2. MECHANISMS FOR BEHAVIOUR CHANGE
2.1 Evaluating policy initiatives in terms of success or failure is problematic. What constitutes success is dependent on the criteria and scope of the evaluation: what appears successful along one criterion can seem less persuasive along others. For example, campaigns for selling light bulbs based on price incentives and marketing have been successful in terms of sales (Munasinghe et al., 2009), and total energy consumption from standard light bulbs began to drop in 2007 (see Figure 2.1). However, the energy savings made from straight substitutions between standard and energy-saving light bulbs is partly offset by the increasing energy consumption from halogen light bulbs (see the disparity between the red and dark blue lines in the Figure). In this case, purchase behaviour is changed (success) but environmental impact is not significantly reduced (failure). While it could reasonably be argued that energy consumption through lighting would have risen at a greater rate were it not for sales of lower energy bulbs, the more important question is why do people appear to consume more light? Answers might be found in accounts of emerging social tastes for ambient low-lighting that has resulted in there being many more light bulbs per room in European countries than was the case in the past (Wilhite and Lutzenhiser, 1999). In other words, behaviour change mechanisms need to be combined so as to address the use of goods and services as well as shifting consumption patterns toward more sustainable substitute products.
Figure 2.1. Total electricity consumption by household domestic appliances 1970 to 2009
Source: Department of Energy and Climate Change (2010) 'Energy Consumption in the UK'.
2.2 It is useful to summarize some of the key behaviour change mechanisms available to policy makers in order to identify what combinations are most effective. Such mechanisms tend to address at least one context in which behaviour might be changed:
- The individual, which refers to initiatives that focus on influencing the attitudes of individual consumers so as to change their behaviours and choices;
- The social, which refers to the social norms, cultural conventions and shared understandings of consumer practices;
- The material, which refers to the objects, technologies and infrastructures that both enable and constrain ways of behaving.
2.3 These three contexts represent a good starting point for isolating behaviour change mechanisms and better understanding the rationale that underpins them.
The individual context
2.4 The individual context broadly covers initiatives that primarily seek to change the attitudes and choices of consumers in ways that encourage more sustainable behaviours. Most prominent are economic incentives. Increasing the monetary cost of environmentally damaging activities or offering financial incentives to undertake less environmentally damaging behaviours (such as air travel and home insulation) is a common, if politically problematic, mechanism. Other than the difficulties of interfering with market mechanisms, two further problems can be identified. Incentives do not necessarily foster long term changes in behaviour. For example, a study conducted as part of the Danish Environmental Research programme gave free one-month bus passes to 400 car-driving commuters. Following a significant increase in the number of journeys made by bus rather than car, the number of bus journeys declined and people largely reverted back to driving their cars after the trial period (Thogerson & Moller 2008). Second, monetary penalties or disincentives can work to legitimate the behaviour being discouraged. For example, when parents at a kibbutz school were fined for turning up late to collect their children the number of late pick-ups rose, because parents felt that they were paying for the right to be late (Gneezy and Rustichini 2000).
2.5 Offering and actively promoting environmentally friendly alternatives to unsustainable practices is another mechanism that addresses individual consumer choices. This can involve 'nudging' choices by making low-carbon goods and services easier to find and adopt. For example, when Tesco moved their organic products from a specialist area and integrated them with all the other produce, it became easier for consumers to choose organic, and sales of organic produce rose by 35% (Munasinghe et al. 2009). Another form of nudging is default opt-ins for discretionary charges to finance carbon off-setting schemes (such as customers having to opt out of a discretionary charge associated with a product, say an airline ticket, for funding carbon off-setting programmes).
2.6 Informing the consumer, thus not trying to overtly manipulate (through price incentives and nudging) but to change attitudes through education is a third mechanism. In addition to product labelling (see Box 2.1), marketing campaigns are often used to raise awareness of environmentally problematic activities. Such approaches can be expensive in relation to their effectiveness. For example, during the 1980s Californian utility companies spent $200 million to advertise residential energy efficiency measures but household energy use changed only marginally. In one case, the Pacific Gas and Electricity Company spent more money advertising the benefits of home insulation in California than it would have cost to install the insulation directly into the homes of every person it was trying to reach (McKenzie-Mohr 2000).
Box 2.1. The Casino Carbon Index (case study 28, Appendix 1)
A French labelling scheme advising consumers of the carbon footprint associated with the purchase of over 200 products. In the first two years of the scheme over 20 metric tonnes of CO2 equivalent are calculated to have been saved as a result of customers substituting higher for lower carbon products. This represents some success but it is important to recognise that this is the level of annual CO2 emissions associated with the energy use in two homes in the United States. It should be noted however that the impacts of carbon labelling would not be expected to show significant returns before all products are fully and consistently labelled.
