terms of young people displacing them from the buses. This is not to say that there were no complaints about behaviour on the buses: indeed, some older citizens did report finding younger passengers on the bus network intimidating, and a few reported having changed their behaviour in response to the noise, crowds or jostling of school children on the buses:

Well you would have near enough got a seat straight away. Now you don’t. They’re so busy shouting and drinking and, because they all bring these bottles of stuff with them and talking to each other about what they’re going to do and on their mobiles as well, that they just don’t take any notice of you. And they’re, they’re in the way as I say, these great big pack bags, you can’t get past them at all. It does spoil your journey really.

Sut, 80–84 years

I don’t go out when the children are coming out of school.

H&F, 74–85 years

I think that most older people like me, over 60s, think the same as I do, that they’re gonna wait for the 9.30 because the rush oh, the rush hour’s over by then. And they can go whenever they wanna go

Isl, 70–74 years

However, more typically, there was a general tolerance of the behaviour of young people, and even in instances of (as above) ‘discomfort-based’ displacement, these were reported to result in changed, or less pleasant, rather than abandoned, journeys. There were very few reports of fear-based exclusion from transport. Despite many not enjoying travelling on buses used by school children, most pointed out that the noise and jostling was what one might expect from excited young people. Crucially, there was general agreement that this did not put them off using the services.

So it’s not something you think it puts you off using the bus because there’s children . . . ?

Oh no, no, no, no

No, no.

No, no, never, we never think you know sometime you just, most of the time it happens so we are at that time about three o’clock, between two and three is when they come out, four they come in. But it doesn’t matter it’s a day . . .

It doesn’t bother us really us no, no.

It doesn’t bother us.

H&F, 70–79 years

Probably we done it when we were kids but we don’t notice it you know

It’s high jinks isn’t it? It’s high jinks more than, we’ve never been involved in anything any time on the bus where there’s been

Trouble

You know a stabbing or something like that or somebody’s got beaten up. They seem to interact with one another, boisterous but not vicious.

Isl, 70–74 years

Indeed many older citizens, from all boroughs, reported general civility from younger people, who would offer seats and allow them on first:

Yeah, I must say, another thing, following what M2 said, I find the young people tremendously

Helpful and kind.

Willing to get up and give you their seat.

H&F, > 90 years

Because you’re standing at the bus stop with hundreds, it seems like hundreds of girls screaming around you, the bus pulls up, they are polite, they do at least let you go first.

Sut, 70–74 years

Young kids you’d think wouldn’t let you off, but in fact I generally find, they may not be terribly nice to each other but to old, what they regard as old people they’re quite good I have to say. Another example we were riding a bus quite late when we, 7.30, Ann and I were, and there were some young lads kicking the bus shelter. We didn’t think this was the best place to be. When the bus came in they stood back and said, oh no, after you. You were first . . .

Sut, 60–69 years

In part, the lack of reported displacement or explicit conflict with school children resulted from a normative assumption that school-aged children and older people used buses at different times and in different ways. Specifically, apart from avoiding school journey times, older people reported rarely using the upstairs of double decked buses, or the back of single deck buses:

Well I can’t say I have been on buses that often when schoolchildren, when youngsters have, but occasionally I do. And yeah, sometimes they’re a bit noisy and so on but then I would probably go downstairs if they’re all rushing upstairs.

Sut, 60–69 years

And would you ever go upstairs?

Oh no

No?

No. [Laughter]

Is that because of?

It’s not because of the children it’s because I can’t be bothered walking up the stairs

I just don’t bother about going up, if I had to go upstairs we’d have to but we never ever go up there we’d rather stand

Isl, 70–74 years

The reluctance to use the upstairs of the bus reflected a consensus about where and when was a reasonable and comfortable time and way to travel, with older people avoiding the school run times, and the places on the bus where young people preferred to sit (upstairs, or downstairs at the back). These preferences were, as in the quote above, rarely attributed to intimidation, but rather to decisions about maximising comfort and convenience. They also, of course, became part of a culture of expectations about where and how to travel, with ‘taken for granted’ nature of the division of space and time according to age on the buses evident in the fact that breaches of the norm were often a source of amusement:

If I’m out at the weekend with my friend, if we’re going say to Kingston, we’ll go upstairs on the bus. [Laughter] We behave like kids.

Sut, 70–74 years

I will pick my, grandchildren up, sometimes . . . And they always want to go upstairs [general laughter].

Isl, 65–89 years

Data from young people reiterated these normative expectations about the arrangements of seating on the bus (what one young person called ‘the segregation’):

[T]he reason why I’d go on the top deck [is] because sometimes the bottom’s really congested yet there’s so many seats available on the upper deck . . .

And that’s why . . . most children are on the upper deck, more seats for the elderly and adults down on the lower deck . . .

[Y]ou’re doing it so there’s more space at the bottom deck for the elderly and that.

Sut, 13–16 years

Older citizens’ choices of when to travel were, then, largely rationalised as made for comfort and convenience reasons, and the expectations of where different passengers preferred to sit were in general shared by younger and older passengers. There is little sense of ‘exclusion’ based on fear. Furthermore, if ‘discomfort-based’ exclusion from certain journeys, such as school run times, was an issue, this was as likely to be generated by the desire to avoid crowds of commuters as young people.

I would avoid the peak times, if I can

[noises of agreement]

And, um, in the morning when everybody’s rushing to work. I mean we’ve been there [we’ve worked], we know what it’s going to be like.

Yeah, yeah [laughing]

And, um, coming home times if you can avoid them

And I, yes

And I think, and I, I do try to avoid the school bus.

[General agreement and laughter]

And you see loads of us have got all day, we, we, we know the times to go, you know

When the school get out.

Isl, 65–89 years

Indeed on occasion, children were overtly preferred as fellow passengers:

Yeah, yeah, between the office mob and [the young people] . . . if I had to choose who to travel with I’d choose the school boys . . . I used to avoid, um, travelling early in the morning, because obviously people are going to work and, and they’re all sort of focused on going to work, so they, you know, there’s no give-and-take then. And I would avoid school runs, obviously, and then the same in the evening. But now, in between, you’ve got these mums, as well . . .

Isl, 70–74 years

Avoiding rush hour commuter and school travel meant that older people were more likely to be in conflict over space and seats on the bus with other adults, and specifically (in the accounts we have of displacement) parents with young children, particularly those with large pushchairs. Indeed, the final comment in the quote above is typical in that the most likely conflicts over the space on buses reported were around the space reserved primarily for wheelchairs or pushchairs, which was typically near the more accessible seats at the rear door, simply because those using them tended to travel at the same times of day as older travellers and be ‘competing’ for the same restricted space. These conflicts were often reported with vigour and irritation:

And do you know that bus driver . . . came up to me, and asked me to get out of my seat and give it up for the lady with the buggy. He wouldn’t move [the bus] off until I got up off that seat . . . I got off the bus because I was like [gestures angrily with clenched palms]

Isl, 65–89 years

This is in contrast to the tenor of reports of conflicts with younger passengers, which had a sense of routine disapproval, but rarely anger. Indeed, rather than being a source of comment on fear or risk, stories of encounters with younger travellers, particularly children, were often used in interviews as a way of demonstrating strategies for dealing with potential difficulties. Older passengers were clear that they could ‘manage’ conflicts, and were not intimidated by the crowded and on occasion confrontational bustle of a busy city. This woman, for instance, talking of how difficult it can be to board a bus in the mornings, goes on to say that she personally has few such problems:

But I can get to the front of the queue quite easily.

Laughter

OK, what tricks do you use to get to the?

Shopping bag. Well you have, you have to. But I notice we all use the same techniques and the bus drivers are very good. If they see an elderly woman there and all these children they come right where the elderly person’s standing.

Sut, 70–74 years

This ability to ‘manage’ and the strategies used to negotiate access to their rightful place in the queue or a seat on the bus meant few older participants reported limitations on their ability to travel that resulted from other passengers. Typically participants ended comments about rowdiness or unpleasantness from young people with a comment asserting that they (if not necessarily other people) were able to cope:

Yeah, yeah young people it doesn’t really stop you going anywhere or whatever

We go wherever we want don’t we?

Yeah that’s right.

It doesn’t stop us.

Isl, 70–74 years

This is not so say that managing such navigation and the potentially difficult interactions with other bus users was necessarily pleasurable. Some participants did report limited, or decreasing, confidence in using the bus service, and talked of avoiding certain routes or times of day. 124 Indeed, a number talked of the challenges of access as physical impairments began to limit ability to stand, or balance on a moving bus. However, the presence of larger numbers of young people were not reported as a significant cause of anything more than unpleasantness, and even stories of challenges had some positive implications for older citizens’ self-esteem simply because they provided opportunities to both exercise, and then account for, valued agency in the world, as a competent, coping and even ‘hardy’ traveller. The presence of young people on the buses might be a reason for avoiding the school run if possible, but so was the presence of large numbers of other travellers. It was never given as a reason for avoiding buses.

