Scottish Government

BARRIERS TO MODAL SHIFT

CHAPTER TWO Literature Review

2.1 There is a growing body of literature on the social, economic, cultural, and environmental factors affecting travel choices. Yet progress to incorporate this growing body of knowledge within practical transport planning and transport modelling has been slow. Consistent with the aims of this project, this literature review considers what lessons can be drawn about barriers to mode shift from the car to other modes, and policies to overcome these barriers based on published research on travel behaviour. It then reviews experiences on the application of these lessons within transport modelling and analysis.

Part 1 - Barriers to mode shift

2.2 In considering barriers to mode shift from car travel to other modes there are a number of strands that need to be considered. Firstly it is necessary to consider the main gaps between public policy aims and public attitudes and behaviour. If there is a gap between what is socially desirable for mode choice and what people desire, then it is necessary to understand and address the factors which define this gap. The review therefore starts by considering car dependence and the costs and benefits of travel. It then looks at the components of these costs and benefits under three main categories:

• Time and money costs of travel, car ownership levels, and land use plans which are measured or estimated and taken account of in conventional transport models, sometimes called " hard factors".
• Other factors affecting travel choices not currently included in travel demand models (e.g. travel information) sometimes called " soft factors" (Halcrow 2002).
• Complementary factors such as the need to carry loads or young children, lifestyle issues, the weather etc.

Components of car dependence

2.3 In considering why people do not shift mode from the car it is useful to start with the literature on the reasons for car dependence. Car dependence has been classified as (Farrington et al 1998):

• Structural dependence - Those who are dependent on their cars because they have no other viable alternatives.
• Conscious dependence - Those who rely on cars but could realistically undertake their journeys by other modes.

2.4 Market distortions that contribute to car dependency have been classified as (Litman 1999):

• Limited choice - Consumers often lack viable choices, or information about those choices so are unable to make efficient decisions. Lack of choices include non transport factors such as limited availability of attractive non car dependent locations to live and transport factors such as the lack of reliable public transport, or pleasant walking routes.
• Limited competition - Lack of competition in public and privately provided roads and public transport limits incentives for innovation, quality and value in provision.
• Cost based pricing - Consumers do not bear the full costs they impose with development of automobile dependent land being underpriced and some transport choices not covering their full environmental costs.
• Wider economic factors - There are many public policies (laws, taxes, subsidies, and investment policies) that support car dependency.

2.5 Car dependence is therefore context specific and, in considering the potential for mode shift from cars, the effectiveness of policies will be highly dependent on the correct targeting of measures. It has been suggested that such targeting can be linked to types of locations, such as rural areas, types of people, such as those with a mobility impairment, and types of trip such as supermarket shopping (RAC 1995, Stradling 2002). Car dependent places are discussed below under "hard factors", and car dependent people and types of trip are discussed under "soft factors".

Costs and benefits of travel

2.6 The most commonly identified reasons why people use cars rather than other modes is that car travel offers opportunities not available by any other means (AA 1998). The last fifty years have seen large reductions in the cost and time of travel by car, which has opened up many new opportunities. Over the same period, few new opportunities have opened up for public transport travel, and in some places public transport opportunities have declined (SCC 1998). With car ownership levels in Scotland currently just over half the level that they might be expected to reach in the future (SE 2002a, Goodwin 1997), the challenge for the next fifty years is to develop a balanced approach to the costs and benefits of travel by promoting policies which achieve a sustainable approach to economic, social and environmental improvements.

2.7 Overall in the UK, travel costs for consumers are a higher percentage of GDP than for many other developed economies (SACTRA 1999). If, through market mechanisms, taxation and current public funding, society is delivering less than optimal solutions, then the challenge for transport planning is to find more socially acceptable, economically beneficial and environmentally sustainable approaches to the development of transport systems.