The general challenge for carbon labelling is to ensure that the information contained within labels is meaningful, easy to understand, standardised across products, and that consumers are motivated to want to take action. So while this could be transferred to the Scottish context; it is not immediately clear that labelling is an effective was of reducing individual carbon footprints.
2.7 Information campaigns are most effective when targeted at particular groups. Segmentation models (such as those developed by DEFRA) make it possible to direct marketing campaigns at groups according to their attitudes and current behaviours. Targeted marketing also opens the opportunity to develop campaigns attached to values that are not necessarily pro-environmental but which, nevertheless, still foster more sustainable behaviours (see Box 2.2).
Box 2.2. Save the Crabs - Then Eat 'Em (case study 19, Appendix 1)
A campaign in Virginia and the Greater Washington DC area persuaded homeowners to fertilise their lawns differently and so reduce the nutrient pollution flowing into the nearby Chesapeake Bay. The information campaign did not position this as an environmental issue, rather it stressed such actions were necessary to ensure that seafood yields, important for the local economy and therefore local communities and households, did not diminish.
It cost $550,000 to initiate and 46% of those exposed to the campaign planned to change their behaviour accordingly compared to 40% of those not exposed. While reported as a success, this campaign only achieved a marginal increase of intentions to change behaviour and appears to be quite expensive. More interesting is the potential to use other more localised and directly relevant issues to mobilise more sustainable behaviours.
The social context
2.8 Many cultural conventions and social norms underpin consumer behaviour. For example, studies consistently reveal shared understandings of a proper meal being that which constitutes meat, vegetables and a carbohydrate (Mitchell, 1999). Changing cultural conventions is difficult because they are entrenched in ways of life and also vary across social groups. Studies also show that social groups favour different variations of the 'proper meal' according to gender, age, household composition and socio-economic status (Bennett et al, 2009). Addressing the social contexts of consumer behaviour involves attempting to shift the cultural conventions and social norms that underpin different activities. This is both difficult and problematic because it requires shifting the foci of initiatives away from individual consumer decisions and toward shaping and intervening in the shared behaviours of social groups.
2.9 Social institutions, such as the workplace and schools, represent social contexts through which people learn, come to understand and habituate certain consumption behaviours, and can be targeted for behaviour change. This can involve initiatives that seek to better coordinate the temporal rhythms of everyday activities, such as hours of working, and encouraging employee car sharing. Households or families are another social institution through which behaviours can be changed, particularly by focusing on moments of transition. A scheme operated by the Centre Area Transportation Authority ( CATA) in Pennsylvania (case study 16) is a good example. It targeted households who had recently moved to the neighbourhood, offering free bus passes for a trial period, and information about bus stops and services. Importantly, rather than simply seek to shift individual attitudes this scheme focused on moments of life-course transition (in this case moving home; others could include marriage, becoming a parent, leaving the parental home, and so on) in order to reframe the social context in which habitual practices are located. While CATA did not directly evaluate the success of the scheme, they did report a sustained and significant increase in passenger numbers. Using social institutions to re-set social norms can also lead to wholesale shifts in cultural conventions (see Box 2.3). As the Cool Biz example demonstrates, governments have a direct opportunity to re-frame the social context of behaviour through attention to the building stock and practices of its institutions.
Box 2.3. Cool Biz (case study 22, Appendix 1)
A campaign initiated by the Japanese Ministry of Environment aimed at reducing energy use in buildings by setting air conditioners at no lower than 28oC throughout the summer months. To make this more comfortable for workers a new dress code was implemented. This involved removing ties and blazers and wearing clothing made from 'breathable' fibres. Not only was the regulation of air conditioning achieved but the social norms surrounding workplace clothing attire changed. The Ministry estimated a 1.14 million ton reduction of CO2 emissions (equivalent to the annual emissions associated with the average energy use of 85,000 US homes), although sales of neckties declined by 36%! While the ministry's figures need to be interpreted with caution, this intervention highlights the potential for governments to shift behaviours through attention to the buildings and practices of its institutions. It is also reasonable to expect that interventions such as this are relatively inexpensive and could perhaps even save money thanks to lower energy bills.
2.10 A second mechanism for addressing the social contexts of behaviour change is cultural tastes, which by definition are shared (otherwise it would be impossible to distinguish good from bad or identify personal tastes - see Gronow 1997). Here the focus is less on influencing the decision making of the individual but generating shared cultural understandings of what is fashionable and appropriate. In Europe and the US, for example, celebrity chefs have made a significant difference to the eating aspirations of large sectors of the population. But celebrity endorsement is not the only way of influencing cultural tastes. Often, early adopters of new styles can set the trend. Think of the gastronomes who, in the 1970s, championed eating lamb and duck when it is pink rather than cooked through, creating a demand for a way of cooking that has since become the mark of good taste in restaurants (Warde 2009). The 'New Nordic Diet' (case study 2) is attempting to use such mechanisms to change eating tastes in Denmark, using a celebrity chef and media exposure, not to mention the production of recipe books and cookery lessons, to shift the cultural understandings of what constitutes 'good Danish food'. While this (expensive - circa £10 million) campaign is in its early stages, concerns about trying to 'dictate' tastes and reactions against the diet as 'not being for people like me' demonstrate the difficulties of trying to 'design good taste' (see Bere and Brug, 2009, for more details).