Conclusion

We found no quantitative evidence that older passengers had been ‘displaced’ as a result of the introduction of free bus travel for young people, and our qualitative data provided further evidence that displacement due to overcrowding or fear was not a major issue. Our qualitative data may have underestimated the impact of young people on older people’s travel behaviour simply because we are less likely to have included older people who did not travel, and we therefore may have missed examples of fear-based exclusion. However, the older citizens who participated came from a range of backgrounds and settings, and were in general articulate on their complaints about other passengers, and about young people’s entitlement to free travel. What was notable was the lack of any overt complaints about young people making them unwilling to travel, in comparison perhaps to the more open commentary on the unpleasantness of conflicts with other passengers. This is largely because older citizens and zip card users are not competing for the same ‘space’: older citizens preferred different seats on the bus, and tended to avoid times of day when young people (and commuters) are using the buses anyway.

In summary, there is no evidence that free bus travel for young people had an impact on older people’s travel behaviour or, therefore, the travel-related determinants of their well-being.

Aim and perspective

This chapter describes the methodology and results of an economic evaluation of the free bus travel scheme. The aim of the economic evaluation was to determine whether or not the intervention ‘free bus fares for 12–17 year olds’ represents value for money if we take public health effects into account, by comparing two alternative scenarios in terms of their costs and outcomes. These two alternative scenarios are:

• Scenario 1 – do something – the bus network is free for 12- to 17-year-olds.

• Scenario 2 – do nothing – the bus network is not free for 12- to 17-year-olds.

Before free bus travel was introduced in 2005, secondary school-aged children paid a reduced flat fare for journeys on the London bus network. Therefore, we assume in scenario 2 that the target group, 12- to 17-year-olds, are charged half the adult fare. Although scenario 1 can be observed in recent years since the intervention, scenario 2 requires that a counterfactual be created; describing the state of the world if the intervention could be removed. This is a standard requirement in ex post evaluation.

In terms of determining the impact of the policy the research has utilised the methodology of developing counterfactuals to determine what would have happened in the absence of the policy (scenario 2). The core counterfactual has been developed based on a comparison with what happened with the age group 25- to 59-year-olds in London over the same time period. This has the advantage that this age group would have been influenced by all other interventions put in place in London over the same time frame (e.g. road user charging), but were assumed not affected by the free bus fares policy for 12- to 17-year-olds. The implication is that where a statistically significant difference between the groups is observed this difference can be attributed to the intervention. Where data were available additional counterfactuals have been calculated for certain outcomes.

A number of potential outcomes of this intervention were identified by the study and reported in previous chapters. Those that are included in the economic evaluation are described in Table 6. A cost–benefit analysis (CBA) framework was developed to allow a range of different impacts of the intervention to be included and so to determine whether or not the intervention represents good value for money. Using well-established literatures on the monetary values of key inputs we compared the net monetised benefits of the intervention with the costs of implementation.

There are some limitations associated with the use of CBA as an organising framework, which are addressed in Limitations. Nevertheless, CBA is a useful tool for combining outcomes across a number of government departments/areas of public service. 125

Time frame and evaluation framework

The outcomes and costs in Evaluation framework results have been calculated for a representative year (2009) and for the intervention applied to the target age group (12- to 17-year-olds). An annual appraisal is feasible because the costs arise on a recurring, annual basis. To use any longer appraisal period would introduce unnecessary speculation about the duration of the policy. The evaluation framework used is shown in Table 7. In the following sections the individual elements of this table are described in detail. The report summarises the results of the monetisation of the outcomes in Evaluation framework results, and then discusses limitations and implications for future research in Limitations.

Safety

The study hypothesised that there will be a change in casualties as a result of the introduction of free bus passes for 12- to 17-year-olds (see Chapter 6). This section uses data from Chapter 6 to estimate the benefits from changes in casualties.

User benefits

The study has hypothesised that there will be user benefits of the scheme, namely a welfare gain for individuals who currently travel for free and previously had to pay and an increase in accessibility (see Chapter 7), as a result of removing bus fares for 12- to 17-year-olds. This section uses data from TfL to estimate those user benefits.

Methodology

The methodology that will be used to determine the change in user benefits involves calculating the change in consumer surplus as a measure of welfare change, using the user benefits approach described in DfT. 127 The user benefit measure is designed to capture the benefits from an accessibility gain as measured by the generalised cost (GC) function including accessibility gains from reductions in cost, reductions in travel time, improvements in journey quality and benefits to any new users who travel more due to the reduction in GC.

Measuring GC from i to j will usually follow the structure below:

⁽¹⁾ GC i j = Fare i j + Time Cost i j + Other Disutility i j

where:

• ‘Fare’ is the money cost of a trip (e.g. the bus fare)

• ‘Time Cost’ is an amount calculated using ‘values of time’ which represent the inconvenience of time spent travelling

• ‘Other Disutility’ is a term which includes the individual’s valuation of any other trip characteristics which are relevant (e.g. inconvenience of a high floor to a bus user).

The relationship between GC and user benefits is shown schematically in Figure 17. Before a scheme is introduced the GC to the user is GC₀ and at this cost level T₀ trips are made. If, for example, the fare cost is removed (as is the case in this scheme) this reduces the GC by (GC₀ – GC₁). The result of this reduction in GC per trip is an increase in trips made from T₀ to T₁. The user benefit resulting from the introduction of the scheme equates to shaded area A for current users and shaded area B for new users (increased travel or accessibility). The demand curve is shown as a straight-line which is consistent with a common assumption in transport appraisal known as the ‘rule of a half’ which is a rule of thumb used to interpolate between two known points when the rest of the demand curve is unknown.

FIGURE 17.

New user and current users’ benefits from the introduction/enhancement of a scheme that reduces GCs.

If all the required data points are known then the value of both of these areas can be calculated and the impact on accessibility determined. The calculation to determine the user benefits for areas A and B is completed using the following formula:

⁽²⁾ User Benefits ( A + B ) = 0.5 × ( GC 0 − GC 1 ) ( T 0 + T 1 )

Data were collected from TfL on the number of free journeys in the 12–17 years age group, in the year 2009. There were 247,297,000 free journeys (T₁). The single bus fare in 2009 was £1 (if paid using an Oyster Card). 128 [It should be noted that the data used to populate the number of trips for 12- to 17-year-olds in 2009 were taken from data collected by TfL through zip card use. This is a different data set from that used in Chapter 3, which drew on travel diary data relating to term time (Monday to Friday) only. The TfL data set was not available for the ‘before’ period of the intervention, prior to data collection via zip card use.] It is assumed for this analysis that young people would have been charged half this fare (50p) in the absence of the intervention. Therefore the difference between GC₀ and GC₁ is £0.50. The key input that needs to be determined to allow the user benefits to be calculated is T₀. This is the trips that would have happened in the absence of the introduction of free fares (the counterfactual).

Two counterfactuals were calculated to assess what would have happened over the same time frame in the absence of the intervention:

• Counterfactual 2a: assumes that expected bus travel is based on a statistical analysis of what happened in the same period to adults in the 25- to 59-year-old age group in London.

• Counterfactual 2c: assumes that expected bus travel is based on demand elasticities and trip generation factor (TGF).

Counterfactual 2a assumes that in the absence of the intervention the pre–post change in bus use would be similar to the pre–post change in bus use in adults in the 25- to 59-year-old age group. Data from the LATS were used to determine the change and are reported in Chapter 3. It should be noted that the LATS data only included data from term-time weekday trips. It is therefore likely that this has underestimated the change in trips for 12- to 17-year-olds, if an increase in trips (outside the journey to school) takes place at weekends and out of term time. The pre–post ratio of adults trips was 1.36 (95% CI of 1.25 to 1.46). This indicates that over the same period bus trips for adults increased by 36% (95% CI 25% to 46%). The number of bus trips made by the 12- to 17-year-old age group was not collected in the pre-period. Instead, the ‘before’ trips have been derived from the LATS and are provided in Table 13. The resulting change in bus trips for the 12- to 17-year-old age group is 22 million within a 95% CI of between 5 million and 40 million bus journeys.

TABLE 13 - Counterfactual 2a: expected bus journeys based on a statistical analysis of what happened in the same period to adults in the 25- to 59-year-old age group in London

Outcome	12- to 17-year-old trips in the pre-period (A)	12- to 17-year-old trips in the post-period (B) (2009)	Counterfactual: pre–post ratio of adult trips (C) (95% CI)	Expected 12- to 17-year-old trips under counterfactual (A × C)	Change in 12- to 17-year-old trips attributable to the intervention: observed – expected (B – (A × C))
Bus trips	165,754,820	247,497,000	1.36 (1.25 to 1.46)	225,426,555 (207,193,525 to 242,002,037)	22,070,445 (5,494,963 to 40,303,475)

Counterfactual 2c uses the data reported in Balcombe and colleagues74 to estimate the change in trips associated with the price elasticities of demand and implied trip generation rates from the change from a half-fare charge per trip to a zero charge per trip. (Note this scenario does not use LATS data results from Chapter 3. It is based on what is expected to happen when there is a price change using price elasticities.) Figure 18 shows the demand curve given the price levels. P_A’ is the full fare that an adult would pay, P_A is the half-fare that a young person would pay. P_B is the zero fare. Balcombe and colleagues74 provide data on expected changes in trips as a result of changing from a half-fare to a zero fare and the difference in elasticity of demand associated with a child compared with an adult.

