House of Lords - Behaviour Change - Science and Technology Committee

Behaviour Change - Science and Technology Committee Contents

CHAPTER 5: THE GOVERNMENT APPROACH TO CHANGING BEHAVIOUR

5.1. In their evidence to us, the Government emphasised three aspects of their general approach to policy making which impact on the use of behaviour change interventions: a preference for non-regulatory policy tools, engagement with a range of organisations through partnership working, and a greater role for local authorities. In this Chapter, we consider the implications of each of these aspects on the effectiveness of interventions to change behaviour.

An emphasis on non-regulatory interventions

5.2. The Government's emphasis on non-regulatory behaviour change interventions can be traced back to the coalition agreement:

Programme for Government

118

5.3. The Minister, Oliver Letwin MP, echoed this sentiment, noting that "over very many years, governments of different persuasions have assumed that the way you achieve change ... is to legislate and then administer".[119] He said that while "there is a considerable place for legislation and regulation ... where we can achieve an effect that otherwise you would achieve by legislation, either directly or through nudge, without having to regulate, we prefer that route ..."[120] BIT was established in order to help achieve this (see Box 7, page 32 below). Dr Halpern, Head of BIT, agreed that governments have tended to use "a relatively limited menu" of policies to influence behaviour and that this meant that an "additional suite of approaches" which reflected a "more nuanced model of what actually drives behaviour change" had been missed.[121]

BOX 7

The role of the Behavioural Insights Team

BIT, a small team of civil servants and academics, is based in the Cabinet Office and led by Dr David Halpern. BIT has a steering group chaired by Sir Gus O'Donnell, Cabinet Secretary, and works with a variety of external experts, including Professor Richard Thaler, co-author of Nudge. BIT was established with a two-year sunset clause and so will cease to exist in the summer of 2012.

At present the team is working, in particular, on promoting organ donation, smoking cessation, car labelling, food hygiene and charitable giving. The commonality between the projects, according to the Government, is that they do not involve regulating and involve both prompted choice and partnership with the private sector.[122]

Dr Halpern identified the origins of what is now BIT in "deregulatory thrust", in part linked to the Better Regulation Executive.[123] He understands the team's role as raising awareness of "less cognitive, less familiar approaches" as alternatives to legislation, pricing mechanisms and advertising and social marketing.[124] Mr Letwin said that BIT was created in order to help Government departments think about "non-regulatory means of achieving behaviour change".[125]

5.4. Mr Letwin gave four reasons for emphasising a non-regulatory approach: effectiveness, cost-effectiveness, less rigid imposition on individuals and reduced burden on business.[126] We discuss the latter within the section on voluntary agreements (see paragraph 5.20 below).

THE GOVERNMENT'S ARGUMENTS FOR A NON-REGULATORY APPROACH

The effectiveness of non-regulatory and regulatory approaches in isolation

5.5. Occasionally, non-regulatory approaches might be the only reasonable way to achieve behaviour change. Professor John Britton, Director of the UK Centre for Tobacco Control Studies, gave an example: "we cannot legislate to stop people smoking in their home but we can educate and nudge people to change".[127] In some policy areas, particularly crime prevention, legislation might already prohibit certain behaviour and the challenge is to achieve greater compliance. In these cases, further regulation may not be a realistic option.

5.6. Aside from these sorts of circumstances, we were given no examples of significant change in the behaviour of a population having been achieved by non-regulatory measures alone,[128] confirming the view of some witnesses that non-regulatory measures in isolation could have little or no effect and that the most effective means of changing behaviour at the population level was a package of different types of interventions. Findings from our case studies supported this view, as did Defra's work on energy efficient labelling (see Box 6, page 27). Professor Michie also observed:

129

5.7. Similarly, as Professor Michie suggested, regulatory measures may also be less effective when used in isolation rather than in a comprehensive package of interventions. Professor Kelly and Professor Britton agreed that legislation is likely to be more effective when the public understand the reasons behind it; this means that non-regulatory measures should be used as a means of "explaining and promoting the idea beforehand".[130] Professor Kelly cited the "Clunk Click" marketing campaign encouraging people to wear seat belts, which accompanied seat belt legislation, as an example of such a measure.[131] Mr Letwin made the further point that regulatory measures could sometimes have unintended consequences: that sometimes governments "have discovered, to their horror, that the effect that they sought to achieve has not been achieved and that instead some other effect has occurred—perhaps benign, perhaps counterproductive".[132] Professor Kelly agreed and referred to the counterproductive effects of using strict controls on the price and availability of alcohol in Scandinavia (see Box 3, page 21).[133]