2.8 Traditionally, transport planning has looked at social and economic factors such as land use development plans and car ownership growth in order to predict the demand for travel. This has considered travel as a purely derived demand and constrained the range of solutions to transport factors such as new roads, buses, or trains (Ecotec 1993). In Scotland since the early 1990s, there has been a growing realisation that the demand for travel needs to be managed (SO 1992, SE 1998). If transport planners now wish to manage travel demand then a broader approach needs to be taken which includes both transport and non-transport levers to facilitate change. Rather than simply predicting the demand for travel, transport planners now need to understand travel behaviour (Jones 2002).

2.9 A broader approach to considering the costs and benefits of travel has been proposed based on travel behaviour concepts (Mokhtarian et al 2001). This suggests that travel needs to be viewed in much the same way as other "goods" having both positive and negative characteristics. People weigh the pros and cons of travel and non-travel alternatives and choose an option that involves travel where the costs or disbenefits of travel are exceeded by the combined benefits of:

• The utility of arriving at a destination - This is the accessibility benefit, and depends on both the quality of the opportunity that can be reached and the travel or telecommunications options available.
• The utility of activities that can be conducted whilst travelling - In most cases these activities can be undertaken without travelling (e.g. listening to music, working, talking to friends, reading, thinking) but they help to increase the utility of a particular choice over other travel choices or non-travel alternatives.
• The utility of travel itself - Numerous sports and hobbies revolve around the travel for its own sake (e.g. hiking). Almost by definition undirected travel is largely a leisure activity, so this element dominates for leisure travel.

2.10 From the large surveys carried out (Mokhtarian et al 2001) it has been shown that trip purpose and people group are particularly significant factors in affecting the balance between these three components of travel. "Travel liking" and "adventure seeking" personalities behave very differently from other personality types such as "calm". To assess how the costs and benefits of travel for each behavioural class are perceived, more widespread surveys of travel attitudes and behaviour will be needed to support future analysis. The Scottish Household Survey already includes some questions on attitudes to travel, but there is scope for these to be rationalised and strengthened (SE 2002b).

2.11 Overall, a behavioural model which considers the costs and benefits of travel on a continuum, from directed to undirected travel, should have the capability of reflecting observed patterns, and the scope for behaviour change, more realistically than traditional approaches considering all travel as a derived demand.

2.12 This project is looking specifically at mode shift from car to public transport, walking and cycling. The "hard factors" are most relevant to directed travel, and the "soft factors" are most relevant to the travel experience and the activities that can be undertaken whilst travelling. The "complementary factors" provide the context within which each travel decision is made. This review therefore considers each of these in turn, within the general theory of travel behaviour summarised above.

Hard factors

2.13 Probably the greatest influences on mode choice are the money costs and travel time costs of each alternative. For many car journeys, the main barriers to mode shift are simply that the time and/or costs of alternative modes are not competitive. When considered on a trip by trip basis, public transport is very often more expensive than car travel (RCEP 1994, SCC 1998) in terms of money costs. For these trips, public transport would need to offer other benefits in terms of travel time or comfort for cost barriers to mode shift to be overcome.

2.14 For making the trade offs between different modes on the basis of cost and time, it is convenient to convert time to costs by using values of time, to create 'generalised costs' for each modal alternative. There is a very extensive literature on the valuation of travel time (Wardman 1998) and, although there is an ongoing need to update and refine this (SACTRA 1999), a number of key points emerge from the existing research as follows:

• The use of a single value of time, as applied in most current public sector appraisals, does not reflect real travel behaviour. Abandoning the single value would greatly increase the complexity of modelling and introduce equity problems for cost benefit analysis, but might be more realistic and insightful.
• However, in schemes involving cash payments (e.g. bridge tolls), such disaggregation is necessary to derive demand curves and so predict levels of use at different price levels.
• Value of time varies by income level, age of traveller, journey length, trip purpose and mode.
• The influence of mode on value of time is highly correlated with income, but within income groups bus travel attracts higher values of time and rail lower values than car, reflecting the relative perceived comfort of these modes.