FIGURE 18.

Trip generation factor.

For a reduction in child fares from a flat (or half) fare P_A to free fare P_B, TGF = B/A. Given TGFs for the reduction from full fare to flat fare, and from full fare to free fare, the TGF of interest can be deduced:

⁽³⁾ T G F F u l l F a r e → F l a t F a r e = A A ′

⁽⁴⁾ T G F F u l l F a r e → F r e e F a r e = B A ′

⁽⁵⁾ ⇒ T G F F l a t F a r e → F r e e F a r e = B A

Given values of 1.33 and 1.60 for (3) and (4) in general demand, we infer (5) = 1.20,74 TGF = B/A for child = 1.20. This should be considered a broad-brush estimate. Therefore, it should be expected that a reduction from half-fare to free fare for a child would generate 20% more trips. It would imply that the whole change in number of 12- to 17-year-old bus trips could be attributed to the intervention. Under counterfactual 2c, T₁ is 247,497,000 and T₀ would be 206,247,500, with the increase in trips due to reducing from half-fare to zero being 41,249,500.

Decongestion benefits

The study hypothesised that one of the benefits to society from the introduction of free bus travel would be a reduction in congestion, as there would be a switch away from using the car. Consequently, there would be a reduction in the negative impacts of congestion [e.g. air pollution, noise, carbon dioxide (CO₂) emissions, etc.] which have been identified by the literature as having an impact on public health. 20 This section assesses the benefits from decongestion of the free bus travel scheme.

Active travel and physical activity

The study hypothesised that there would be a change in levels of active travel (walking and cycling) as a result of the intervention of free bus fares for young people (see Chapter 4) which may lead to a change in physical activity. This section examines whether or not we can estimate the benefits from a change in physical activity as a result of the scheme.

Crime

The study hypothesised that there would be a change in assaults as a result of the intervention (see Chapter 6). This section uses data from TfL and Chapter 6 to estimate the benefits from changes in assaults.

Revenue

One of the consequences of the intervention is that those 12- to 17-year-olds who would have previously paid half-fare in 2009 are now travelling for free. This reduces the revenue that would be received by TfL. The methodology used to calculate the change in revenue is described next.

Administrative costs

The introduction of free bus fares resulted in additional administrative costs for TfL. One element of this is the issuing of travel cards to allow 12- to 17-year-olds to be able to travel for free. (Since 2011 this policy has been changed and 12- to 17-year-olds are now charged £10 per card to cover these costs. This was not the case in 2009 and has not been included in the CBA.)

Bus operating costs

It was hypothesised that the introduction of free bus fares to the 12- to 17-year-old age group would lead to additional marginal operating costs for the operator. This section provides the methodology for calculating these.

Evaluation framework results

The framework initially set up in Table 7 can now be populated with the costs and benefit results, so giving the net present value (NPV) and benefit : cost ratio (BCR) of the intervention (see Tables 25 and 27). The results for the counterfactual where all the inputs are based on a comparison with the change in 25- to 59-year-olds (counterfactual 2a) are shown in Table 25. The benefits from the intervention to society as a whole are found to be greater than the costs incurred: at £13.8M the NPV is substantially greater than zero. Moreover, the value for money offered by the intervention appears to be high. DfT140 provides a value for money-rating scheme for transport interventions for BCR (Table 26) – based on this the intervention exhibits high value for money and would have a high chance of funding (BCR between 2 and 5).

Alongside the main results using counterfactual 2a, a set of sensitivity tests were conducted to examine how changes in the assumptions would alter the results, and to assess the robustness of the results (Table 27). These include alternative counterfactuals and additional outcomes. To interpret the table, numbers in italics should be treated as alternatives to the entries in the first three columns, or as additional where there is no entry in the first three columns.

Two alternative counterfactuals are shown. The first substitutes the base case safety outcome values with the counterfactual based on a comparison with 12- to 17-year-olds outside London. This leads to a substantial increase in NPV and BCR of the project (BCR with a mean of 6.6), and although this is judged to be a less likely outcome than the base case result (given that it is not based on a London setting), it does add to confidence that the true BCR is > 2.

The second alternative counterfactual uses an estimate of changes in trips based on demand elasticities, rather than using the 25- to 59-year-old age group as a control. This alters the user benefits, revenue and bus operating costs. The results again imply a high value BCR (3.5), which is encouraging from a value for money perspective.

The second set of tests deal with additional outcomes. These were carried out to explore the implications of including costs and benefits which had been excluded from the main results (see Table 25) because of limited confidence in validity or attributing causality.

• The study identified an increase in assaults, although there remains a question over causality and hence the usefulness of the results for policy analysis. Including the estimated disbenefits associated with assaults would marginally reduce the BCR (to 3.2) but not threaten the high value for money rating.

• Decongestion benefits, including reduced delay, CO₂ emissions, noise and local air pollution, would add further to overall benefits of the policy. An additional £165,000 (range £119,000–212,000) would be added to the NPV, and the expected BCR would remain at 3.6.

• Finally, an attempt was made to estimate the potential marginal capacity costs to bus operators by comparison with the Freedom Pass scheme for older and disabled users. The results were found to be highly uncertain; nevertheless, if they are included, they act to reduce the NPV to –£23.5M and the BCR to 0.4. Hence, the finding is that the results of the evaluation are sensitive to the treatment of bus operator costs.

In all bar one of these sensitivity tests the analysis has resulted in a positive ‘high value’ BCR, indicating that the intervention (given the outcomes considered) has represented good value for money taking into account the impact on public health. It also implies that there would have to be significant negative outcomes not already represented in the framework to result in a negative NPV and BCR < 1. Connecting this result with the qualitative findings it can be identified that there were additional positive benefits such as fostering social inclusion (see Chapter 7), which if presented alongside have the potential to make the results stronger in favour of the intervention.

Limitations

This study has sought to develop an economic evaluation of the public health impacts of the intervention, focusing on the costs and benefits for 1 year (2009) following the implementation of the scheme in 2005. The evaluation is based on the quantitative research within this study plus available secondary data and valuation evidence. The evaluation is wide-ranging compared with most transport policy evaluations; however, it does have some limitations which are worthy of discussion.

One limitation is that the study has focused on available data for the immediate ex post period 2006–9, rather than seeking to forecast into the future the longer-term health impacts that the intervention might bring. More research is needed in both public health and economics to translate an increased used of bus travel into longer-term public health impacts, particularly with respect to outcomes such as physical activity, which are expected to have most of their benefits in the future.

Similarly, the long-term impact on life chances and independent living, as a result of increased mobility between the ages of 12 and 17 years, has not been possible to quantify and hence to include in the economic evaluation. There are some hints of it in the qualitative assessment. This topic would also benefit from targeted research.

Another limitation that was raised during the research was the impact of using values that were designed to be applied to adults to the context of young people aged 12–17 years. This was highlighted in the context of the WHO130 evidence for changes in physical activity, but is also relevant for the change in safety and intentional assault, whereby the value of statistical life (VOSL) calculated based on an adult has been applied. Further research is required to determine if this value should be different and what values could be applied to children.

Particular challenges in attributing causality arose with assaults. The question is whether or not the policy impact is swamped by background trends (e.g. in youth crime) which are not picked up in the counterfactual, hence they appear as ‘effects of the policy’ when in reality they are not. Particularly in the case of assaults, there is a need for better data linking the hospital episodes to bus travel, or a wider-ranging analysis of assaults in which 12- to 17-year-olds’ mobility is set alongside other factors and trends. Conversely, it is easier accept the findings on mode shift and trip generation based on well-understood relationships in transport demand. In general, the authors take the view that the counterfactuals do serve the intended purpose; however, in the case of assaults there is a case for further investigation. Having said this, the magnitude of the ‘change in assaults’ that was found makes it doubtful whether or not any changes to the analysis of assaults would materially affect the case for the policy.

Wider economic impacts are another difficult area, linked to the increased access to employment and training provided by the policy. A more detailed understanding of trip purposes for 12- to 17-year-olds, starting with data gathering, would be helpful in providing a more concrete assessment of these impacts. Moreover, WebTAG has some limitations for application in economic evaluation of public transport and public health interventions, where there are intended and unintended public health consequences. Further work is needed to resolve this issue.

Finally, although a CBA framework has enabled the range of outcomes considered in the quantitative work to be valued and included in overall NPV and BCR, one of the key weaknesses of a CBA framework is that it excludes those outcomes that cannot be monetised. The qualitative research (see Chapter 7) identified that there were other potential outcomes from the intervention that this approach has not been able to capture. For example, the universal availability of free travel seems to be particularly valued (see Chapter 3), as young people can be confident that any and all peer group members can travel in that way. It was not possible to monetise that effect within the scope of this study, and additional survey work would be require to attempt to capture it. That evidence would then need to be considered alongside the results of the current CBA to provide a more comprehensive description of the intervention. For the moment, it remains necessary for the person using the evaluation results to weigh the CBA against the qualitative and unmonetised impacts.