2.11 Community-based initiatives often seek to change consumer behaviour by influencing social norms (see Box 2.4). Such schemes tend to focus on the importance of social networks for circulating information and expectations regarding appropriate behaviours. They seek to support individual efforts to live more sustainably just as they might operate to 'police' tastes and so govern both what it is that 'we' like to consume as well as what 'we' understand to be bad or appropriate conduct.
Box 2.4. Backyard Composting (case study 20, Appendix 1)
In 2000, Nova Scotia province in Canada introduced an initiative to encourage composting by persuading households who already engaged in this activity to put a sign on their bin to make it more visible to their neighbours. They were also asked to speak to their neighbours and friends about composting and dispel common fears that it was unpleasant. Seven months after the trial began 80% of those surveyed who had not previously composted had taken up the activity.
This is an example of a very cheap and simple initiative that uses social norms to trigger behaviour change, and with impressive results. Such schemes have much potential to be applied in different locations due to their sensitivity to local issues and cultures, and the Scottish Government's Climate Challenge Fund represents a useful vehicle for facilitating such activities.
The material context
2.12 The material context refers to the technologies, infrastructures and material design of the goods that almost all forms of consumption involve or rely upon. Infrastructures and technologies permit particular forms of consumption, whether that is travel, cooking, bathing or surfing the internet, but also lock people into behaviours that can be very difficult to change. The freezer, for example, has developed alongside an entire frozen-food infrastructure that includes the changing design and use of houses and kitchens, the development of the out-of-town supermarket, and subsequent decline of local food stores, which renders household food provisioning without a freezer increasingly difficult (Shove and Southerton, 2000). Shifting infrastructural arrangements are, however, often essential to facilitate individual decision making and shift social norms toward more sustainable behaviours. Recycling in the UK increased significantly once the appropriate waste infrastructure was in place through the provision of recycling bins/boxes and curb-side collection facilities. Such interventions in material infrastructures not only create the conditions for new habits to emerge but have the potential to lock people into sustained environmentally friendly behaviours (see box 2.5).
2.13 The way that objects are designed and combined can shape the way that they are. Classic examples are traffic calming measures such as sleeping policemen which force drivers to slow down or oversized hotel keyfobs which encourage guests to hand keys in at reception when leaving the hotel, reducing the number of lost keys in the process (Latour, 1991). Attention to design of goods and services is therefore a mechanism that can nudge behaviours in particular directions (see Box 2.6).
Box 2.5. BRT and CicloRuta (case study 9, Appendix 1)
The city of Bogotá, Colombia has made substantial investments in a Bus Rapid Transit ( BRT) system and an extensive cycle network. The BRT system involves: dedicated bus lanes with feeder routes into the main system; dedicated terminals that allow for quick and easy boarding and ticketing; frequent and high capacity buses and; an organisation structure that allows for flexible and reliable scheduling. Consequently, bus travel becomes a quick and reliable alternative to car travel and, as a result, it is estimated that over 287,000 tons of CO2 are saved annually - equivalent to the annual emissions associated with the energy use of 22,158 homes in the USA. However, it is not a cheap intervention with the cost of the first phase of implementation having cost $297 million (although it is estimated that $25 million will be made in carbon credits by 2012). The BRT has gone hand-in-hand with the expansion of Bogotá's cycle network which has worked to make cycling safer and easier and in turn, the percentage of residents using bicycles doubled between 2000 and 2007 with an estimated annual saving of 6,500 tons of CO2. Again, this is not a cheap intervention with an initial investment cost of $50.25 million but taken together, BRT and CicloRuta highlight the potential for interventions in material infrastructures to facilitate less environmentally damaging travel behaviours.
Box 2.6. Traffic Signal Timing (case study 25, Appendix 1)
In Portland, Oregon the timings of traffic signals were changed in order to reduce the petrol used by motorists when idling or accelerating. This relatively straightforward intervention is reported to save 15,460 tons of CO2 emissions each year (equivalent to the emissions associated with the annual energy use of 1,194 US households). This intervention is particularly interesting insofar as behaviour is changed through the re-scripting of the practice of driving and carbon emissions reduced without individuals having to make the choice to do so or necessarily be aware that they are changing their behaviour. It cost $533,000 to initiate for which the returns were significant given: 1) the CO2 reductions that it is reported to be responsible for and; 2) figures that estimate annual financial savings for motorists of over $4,000,000.