Travel reliability

2.15 Travel behaviour research has consistently shown that reliability is one of the most important factors affecting individual travel decisions. Despite this importance it has rarely been considered effectively in appraisal (Bates 2001). Recent multi-modal studies have aimed to measure public transport reliability in terms of the difference between timetabled and actual travel times, and for car travel reliability is measured as variability in actual travel times (DETR 2000). In practice most assessments have lacked good quality data (MVA 2002).

2.16 If a delivery of goods is regularly late, or a businessman keeps missing meetings, or an employee is repeatedly late for work due to transport delays then the company, businessman, and individual are accountable for their failure to perform. Reflecting this accountability, the behavioural response of the company or traveller is usually to seek to change those aspects of the transport system over which they have control to allow them to meet their needs better (Halden 2002). As highlighted in paragraph 2.23, car travel is associated with "control" so lack of reliability in transport systems will be a significant barrier to mode shift from car to public transport.

Car dependent places

2.17 The links between urban form, the sets of travel choices available to people in each type of location and the way that these choices influence travel behaviour have been studied in some depth (Handy 1992). Although it has been shown that land use patterns have a very substantial impact on travel behaviour, the relative influence of population density, land use mix, and urban design is not well defined.

2.18 Although studies have compared travel behaviour for different types of urban area and shown significant differences based on area type, the lessons for travel behaviour have been far from clear. In particular, there is a likelihood that part of these observed differences reflect the fact that people choose to live in areas that meet their lifestyle requirements: people who are heavy car users choose locations where parking is easy. Similarly, groups who like to be able to walk or cycle to destinations choose to live in high density areas. Thus, if high density housing is built in order to attract people from suburban areas, they may not display the same travel patterns as those currently living in high density areas due to the personality differences noted in paragraph 2.10.

2.19 One promising approach to understanding the effects of location is to analyse the choices faced by different types of people in different environments. It has been shown that when urban form is characterised in terms of accessibility to various activities, rather than more traditional geographical classifications (e.g. sub-urban), then links to travel behaviour become more apparent (Hansen 1982). A similar finding for rural areas of Scotland was recently demonstrated (DHC 2002).

2.20 However such analysis is very dependent on the variables used to define the choices available to people. Increasing specialisation results in longer trips to more centralised facilities but quantifying this in a meaningful way is problematical since perceptions of people change over time, vary markedly across society, and are dependent on many complex factors and interactions. Despite these many unknowns, it is clear that policies to improve accessibility to defined land uses through transport and land use initiatives can be successful in reducing car dependence (Ecotec 1993).

Soft factors

2.21 In considering the "soft" factors affecting travel choices, the barriers tend to be specific to types of people or types of trip. The barriers relate not so much to the costs and time of travel, or the physical location of activities, but to the way that journeys and modes are perceived and the quality of the travel experience.

Car dependent people

2.22 Whilst travel behaviour and attitudes vary markedly across the population, behavioural classifications do not correlate well with demographic classifications, except for some particular groups such as those with mobility impairments (RAC 1995).

2.23 Psychological attitudes which categorise positive attitudes to car dependence include: independence, variety seeking, privacy, speed, convenience, and control. The people who derive the greatest psychological benefit from car use are the least likely to want to use their cars less. Negative attitudes towards car travel include: stress, discomfort, isolation, lack of exercise, and anti-social. (RAC 1995, Mokhtarian et al 2001, Mackett 2001, Stradling 2002). In these studies there are also many reasons given for car dependence which, although justifying car dependence, have some negative connotations towards car use such as: poor weather, no-choice, and safety considerations.

2.24 Car travel also allows users to express their personality through the type of car they drive and the way they drive it (Stradling 2002). However, the ability to express personality through travel is not unique to car travel. The options to express personality through public transport and walking are certainly more limited than for car travel but are still significant, and many cyclists express their personalities through their choice of cycle and behaviour (DHC 2001) and walkers through their walking style and choice of fashion footwear.