Conclusions

From the perspective of the CBA framework, there is a body of evidence indicating that the policy has reduced RTI, increased bus travel, and reduced car travel, while not reducing levels of active travel in the city. In the base case the benefits have substantially outweighed the costs, providing what the DfT considers ‘high’ value for money. Sensitivity testing was conducted, and most of these tests do not overturn this conclusion. Only a sensitivity test with substantial additional capacity costs of bus operation threatens the positive NPV of the policy, and there is no firm evidence that this was the case.

Introduction

Many policy and infrastructural interventions, among them educational, housing and transport measures, have had a significant influence on public health despite their not having an explicit public health brief. As public health in the UK moves to a new context, with a base in local authorities, understanding the positive and negative effects of such interventions, and their potential to widen or narrow inequalities in health, becomes increasingly important. This study explores the effect of such an intervention, using qualitative and quantitative methods. It evaluates the public health impact of a transport intervention which was not designed with public health improvement as an aim, but which was likely to have a range of effects, positive and negative, on the immediate and future health of young people (the target population) and the wider population in London. At a time when universal benefits in general and concessionary fares in particular are under scrutiny, evidence on these putative health effects are important in informing decision-making. This study therefore aims to provide evidence on the public health effects that could be attributed to the introduction of free bus travel for young people, and to conduct a cost–benefit evaluation which takes these health effects into consideration.

Main findings

Our study provides empirical evidence on key questions in the debates around whether or not such schemes are worth investment, and their effects on outcomes such as active travel, injury rates, social inclusion and sustainability. We first outline the main findings, in terms of the causal pathways hypothesised in Chapter 2 (see Figure 3).

Assessing the evidence and revising the logic model

We have presented the quantitative and qualitative evidence for the pathways hypothesised as potentially linking the intervention (free bus travel) to outcomes related to the determinants of health. We can now revisit the logic model proposed in Chapter 2, to summarise the impact of free bus travel on public health in a graphic way, using a revised logic model (Figure 19).

FIGURE 19.

Revised model of pathways between intervention and determinants of health.

First, we now clarify the meaning of the intervention. The key elements of the intervention were that it provided free bus travel for all young people in the context of broader policies that improved the bus system in London. What we evaluated was ‘universal free bus travel in the context of a good bus service’. Our findings relate to this, and we have specified the intervention (access to universal free bus travel for young people) and major elements of the context (efficient, accessible, bus network) which are preconditions for the effects we have identified from the intervention.

This general context also had some effects which interacted with the effects of the free bus scheme, such as the increasing use of bus travel for the whole population, contributing to normalising bus travel, so we have added lines indicating this.

The context also has effects that act on the study outcomes of interest, independently of the free bus scheme. One we have included here, as data were generated as part of the research on older people’s use of bus services, is a positive impact on the well-being of older citizens in London. 124

The quantitative elements of this study attempted to control for this context to some extent by identifying specific effects of the intervention on the target population of the scheme itself. Evidence for these links (e.g. for an increase in short trips by bus) is indicated by black lines.

Our revised model indicates where there was no evidence found from the quantitative or qualitative analysis for an impact on public health (displacement of older people from the bus network; increase in assaults), and where there was weak evidence, such as suggestive evidence from the qualitative component only (future dependence on cars). This model is not an exhaustive description of the many pathways that link free bus travel to the public health, or an attempt to evaluate whether the scheme overall has been ‘good’ or ‘bad’. The outcomes on the right of the model are diverse, and adjudicating between them both value-laden and often linked to disciplinary, occupational or lifecycle priorities. For example, physical activity outcomes are prioritised by public health researchers, whereas road engineers tend to prioritise injury outcomes, and young people themselves prioritise social inclusion. We have mapped outcomes that were of interest to stakeholders and were theoretically and plausibly related to determinants of health and present a summary of the available evidence from this study for these.

In summary, this model suggests that where free bus travel is offered to all young people in the context of a good bus transport system, then there will be public health gains in terms of young people’s social inclusion and ability to travel independently; there will be minimal impact on the overall amount of active travel young people do; some indication that dependence on car travel is reduced and no evidence that there is a consequent impact on older citizens’ use of bus transport.

Strengths and weaknesses of the study

Methodological findings: using mixed methods to strengthen internal and external validity

In Figure 19, many of the claims made are reliant on qualitative data. In terms of assessing the credibility of causal claims made, this is problematic from the perspective of current guidance on evaluating natural experiments,81 which is largely concerned with statistical methods for defending non-randomised designs from the threats to confounding discussed in the section above. However, these guidelines do not currently include other ways of strengthening credibility. Our approach was to use a pragmatic, iterative and inductive design which drew on mixed methods, primarily secondary analysis of existing data sets, analysis of primary qualitative interview data and a systematic review. Rather than taking an epidemiological approach of assessing effectiveness as the primary aim of an evaluation, our approach integrated epidemiological questions of effectiveness within a more inductive mapping of ‘what is going on’, utilising the best available methods and designs for each element of the evaluation. 148

Our final study aim was to contribute to methods for strengthening causal inference, and we turn now to our methodological findings on the challenges of integrating different research designs within one evaluation.

A number of models have been proposed for integrating data in mixed-methods studies. 149 We integrated qualitative data with findings from quantitative data sets using an inductive approach, in addition to a thematic content analysis, to enable an analysis not just of what young people said about the intervention, but also how their accounts of travelling in London reflected tacit knowledge and taken-for-granted understandings. This analysis was used in four ways:

• To refine our understanding of the concepts (active travel) and the indicators we have chosen (walking, cycling), and of the intervention itself.

• To assess the credibility of causal attribution through delineating mechanisms by which the intervention may have influenced young people’s behaviour.

• To provide some insight into important health and well-being outcomes that cannot, currently, be measured or monetised.

• To understand the context within which the scheme was provided, to help elucidate the necessary and sufficient conditions for its potential effects.

A major challenge with this iterative approach is that the logic of good epidemiological design, with published protocols specifying the hypotheses and analysis plans, sits uneasily with the logic of a more inductive qualitative design, where good practice suggests developing plans for further sampling, data generation and analysis plans as hypotheses emerge from ongoing analysis. 97 An example from this study is the decision to use a change-on-change analysis using adults as a ‘non-exposed’ control. Although this provides some control for broader changes in the transport system which also influence young people’s travel, ongoing qualitative work suggested some limitations of some of the comparisons we had specified. For example, qualitative interviews shed light on the importance of discretionary travel, and suggested that this had implications for the uptake of free bus travel that might be inadequately captured by available travel diary data (which was restricted to term-time weekdays for our ‘pre’ period). Further quantitative analysis is possible, and we did conduct a follow-up study of the different correlates of young people’s travel throughout the week (see Steinbach and colleagues58). To make sense of the evaluation, we would want to take seriously the implications of the qualitative work that the scheme did increase young people’s tendency to use the bus as a ‘default’ mode. However, our predefined protocol specified that the change-on-change analysis (comparing bus travel before and after in adults and young people) would be used to identify whether or not the scheme had ‘caused’ an increase in bus travel. Taking other sources of data into account risks being perceived as introducing the biases from ‘data dredging’ that pre-published protocols are designed to avoid, yet not to do so risks eliminating useful information from the evaluation.

This is not an issue unique to natural experiments. Similar issues have been raised in relation to social science research within RCTs,150 in which trial protocols and procedures designed to maximise intervention fidelity preclude the incorporation of insights from ongoing qualitative research on issues such as how best to operationalise outcome variables in in trial questionnaires, or how to collect data. The challenge is not integrating qualitative and quantitative evidence, but integrating the rather different overarching questions that the two methods relate to: respectively (in our case), ‘how has free bus travel affected the public health?’ and ‘how confident are we that free bus travel brought about this range of outcomes?’

One potential solution to this tension is to utilise a ‘capacity approach’ to evaluating programmes, changing our question from does ‘X causes Y?’ to ‘how stable is X’s capacity to effect a change in Y?’ as suggested by Cartwright and Munro. 151 A focus on the ‘stability’ of the capacities of phenomena (such as policies or programmes) does not preclude attention also to the causal effects of those phenomena, but does mean we are asking a somewhat broader question about ‘what happens’ when a policy is introduced. This requires marshalling a broader range of evidence to develop a credible argument about the likely limits and possibilities of transferability.

Given that the criteria for assessing the stability of capacities are not as well developed as those for assessing the causal claims in RCTs, the claims made from such an approach are likely to be more contentious. There is consequently more reliance on demonstrating how X causes a change in Y rather than just that it does (i.e. the ‘mechanisms’ in epidemiological terms), and on evidence to delineate what other factors can promote, inhibit or otherwise affect the ability of X to affect Y. 151 Thus, suggest Cartwright and Munro,151 an evaluation might usefully ask a set of questions such as: how does X operate to promote Y?; What is needed for X to promote Y?; What can stop/inhibit the operation of X?; What other capacities promote/inhibit Y?; What happens when capacities interact?

Although our evaluation was not designed with this approach in mind, we have used elements of this logic. One particularly important element for external validity (especially with regard to the transferability of evaluations) is the need for greater attention to the theoretical links between concepts (such as between using public transport and active travel), rather than simply the links between indicators (such as the relationship between travel diary reports of bus travel and diary reports of walking and cycling). The indicators for the variables in our causal pathway model in Figure 3 were inevitably chosen pragmatically, from data sets available from before and after the intervention. Indicators such as ‘number of kilometres walked’ as recorded in travel diaries can only ever be partial measures of complex concepts such ‘active travel’. Methods for RCTs have largely been developed in the context of well-understood biological mechanisms, and where reasonable consensus exists around how to judge the reliability and validity of indicators. Extending the logic of these methods to natural experiments of poorly understood and undertheorised systems such as transport carries a risk of focusing our attention on the causal pathways between low level indicators, rather than on the links between concepts and indicators, or between concepts themselves.

This is where qualitative analysis can make an important contribution. Here, we have used the qualitative analysis to shed light on the relationships between concepts and indicators (in, for instance, exploring whether or not bus travel is properly identified as a ‘passive’ mode); on the limitations of current indicators (in, for instance, shedding light on the limitations of using our only available data set, on term-time weekday travel, as an indicator for mode shift); and on the plausibility of causal attribution (in, for instance, the finding that the scheme itself is likely to have increased bus use among young people).

In developing the methodologies needed for evaluations of natural experiments, we suggest a ‘capacities’ approach might be a fruitful avenue for designing studies which maximise both internal and external validity. Where mechanisms are complex, contested and poorly theorised (which applies to most up-stream determinants of health), more attention to adequate theorisation of the conceptual links between determinants of health and health behaviours or outcomes, and the links between concepts and indicators, is essential. The challenge is that this will develop incrementally throughout an evaluation. It is not simply a matter of qualitative work informing the research questions and illuminating quantitative findings, or early quantitative results informing the qualitative interviews. Insights and understanding will inevitably emerge in an incremental way. Developing the evidence base on the structural determinants of health may require rethinking our reliance on models of research deriving from evaluations of effectiveness. In particular, it will entail taking more seriously judgements about the capabilities of interventions, and drawing on a range of research designs148 and sources of evidence (both qualitative and quantitative), rather than relying solely on research designed test hypotheses about ‘what works’.

The implications of our findings for public health

Generalisability

In terms of adding to the public health evidence base, to what extent is credible evidence linking the introduction of free bus travel to health outcomes in London generalisable beyond one (atypical) setting? London has relatively good public transport networks, and there are indications that, for instance, there is less attachment to car transport as a socially prestigious mode than in other parts of the country. 40 Here, a randomised controlled design, even if such a study were theoretically possible, would not necessarily be a ‘stronger’ design. As Cartwright argues,151,175 there are inherent weaknesses in terms of external validity from the focus on overly refined causal chains which are insufficiently theorised in terms of links between more abstract concepts. In this setting, the risk would lie in making a link between, say, ‘free bus travel’ and ‘fewer miles walked’ with insufficient conceptualisation of what those links actually mean in this context. Is ‘free bus travel’ an empirical indicator of: a financial saving, a plastic card bearing a photo, the ability to travel across London or the status that derives from an entitlement as a citizen of London? Our qualitative work has mitigated this threat to generalisability to some extent, through unpacking what the various indicators in our model might mean (see Figure 19).

The public health implications of transport planning in London are typical of those in many cities across the world. Transport systems which foster ever increasing car dependency are not sustainable107 and a growing public health literature identifies the short-, medium- and long-term implications for health of not addressing this dependency. We have evaluated a scheme which, as part of a broader raft of policies that encouraged public transport use, had promising implications for mitigating some of these effects. Specifically, we identified evidence that the scheme played a role in ‘normalising’ bus transport for young people, an important precondition of reducing future car dependency, and one faced in other urban environments in which car drivers typically have negative views of public transport deriving from lack of experience using it. 176 London may be unique in its transport system and the governance of that system, but we have outlined the key features of this ‘uniqueness’ that contribute to the effects identified of the scheme: that it was introduced in the context of general improvements in the bus service and other policies to reduce private car use. However, in a UK context, a key question is the generalisability of the scheme overall to settings with deregulated bus services, where levers for improving public transport lie primarily with operating companies rather than transport authorities. A recent Passenger Transport Executive Group report177 addressed bus transport for young people in urban areas outside London, noting the importance of good bus provision for their current access to activities necessary for well-being and to ensure that they continue to see public transport as a viable option in adulthood. Citing a number of concessionary schemes that have achieved gains in bus ridership and acceptability to young people, the report suggests that simplicity in fare structures and developing services in tandem with young people are key factors. Detailed comparisons of the London free bus scheme with the outcomes of schemes in other settings, particularly those in which there are Integrated Transport Authorities, will help elucidate what can be gained in terms of ‘normalising’ bus use in urban centres with varying transport governance arrangements. In the case study described here, there is no reason to suppose that many of the benefits reported would not be replicable in other metropolitan contexts where public transport is destigmatised. Although London is unusual in that public transport tends to be used right across the social spectrum, in a different context, Baker and White178 found that those with access to cars were also making significant use of buses following the introduction of concessionary travel. This, they point out, may have implications for wider policies aimed at traffic reduction. In a global context where the need to introduce measures to address carbon emissions are largely recognised in high- and middle-income countries, there may also be wider benefits to extending concessionary schemes in differing social, economic and political contexts. Debates about how to not only ‘normalise’ but to privilege public transport are now gaining international policy traction with, for instance, a Delhi High Court judgement rejecting a proposal to abolish the Delhi Bus Rapid Transit System with the words: ‘A developed country is not one where the poor owns cars but it is one where the rich use public transport’. 179

We have presented evidence of the effects, mechanisms and influences at work when free travel is offered to young people in the context of a good bus network, used by the broader population. Future users of this evidence can use it to help assess whether or not those mechanisms and influences are similar enough to their own setting to predict likely effects.

Implications for further research

We have identified a number of areas where more evidence would be useful to inform policy.

Funding acknowledgement

This project was funded by the National Institute for Health Research Public Health Research programme (project number 09/3001/13). AG contributed to this research while funded by an NIHR post-doctoral fellowship.

Acknowledgements for contributions to the project

The team would like to thank the following people:

Avril Porter, project administrator, for her invaluable input to the administration and management of the study.

Chris Lines (Transport Consultant) for his support and enthusiasm for the project throughout; his skill and insights as chair of the study Steering Committee and for chairing a stakeholder workshop.

Suzanne Lutchmun (NHS Camden), member of the Steering Committee, for her input into the original idea for the project and for bringing public health insights to the discussions.

Alex Phillips and Richard Jeremy (TfL) for their support as members of the Steering Committee, generosity with their time in providing information and data and helpful input to the report writing.

Steven Salmon (Confederation of Passenger Transport) for his invaluable input to the Steering Committee.

John Barry (TfL) for input to stakeholder workshops and helpful advice.

Zara Todd for help in identifying participants and advice on young disabled travellers.

Alan Shiell (Executive Director of the Centre of Excellence in Intervention and Prevention Science, Melbourne, Austrailia) for health economics input, support and helpful discussions throughout and for chairing a stakeholder event.

Lucinda Saunders (GLA) for her contributions to the project while based at LSHTM, in particular for conducting a literature review of the impact of active travel on health outcomes and for providing a commentary for the stakeholder event ‘On the Buses Workshop’.

Bhash Naidoo (NICE) and Kenny Lawson (Glasgow MRC) for their presentations at the Stakeholder Event on ‘Economic Evaluations for Informing Public Health Policy’.

Rebecca Fuller (Passenger Transport Executive Group), Risan Nishori (Robert Clack School) and Reece Weaver (Robert Clack School) for their commentaries on our draft findings at the Stakeholder Event ‘On the Buses Workshop’.

We are also grateful to the participants from the DfT, TfL, London Boroughs, Kilburn Older Voices Exchange, Living Streets, Confederation of Passenger Transport UK, NHS organisations and academic colleagues who contributed to these events.

All the young people who conducted research inspired by On The Buses as part of LSHTM’s Young Scientists’ Programme.

Monisha Sodha, Nancy Williamson and Shirley Solomon for their administrative help on the project.

Finally, our greatest debt is to all the young people and older citizens who kindly talked to us about their travel in London, and the various schools, community organisations and local councils who helped with contacting those individuals.

Prior publications

This report draws on the following work already published:

Jones A, Steinbach R, Roberts H, Goodman A, Green J. Rethinking passive transport: bus fare exemptions and young people’s wellbeing Health and Place 2012;18:605–12.

Jones A. Free for some: setting the context for the ‘On the Buses’ study. Occasional papers in transport and health (1). London: LSHTM; 2010.

Wilkinson P, Edwards P, Steinbach R, Petticrew M, Goodman A, Jones A, et al. The health impact of free bus travel for young people in London: protocol for an observational study. Occasional papers in transport and health (2). London: LSHTM; 2011.