2.25 Reasons given in the Scottish Household Survey (SE 2002b) for not using public transport include: laziness, inconvenience, insufficient reliability of public transport, too slow, too costly, don't feel safe, too crowded, uncomfortable, too much to carry, dislike waiting, filthy, need a car at work, health reasons, difficult access, and infrequent or unavailable service.

2.26 However, stated perceptions of the characteristics of the choices available often differ from the real choices, for a number of reasons such as: a lack of information; because people feel a need to justify their car dependence (SCC 1998), resulting in the phenomenon of 'cognitive dissonance', where drivers adjust their perceptions and attitudes in order to support their current behaviour (e.g. Tertoolen et al, 1998); and sometimes because the commonly used objective measures do not reflect the ways in which consumers perceive the services offered. For example, bus reliability may be characterised in transport models as average excess waiting time, but it is the extreme values that more commonly shape passenger perceptions.

2.27 In terms of person characteristics, some clear patterns can be found in the National Travel Survey. Overall, men still have higher rates of driving licence ownership than women (though this disparity is greatly reduced among younger people), and consequently make more trips as car drivers. On average, women's work trips are shorter than men, partly reflecting the greater incidence of part-time working, and the greater need to fit around school hours and other daily constraints. Women with cars in families with children often exhibit more complex trip chains than their male partners, and are more car dependent as a consequence.

2.28 There has been little published research in the UK on the influence of ethnicity and faith on travel patterns. A current study commissioned by DfT (Social Research Associates 2002) is seeking to: identify patterns of use of public transport by minority ethnic and faith communities and how this is related to use of cars and other modes; analyse how religious, cultural and racial attributes affect use, non-use and times of use, and draw out transport needs and requirements, and propose a range of measures to encourage ridership. Preliminary findings highlights some key differences such as trips to churches and other religious centres, and the unacceptability in some communities of certain groups of women travelling alone.

Trip characteristics and car dependence

2.29 In paragraph 2.9 trips were defined as directed or undirected. For trips that are predominantly directed, the trip purpose is explicit but for predominantly non-directed trips the purpose can be more complex. However, all trips must be worthwhile for the people who make them, otherwise they would not be made. The concept of car dependence implies that part of the value or purpose of a trip is the use of the car. In considering barriers to mode shift within this broad view of travel behaviour, the most important issue is to understand what travel or non-travel choices might substitute for a car based travel choice.

2.30 In most demand modelling, the primary classification by trip purpose is into work and non-work trips. This reflects real behaviour since non-work travel decisions are much more complex than work journeys, allowing a wider range of choices including mode, timing, frequency, destination and route. Work trips have more constraints on timing, frequency, and destination so the mode and route choices available to people are relatively more important (Jones 1991).

2.31 One of the most significant trip characteristics affecting car dependency is trip complexity. Physical (e.g. the length of a walk), cognitive (e.g. knowing where to find the bus stop) and affective effort (e.g. concerns that the bus may be late) are all needed to plan and undertake travel (Wardman et al 2001). Time and money costs have been the focus of efforts to promote mode shift, but affective effort caused by the uncertainty of travelling by public transport can be the greatest barrier particularly where interchange is involved.

2.32 Trip complexity is also evident in the formation of trip chains. Rather than simply travelling between, say, home and work and back, people add in additional stops (to go shopping, visit friends, etc.). Many studies have found that trip chaining is more common and more complex among car users than public transport users, and this makes it difficult to switch from the former to the latter, without a more basic restructuring of travel into simpler trip chains (Hensher and Qeyes 2000, Jones 2002a). This requires considerable cognitive and affective effort.