Green J, Jones A, Roberts H. More than A to B: the role of free bus travel for the mobility and wellbeing of older citizens in London. Ageing Soc 2012. http://dx.doi.org/10.1017/S0144686X12001110

Edwards P, Steinbach R, Green J, Petticrew M, Goodman A, Jones A, et al. Health Impacts of free bus travel for young people: evaluation of a natural experiment in London. J Epidemiol Community 2013. doi:10.1136/jech-2012-202156

Saunders L, Green J, Petticrew M, Steinbach R, Roberts H. What are the health benefits of active travel? A systematic review of trials and cohort studies. PLOS ONE 2013;8:e69912.

Goodman A, Jones A, Roberts H, Steinbach R, Green J. ‘We can all just get on a bus and go’: rethinking independent mobility in the context of the universal provision of free bus travel to young Londoners. Mobilities 2013. http://dx.doi.org/10.1080/17450101.2013.782848

Contributions of authors

Dr Judith Green (Reader, Sociology of Health, LSHTM) was the principal investigator. She co-ordinated the study, contributed to the study design, qualitative fieldwork and analysis, interpretation of data and writing the report.

Ms Rebecca Steinbach (Research Fellow, Social Epidemiology, LSHTM) prepared data sets (London travel surveys, STATS19 road casualty data and HES), conducted statistical analyses and prepared data for the economic evaluation. She contributed to the qualitative fieldwork, qualitative data analysis and the writing of the report.

Dr Alasdair Jones (Research Fellow in Qualitative Research, LSHTM) contributed to the design of the qualitative study, led the qualitative fieldwork, and contributed to analysis, interpretation of findings and writing the final report.

Dr Phil Edwards (Senior Lecturer, Statistics, LSHTM) contributed to the study design, prepared data sets (London travel surveys, STATS19 road casualty data and HES) and conducted statistical analyses. He contributed to the writing of the report.

Ms Charlotte Kelly (Senior Research Fellow, ITS, University of Leeds) contributed to the economic analysis, interpretation of data and writing the report.

Dr John Nellthorp (Senior Research Fellow, ITS, University of Leeds) contributed to the economic analysis, interpretation of data and writing the report.

Dr Anna Goodman (Research Fellow, Social Epidemiology, LSHTM) contributed to the qualitative data collection and analysis, interpretation of the data and writing the report. She led the youth engagement strand of the project.

Professor Helen Roberts (Professorial Research Fellow, Medical Sociology, UCL Institute of Child Health) contributed to the study design, qualitative fieldwork, interpretation of findings and writing of final report.

Professor Mark Petticrew (Professor, Public Health Evaluation, LSHTM) contributed to the study design, interpretation of findings and writing of final report.

Professor Paul Wilkinson (Professor, Environmental Epidemiology, LSHTM) contributed to the study design, interpretation of the data and the writing of the final report.

Disclaimers

The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The PHR editors and publisher have tried to ensure the accuracy of the authors’ report and would like to thank the referees for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this report.

The views expressed in this publication are those of the authors and not necessarily those of the PHR programme or the Department of Health.

The intervention: free bus travel for young people

The specific interventions concerned are the introduction of free bus fares for 12- to 16-year-olds in September 2005 and the extension of this concession to 17-year-olds in full-time education September 2006. Fares are exempted for London residents in this age bracket (12–17 years) who apply for a ‘zip card’ from TfL and use it according to the terms and conditions of the issuer. 196,197 These fare exemptions for young people were introduced during the tenure of the previous Mayor of London, Ken Livingstone (Mayor 2000–8). When the first intervention concerned was launched (2005), secondary school-aged children had paid a reduced, 40p flat fare for journeys on the London bus network (Table 28). As well as granting the cardholder unlimited free travel on all buses and trams displaying the London Buses symbol (both within and just outside London),12 zip cards also act as conventional ‘Oyster’ cards and can be loaded up with pre-pay or travelcards for the cardholders to use other parts of the TfL network (tube, DLR, London overground and most National Rail services operating in the capital) at a variety of discounted rates (pp. 6–11). 14

When these fare exemptions were unveiled, the stated rationale for universally eliminating bus fares for young people in London was to help them to continue studying, improve employment prospects and promote the use of public transport (p. 7). 13 That is, it was aimed first and foremost at mitigating the potential social exclusion effects for young people of fare-based urban transport systems. 15 As it has been stated more recently on the TfL website:

Granting young people free travel is part of the Mayor’s strategy to embed more environmentally sound travel habits from an early age while helping young people to unlock education, sport, leisure and employment opportunities

TfL, 200716

By removing any need to pay, at the point of use, for travel on buses, the argument went that young people would be better (and more equally) able to access goods and services (schools, libraries, leisure facilities, etc.) and so reduce the chances of their suffering from transport poverty. At the same time, it was hoped that by encouraging bus use from an early age more environmentally sustainable travel practices would become ingrained.

Young people with disabilities living in London can also apply for a Freedom Pass, if they have an eligible disability. 198 These provide free travel on London underground and overground services as well as the bus routes. Most young people with eligible disabilities use the Freedom Pass for public transport use, rather than a zip card.

Bus ridership data and trends: overview

One of the key background trends for this study is the substantial increase in the use of mass public transport recorded in London in recent years (Figure 20). For example, it has been shown that ‘[t]otal passenger-kilometres travelled on [all public transport] services operated by TfL were almost 70 per cent higher in 2008/09 than in 1991/92 . . . this growth . . . was especially pronounced on the bus network’ (p. 45). 31 Specifically, in the same period (1991/2 to 2008/9) bus patronage increased by 93%; moreover, between 2000/1 and 2008/9, recorded bus passenger kilometers increased by 64% (p. 45). 31

FIGURE 20.

Modal share (%) for journey stages, 1993 and 2008 (source: adapted from TfL 201031).

In 2008/9, TfL’s ‘best estimate’ is that 7942 million passenger-kilometres were travelled by bus (p. 47, table 2.5). 31 This compares with 6755 million passenger-kilometres travelled by bus in 2004/5 (just prior to the first of the interventions evaluated in this study), 4709 million passenger-kilometres in 2000/1 (the year that the GLA was established) and 3996 million bus passenger-kilometres in 1991/2. [Between 2006/7 and 2007/8 TfL made a significant change to the methodology used to estimate bus trips and journey changes. In short, from 2007/8 onwards they have used Oyster card validations as the primary source of these estimates rather than ticket sales. The Oyster card validation record, they argue, ‘provides more robust estimates of total bus and tram use’ (p. 48). 31 The effects of the estimation method change in 2007/08 were to ‘add almost 0.4 million bus trips and 0.5 million journey stages to the daily average, and [to increase . . .] the estimate of bus mode shares by 1 percentage point’ (p. 48). 31 See Transport for London: Travel in London: Report 2 (pp. 45–9)31 for further information on this travel data estimation change.] In 2008, using journey stage data, bus travel (including tram travel) accounted for 20% of the modal share of transport in London, the highest for any public transport mode, up from 13.5% in 1993. In the past decade, then, there has been a significant shift in modal choice towards buses. This period is one characterised by a shift to GLA/TfL oversight of London’s transport infrastructure and services. The backdrop to this study is, therefore, one of high levels of and pronounced increases in bus usage in the capital.

Data sources

1. STATS19.

2. HES.

3. London Travel Surveys (LATS and LTDS).

Participation, engagement and involvement of ‘young scientists’ in an ongoing study

On the Buses zip card user: topic guide (focus groups)

On the Buses Freedom Pass user: topic guide (interviews)

Abstract

Background The extension, in September 2005, of free bus and tram travel in London to people 12–16 years of age and, in September 2006, to people under 18 years is likely to have had a range of impacts with implications for public health. The ‘On the Buses’ project aims to evaluate these impacts using a mixed method quasi-experimental design. This paper describes the protocol for the analyses of quantitative data for the study.

Methods/design Analyses will be based on routine travel survey and injury data for London, and will primarily entail comparison of pre-intervention to post-intervention change in the target age-group (12–17 years) against the corresponding change in people aged 25–59 years. The main outcome measures will include frequency and distance of all travel, and of active travel; frequency of independent travel, bus use; percentage of journeys < 1 km travelled by mode; incidence of road injury, and of intentional injury. We will use conditional fixed-effects Poisson models.

Discussion This quantitative study is part of a larger evaluation which draws on qualitative data, economic evaluation and literature reviews to describe the effect of free bus travel for young people on public health. It will also contribute to methodological development in relation to causal attribution in the absence of controlled experimental evidence, and in the use of routine data sets for assessing the effect of interventions on public health.

We hypothesized that the introduction of free bus travel in London would have several important effects, both direct and indirect (Figure A7.1). To test these adequately requires a combination of quantitative and qualitative methods. The quantitative study is one component directed towards testing the active travel, road injuries and intentional injuries hypotheses. Its protocol is described in this paper.