Complementary factors

2.33 Complementary factors can also be important. For example, widespread availability of mobile phones can allow people to stay in touch with the home or office more easily, increasing the attractiveness of travel (Mokhtarian 2002). They can also help to counter the disruption caused by cancellation of bus/rail services or delays on route, since travellers are able to notify those they are meeting of the delay. A study of Manchester to London business rail travellers in the 1990s showed that one of the main triggers for initially purchasing a mobile phone was to counter the effects of service unreliability (TOI 1998).

2.34 With portable computers and printers, cars are increasingly being used as mobile offices. The advantages of public transport, offering opportunities to work whilst travelling, are still significant, but without continued improvements to facilities on public transport these advantages will be eroded by the developing in-car technology.

2.35 Travel choices need to fit with people's lifestyles and trends in society. Car dependence is an individual and social characteristic, and to reduce car use individual trip types need to be considered (SCC 1998). Leisure trips are growing rapidly and it has been suggested that this is associated with the loss of pleasant local opportunities, as walking and cycling have become less pleasant from busier roads (WA 2000). Transport planning which deals only with the consequences of social change will not be successful (Jones 2002). However barriers to mode shift can be overcome through joint working between communities, transport and non-transport sectors to develop coherent programmes for economic, social, and environmental enhancement (DHC 1999).

Lifestyle issues

2.36 With so many variables involved in travel behaviour, the stability of travel time budgets is both interesting and potentially very important for overcoming car dependency (Noble and Potter 1998). There is evidence both in the UK and other countries that daily travel time budgets have remained stable - at around an hour a day per person - over the last forty years, despite large increases in the volume of travel ¹. This has been possible by switching from slower to faster modes, thereby covering a greater distance in a given time. About 70% of the growth in annual car-based distance per person seems to be associated with a growth in car ownership, but around 30% reflects the increase in door-to-door speeds brought about by improved roads and land use decentralisation to less congested areas (Jones 2002c).

2.37 Around 10,000 miles per year appears to be a pivotal level of car use for English drivers below which most drivers would like to use their cars more and above which they would like to use their cars less (Stradling 2002). At an average car speed of about 40mph, 10,000 miles would account for about two thirds of the average one hour a day travel time budget.

2.38 The implications of stable travel time budgets are that if people can be persuaded to switch from car to other (slower) modes, for example, to walk to their local park for leisure purposes instead of driving to a more distant leisure centre, then they might drive less, not just for leisure, but also for other trip purposes, and so reduce overall annual distances travelled. However, the converse also applies. If there were to be fewer pleasant places to walk then it might be that people would want to use their full travel time budget on car travel resulting in greater annual distances being covered.

The complexity of travel choices

2.39 In the preceding discussion on barriers to mode shift, these have for clarity of presentation been separately characterised under three headings: hard, soft and complementary factors. In practice, however, mode decisions can often involve the traveller taking into account all three of these types of factors at the same time. This is illustrated in Figure 2 for school travel (taken from Jones and Bradshaw, 2000).

Figure 2 Barriers for School Travel

2.40 Parents who drove their children to school were asked why they did so and what changes would be needed for them to feel comfortable letting their children travel by an alternative mode. A range of factors were identified, with around half the parents questioned indicating that they would prefer not to have to chauffeur their children to school. One of the key lessons from the study was that - in many cases - there are multiple barriers preventing people switching from car to other modes. Tackling one or two issues may not be enough: change will only be brought about when all the relevant barriers have been addressed.

2.41 This theme was further developed in Scottish research which showed that many people were making school travel choices which they considered to be less than optimal to obtain what were perceived as wider social benefits (Halden 2002). Overcoming barriers to walking, cycling and public transport use therefore often needs to include both transport and non transport interventions.

Summary of barriers to mode shift

2.42 Table 1 summarises the main barriers to mode shift from car, under the three characteristics of "hard", "soft" and "complementary" factors. As previously noted, the importance of each will vary according to local initial conditions, trip purpose and the characteristics of the travellers.

Table 1 Barriers to Mode Shift