Methods/design

The overall aim of this study is to characterize changes in health behaviour and selected health outcomes in children and older adults following the introduction in September 2005 of the London free bus travel scheme for children aged 12–16 years and its subsequent extension to under those 18 in September 2006. There are three specific objectives:

1. To assess the impact of free bus travel for 12–17 year olds on their use of bus and other transport modes and on their non-car travel overall;

2. To assess the impact of free bus travel for 12–17 year olds on the use of bus and other transport modes by older age population groups;

3. To identify changes in the incidence of injuries in young people under 18 following access to free bus travel.

Research hypotheses

We hypothesize that the introduction of the free bus pass scheme is associated with:

1. an increase in bus use and overall ‘independent’ travel (the latter represented from available data by all non-car travel), but a reduction in active transport (walking and cycling) and car use, among the target age-group [Independent travel refers to travel young people do without guardians. With no direct measure of this, we use a proxy measure of all travel except car travel and, for those under 17 years, motorbike travel.];

2. a reduction in bus use and trips < 1 km made by people aged 65+, especially during the hours when children usually travel from, school;

3. a reduction in road traffic injuries in the target age group; and

4. an increase in intentional injury rates in the target age group.

   1. We also hypothesize that:

5. changes will be more pronounced in the inner-London boroughs (with denser bus networks) than in outer-London boroughs;

6. changes will be more pronounced in boroughs with a known higher take-up of free bus travel;

7. changes in distance/frequency of bus travel, independent (non-car) travel, and active travel, and in injury incidence, are greater in households with low income;

8. reductions in car use will be greater in households with high income;

9. changes in distance/frequency of bus travel and active travel, and in injury incidence, will be the same across all ethnic groups.

Outcomes and measures

The outcomes we will use to assess the hypotheses are as follows:

1. frequency and distance of all transport, of active transport (i.e. walking and cycling) and of independent (non-car) transport (i.e. walking, cycling and public transport) in people aged 12–17 years;

2. frequency of bus use and distance travelled by bus in people aged 12–17 years;

3. frequency of bus travel and distance travelled by bus in other age groups;

4. incidence of intentional and non-intentional injuries in people aged 12–17 years.

We will use three main sources of data for our outcome measures.

Power and sample size

The LATS and LTDS samples include data on 3,000 young people before and after the intervention, giving over 80% power to detect a 10% relative reduction in average distances walked daily by young people (i.e. from 0.9 (SD 1.3) km to 0.8 (SD 1.3) km per day) at a 5% significance level. Similarly, the study will have over 90% power to detect a 10% increase in the average distance of bus travel (i.e. from 4.3 (SD 4.1) km to 4.7 (SD 4.1) km per day). For transport-related injury, the study would have 80% power to detect a 10% change, or 90% power to detect a 12% change significant at the 5% level. Statistical power is inevitably more limited for subgroup analyses, but there will be 90% power to detect a 15% change in average distance travelled by bus by young people within the most deprived quartile, for example.

Analyses

(i) Analyses will compare the changes in each outcome variable in the pre versus post intervention time periods in the target age group (12–17 years) to changes in the outcome variable pre and post intervention in 25–59 year olds (see Table 1 and Figure A7.2):

(ii) We will conduct similar analyses to compare pre–post intervention changes in outcomes by subgroups: area of London (inner versus outer; areas of high intervention take up versus low intervention take up); deprivation group (most deprived fifth of population versus least deprived 80%); household income (< £15k per year versus > £50k per year (for travel patterns only)) and ethnicity (White, Black, Asian, other) – see Table 3.

(iii) To explore whether older citizens are being displaced from buses and travel more broadly, we will compare the pre-post change in older citizens’ travel during post-school commuting hours versus other times (see shell Table 2). (Prior to January 2009, older citizens were not able to use free buses before 9.30 a.m.)

(iv) For STATS19 road traffic injury data we will implement conditional fixed effects Poisson regression using Stata’s xtpoisson command, based on annual counts of casualties and collisions. Robust standard errors will be obtained using jackknife procedures clustering on borough (n = 33). The underlying trends in casualties and collisions will be fitted using linear terms. Analyses will be stratified by age-group, and comparisons will be made between the 12–17 years and older ages.

References

What are the health benefits of active travel? A systematic review

Lucinda E Saunders, Judith M Green, Mark P Petticrew, Rebecca Steinbach, Helen Roberts

Abstract

Background: Increasing the proportion of active travel (primarily walking and cycling) within the population has been widely advocated as a route to reducing obesity levels and achieving other population health benefits. However, the strength of evidence underpinning these assumptions is not known. The objective of this study was to assess the evidence that active travel has significant health benefits.

Methods: The study design was a systematic review of (i) non-randomised and randomised trials, and (ii) observational studies examining either (i) the effects of interventions to promote active travel or (ii) the association between active travel and health outcomes. The data sources were published and unpublished reports of studies in any language identified by searching 11 electronic databases, websites, reference lists and papers identified by experts in the field. The eligibility criteria were prospective observational and intervention studies measuring any health outcome of active travel in the general population were included. Studies of patient groups were excluded.

Results: Twenty-two studies from 11 countries were included, of which six were studies conducted in children. Six studies evaluated active travel interventions. Sixteen were prospective cohort studies which did not evaluate the impact of a specific intervention. No studies were identified with obesity as an outcome in adults; three prospective cohort studies in children found no association between obesity and active travel. Small positive effects on other health outcomes were found for active travel in six intervention studies, but these studies were all at risk of selection bias; modest benefits for other health outcomes were identified in 13 prospective studies.

Conclusions: However there is accumulating evidence that active travel may have positive effects on other health outcomes than obesity. Further trials of interventions to increase active travel are needed to assess its impacts on obesity and other health outcomes.

Keywords: Public health; systematic review; active transport; physical activity

Background

The link between physical activity and health has long been known, with the scientific link established in Jerry Morris’ seminal study of London bus drivers in the 1950’s [1]. There is also good ecological evidence that obesity rates are increasing in countries and settings in which ‘active travel’ (primarily walking and cycling for the purpose of functional rather than leisure travel) is declining [2–3]. Given that transport is a necessity of everyday life, whereas leisure exercise such as going to a gym may be an additional burden for individuals, and is difficult to sustain long term, [4–5] encouraging ‘active travel’ may be a feasible approach to increasing levels of physical activity [6]. It is also plausible to assume that interventions aimed at increasing the amount of active travel within a population are likely to have a positive impact on health. This has been the underlying rationale for recent public health interest in transport interventions aiming to address the obesity epidemic and a range of other health and social problems [7]. In July 2011 the Chief Medical Officers for the UK produced new guidance for physical activity ‘Start Active, Stay Active [8]’ which re-emphasised the known risks to health of the sedentary behaviour of the UK population and the central role of active travel: “For most people, the easiest and most acceptable forms of physical activity are those that can be incorporated into everyday life. Examples include walking or cycling instead of travelling by car, bus or train.” (p17). Active travel is seen by policy makers and practitioners as not only an important part of the solution to the problem of obesity but also for a range of other health issues at a population level. It may also have other important non-health outcomes, including a reduction in traffic congestion and carbon emissions [9].

It has recently been recommended that the public health community should advocate for effective policies that reduce car use and increase active travel [10]. One recent overview concluded that active travel policies have the potential to generate large population health benefits through increasing population physical activity levels, and smaller health benefits through reductions in exposures to air pollution in the general population. However, while a recent systematic review and meta-analysis [11] found that non-vigorous physical activity reduced all-cause mortality, the two studies which looked at active commuting alone [12–13] found no evidence of a positive effect. There are a number of reasons why active travel may not contribute to overall physical activity levels. Studies of young children have found no differences in overall physical activity levels for active and non-active commuters [14–16], perhaps because the distance walked to school may simply be too short to make a significant contribution. For both children and adults, it is unclear how far individuals may offset the extra effort of cycling or walking with additional food intake, or by reducing physical activity in other areas of everyday life. Additionally, there is evidence that the health benefits of exercise are not shared equally across populations, with the cultural and psychological meanings of activities such as walking or cycling potentially influencing their physiological effects [17–18].

A reliable overview of the strength of the scientific evidence will therefore be of value, because the causal pathways between active travel and health outcomes such as obesity are likely to be complex, and promoting active travel may have unintended adverse consequences [19].

Overall, aside from impacts on all-cause mortality, the evidence that active travel will lead to a range of other health and social benefits including a reduction in obesity remains to be determined, and existing studies show a mixed picture on the relationship between active travel and health outcomes including obesity [20]. Recent systematic reviews have focussed almost exclusively on cross-sectional studies [20–22], or one narrow health outcome [23] or combined leisure and transport activity [24]. A clear picture of the health effects of active travel cannot be obtained from these reviews so a systematic review of evidence from empirical studies was carried out with the objective of assessing the health effects of active travel specifically (rather than of physical activity in general, where the evidence is already well-established). It synthesises the relevant empirical evidence from intervention studies and cohort studies in which health outcomes of active travel have been purposively or opportunistically measured. Obesity is a particular focus of this review because the rise in the prevalence of obesity over the past 30–40 years has been in tandem with the decline of active travel and overweight and obesity are now the fifth leading risk for death globally as well as being responsible for significant proportions of the disease burden of diabetes (44%), ischaemic heart disease (23%) and some cancers (7–41%) [25].

Methods

A systematic review was undertaken to identify and synthesise the evidence for the impact on obesity and other health outcomes of active travel (Review protocol and full search strategy is available on request from the authors). Eleven databases were searched for prospective studies of any design (Table 1) (this search will be updated if the paper is accepted). The search strategy adapted the search terms developed by Hoskings et al. [26] (2010 Cochrane Review) and Bunn et al. [27] (2003). Hand-searching of relevant studies was also conducted, and bibliographies of identified papers were checked along with those of papers already known to the researchers. No time, topic or language exclusions or limits were applied.

Two reviewers independently identified potentially relevant prospective studies. If it was not clear from the title and abstract whether the article was relevant to active travel then the paper was reviewed in detail. Papers which were not published in English would have been translated in full if deemed potentially eligible for inclusion. One reviewer [LS] then screened the articles using the following inclusion criteria:

1. Prospective study examining relationship between active travel and health outcomes; or study evaluating the effect of an active travel intervention;

2. Active travel (walking or cycling for transport rather than work or leisure) measured in a healthy population (e.g. using self report measures, or use of pedometers); and

3. Health outcome included.

Retrospective and single cross-sectional studies (e.g. one-off survey) were excluded.

One reviewer (LS) extracted data according to a common table format including information on methods, outcomes (as adjusted relative risks, or hazard ratios; if these were not available or calculable, other effect measures were extracted – e.g. mean changes), populations and setting for each study, and two reviewers then independently reviewed each study using the ‘Effective Public Health Practice Project Quality Assessment Tool’ [28] and discussed differences to produce consensus scores for each study against each quality criterion. A narrative review, taking account of study quality, was then carried out. It was not possible to conduct a meta-analysis as meta-analysis of mixed designs – and particularly observational studies – is not robust and this review includes a range of mainly observational study designs, populations and outcomes [29].

Results

Twenty-two studies reported in twenty-seven papers were included (see Tables 2 and 3). Three were prospective cohort studies with obesity-related outcomes; thirteen were prospective cohort studies with other health outcomes; and six were intervention studies with other health outcomes (see PRISMA diagram for details of excluded studies - further details available on request from the authors). For the prospective cohort studies the results are presented adjusted for covariates, there was variation in what adjustments were made by different studies but the adjustments did not have large impacts on effect size. Details of the methodological assessment of each paper are reported in a separate web table.

Discussion

This review found no prospective studies of active travel with obesity as a primary outcome in adults, and no significant associations between obesity and active travel were found in the studies of children. For other health outcomes small positive health effects were found in groups who actively travelled longer distances including reductions in risk of all cause mortality [41], hypertension [44], and Type 2 diabetes [46].

The evidence is difficult to synthesise as active travel is not defined consistently across studies, and the definition is dependent on what is considered normal in a particular setting. For example Luoto (2000), and Barengo (2004, 2005) [12, 42] considered active travel to be more than 30 minutes per day and inactive travel to be less than 30 minutes per day. Batty (2001) and Hayashi (1999) [44] however considered active travel to be more than 20 minutes per day. Differences in health outcomes between people who actively travel 29 minutes per day and those who travel 31 minutes per day are unlikely, so differences between active and sedentary populations may be masked by the methods by which active travel is defined and reported. Meanwhile Besson (2008) [55] considered active travel to be >8 metabolic equivalent task (MET) hours per week while Matthews (2007) [51] considered it to be > 3.5 metabolic equivalent task hours per day which may reflect differences in norms between Norfolk and Shanghai in terms of active travel.

It is difficult to generalise from studies conducted in different countries or settings, as the amount of exertion required to travel actively may be greater in some settings than others with the same journey time, due to differences in congestion, terrain, and weather conditions. In the UK only 39% of men and 29% of women achieve 30 minutes of moderate intensity physical activity of any type five times a week so a population change of adding 30 minutes of active travel per day might well produce much larger changes in health outcomes than were measured in non-UK studies. For example in Barengo (2005) [43] 30% of men and 46% of women travelled actively more than 15 minutes each way to work which shows a higher level of active travel in the population than may be seen in other countries, including the UK. The prospective cohort studies also tended to focus on travel to work or school rather than active travel for general transportation, which again may limit generalisability.

The study by Cooper et al. (2008) of school children in Odense, Denmark found that 65% of boys and girls walked or cycled to school, a much higher proportion than would currently be found in the UK. However, journey times were less than 15 minutes for the majority of active travellers so the health effects of active travel for such short periods are difficult to measure in isolation. This highlights one of the difficulties of assuming active travel to school in young people to be a major source of physical activity, as it is common for children only to walk or cycle to school when the journey time is relatively short. While in adults bouts of as little as 10 minutes of physical activity are acceptable to contribute to their weekly physical activity target of minimum 150 minutes, children aged 5–18 are expected to be physically active for a minimum of 420 minutes per week [8] so a short active commute to school will not make a significant contribution to their overall physical activity requirements. The study by Lofgren et al. [49] included a study population with fairly high levels of physical activity overall and half the participants were active travellers which makes it difficult to attribute health outcomes to active travel alone when active travel may not contribute significantly to participants overall physical activity levels.

De Geus et al. (2007) [33] highlighted one of the difficulties of measuring active travel in intervention studies as they found that study participants cycled 13% faster when their fitness was being measured compared to their usual speed on their daily cycle commute. Hence just through the process of measuring active travel there is the risk of over-estimating the health benefits it can confer. Thus issues including terrain, problems of definition, study design and the difficulty of disentangling the effects of active travel from more general physical activity make synthesis difficult. It is in this context that a wider range of studies, including ethnographic and qualitative work, to inform generalisable theory may be helpful. Active travel not only potentially benefits health as a source of physical activity but may also off-set air pollution from motorised vehicles for those journeys and contribute to social and environmental goals such as improving social cohesion and reducing CO₂ emissions. These combined benefits may be a potent argument for promoting active travel, and emphasise the importance of models which incorporate both health and non-health benefits [57–58].

Finally, reviewing this evidence is challenging, in particular designing searches which are both sensitive and specific, and over 70% of the studies identified for this search were found through hand-searching, despite a thorough search of databases. While it is possible that studies may have been missed, our comprehensive search for studies makes it unlikely that a significant body of work has been excluded.

Conclusions

This is the first study to bring together all prospective observational and intervention studies to give an overview of the health effects of active travel in general. Previous systematic reviews of health outcomes of active travel have included primarily cross-sectional studies from which reliable inferences about causality cannot easily be drawn, or have relied on indirect evidence on the effects of physical activity on health, as opposed to the effects of active travel.

The interest in identifying health outcomes of active travel is to determine whether active travel affects health. However this will depend on the context within which individuals are travelling – length of journey, frequency of travel, nature of the terrain, risk of injury, levels of air pollution and so on as well as other aspects of the lifestyles of the participants. For example travelling actively may mean that the individual is more or less likely to be physically active at other times, or they may modify their diet. It may mean that they are more or less likely to strengthen social networks. While the studies identified in this review do not enable us to draw strong conclusions about the health effects of active travel, there is consistent support for the positive effects on health of active travel over longer periods and perhaps distances.

This review highlights the difficulty in measuring health outcomes of active travel in the general population. In prospective cohort studies if the follow-up period is short then it may not be possible to measure health effects that take many years to appear, such as changes in risk of heart disease. Conversely in those studies which have long follow-up periods of many years there is the risk that active travel has not been consistently adhered to throughout the follow-up period.

These cautious conclusions on the health impact of active travel do not, of course, mean that now is the time to confine active travel to the walk from the front door to the car door. The evidence on the effect of physical activity is sufficiently strong to suggest that the part played by active travel in improving health is likely to be significant and is well worth maintaining. Other aspects of active travel, including a reduction in pollution, and in carbon footprint are clear potential co-benefits and likely to become even more so.

Appendices:

PRISMA flowchart

Table 2 – results [available from authors]

Table 3 – summary of outcomes [available from authors]

Search strategy

Quality assessment table [available from authors]

Ethics approval and patients’ consent, funding sources and independence:

Ethical approval was not required as this is a systematic review. This project was funded by the National Institute for Health Research Public Health Research programme (project number 09/3001/13). The views and opinions expressed herein are those of the authors and do not necessarily reflect those of the Department of Health. The funders had no role in the design, conduct or reporting of project findings.

References

Search Strategy

Final Search Strategy for Medline

Search in ‘keyword’

1. (Walk* or cycle* or cycli* or bicycle* or bike* or walking bus* or ecological commut* or ecological transport* or non-auto* or non-motori?e*).mp.

2. (green* adj3 (travel* or transport* or commut*)).mp.

3. ((activ* or health*) adj3 (travel* or transport* or commut*)).mp.

4. (ecological adj3 (travel* or transport* or commut*)).mp.

5. or/1-4

6. (Prospective adj1 (study or studies)).mp

7. (Program evaluation or evaluation research).mp.

8. (randomi$ or randomly).mp.

9. (controlled adj2 (trial or trials or study or studies or experiment$)).mp

10. (before adj1 after).ab,ti.

11. (control adj1 group$).ab,ti.

12. (intervention adj1 group$).ab,ti.

13. (experimental adj1 group$).ab,ti.

14. (comparison adj1 group$)ab,ti.

15. Or/6-14

16. (Body mass index or BMI or body mass gain or overweight or weight gain or weight status or weight management or weight loss or body weight or waist circumference or obesity or calorie* or energy)

17. 5 and 15 and 16

18. 17 in humans