5.8. Nudges are a subset of non-regulatory interventions (see Table 1, page 9) and the points made above about non-regulatory interventions apply. Several witnesses told us that, though some nudges reflect experimental evidence about what influences behaviour, they would be unlikely to have a significant effect if used in isolation. Dr Anable, University of Aberdeen, said in relation to reducing car use, for example, that "nudging will achieve nothing ... over the longer term [and] at the bigger scale".[134] Professor Ray Pawson, Professor of Social Research Methodology, University of Leeds, agreed that "sustained behavioural change is difficult to accomplish and requires more than a well aimed 'nudge' in the right direction".[135] Sara Eppel, Head of Defra's Behaviour Change Centre of Excellence, also said in relation to nudging: "I don't speak up its success … you often need some behavioural intervention to make your policy easier to implement, but you may also end up going for the much harder and faster policies at the end of the day".[136]

The cost-effectiveness of non-regulatory interventions

5.9. The Government have suggested that non-regulatory interventions are a more cost-effective way to change behaviour. Mr Baker, for example, said that "... in terms of value for money, the use of nudge and encouragement, apart from being sometimes as effective as regulation, can also be far more cost-effective for the public purse".[137] The MINDSPACE report also suggests that non-regulatory policy tools could lead to better outcomes at a lower cost.[138] In contrast, Professor Marteau and others, writing for the British Medical Journal, noted in relation to nudges that "[lower cost] cannot be assumed because the cost-effectiveness of nudges has not been evaluated".[139]

5.10. Effectiveness is a necessary prerequisite of cost-effectiveness. If an intervention has no effect then it cannot be cost-effective. Given Mr Letwin's view about nudging that "it is of course open to question whether any of this will have any effect whatsoever",[140] we find it surprising that the Government judge that they are in a position to assert that nudging is generally cost-effective.

Respecting the freedom of the individual

5.11. Finally, the Government argue that non-regulatory approaches are more "respectful of the freedom of the individual".[141] We believe that this is misleading. For example, there is a difference between regulation of the individual and regulation of businesses and only the former will tend to restrict the freedom of the individual (see paragraph 2.18 above). Indeed, an argument can be made that regulating businesses might increase the freedom of individuals by preventing businesses from influencing their behaviour and so creating a more neutral environment in which to make choices. We also draw attention to our conclusion in paragraph 2.20 that the Government should be able to explain why an intervention is a necessary and proportionate means of tackling a problem.

THE NEED FOR A RANGE OF INTERVENTIONS

5.12. Whilst the Government have emphasised non-regulatory approaches, Mr Letwin acknowledged that there were circumstances when regulation was appropriate. The Government, he said, were not arguing "that we can substitute behavioural science and behavioural insights for the entire panoply of regulation. It may well be that there are all sorts of domains in which regulatory action is required to make major shifts—either only regulatory action, or regulatory action allied to other things".[142] As we have said (in paragraph 2.4 above), however, the evidence of officials suggests that the understanding that regulation has its place is not fully appreciated throughout Government departments. Ms Eppel, Head of Sustainable Products and Consumers at Defra, for example, told us:

143

5.13. In general, the evidence supports the conclusion that non-regulatory or regulatory measures used in isolation are often not likely to be effective and that usually the most effective means of changing behaviour at a population level is to use a range of policy tools, both regulatory and non-regulatory. Given that many factors may influence behaviour, this conclusion is perhaps unsurprising.

5.14. We welcome efforts by the Government to raise awareness within departments of the importance of understanding behaviour, and the potential this has for the development of more effective and efficient policies. We are concerned, however, that emphasising non-regulatory interventions will lead to policy decisions where the evidence for the effectiveness of other interventions in changing behaviour has not been considered. This would jeopardise the development of evidence-based, effective and cost-effective policies.

5.15. We therefore urge ministers to ensure that policy makers are made aware of the evidence that non-regulatory measures are often not likely to be effective if used in isolation and that evidence regarding the whole range of policy interventions should be considered before they commit to using non-regulatory measures alone.

Partnership Working

5.16. The Government told us that "the involvement of private and Voluntary, Community and Social Enterprise sector organisations will be crucial" when they are trying to change behaviour.[144] Much of the evidence we received highlighted the benefits of partnership working, where Government initiatives are supported by other organisations, and suggested that the Government could do more to work with industry, the third sector and local communities to deliver multi-faceted behaviour change interventions through the most appropriate messengers.[145]

5.17. Numerous reasons were provided in favour of partnership working: that interventions undertaken by local communities and social enterprises were an effective way to change behaviour because those who are affected by an issue are the most likely to be able to solve it;[146] that individuals often respond best to messages about behaviour from those within their local community;[147] that the resources of businesses and the third sector were not time-limited in the same way as Governments, enabling greater consistency in their work;[148] that other sectors have a range of expertise about how to influence behaviour which the Government could take advantage of;[149] and that the third sector were particularly good at harnessing community spirit and were trusted messengers for behaviour change interventions.[150] Rory Sutherland, President of Independent Practitioners in Advertising, noted that, in the business world, some brands also engendered trust in a way that governments often do not[151]—a point borne out by the contribution of businesses during the Change4Life (see Box 8, page 37) programme in communicating messages to consumers.[152]

BOX 8

Change4Life

The Change4Life campaign involved over 200 partners drawn from the voluntary sector, businesses and local government. The campaign also involved over 50,000 local community groups.[153] The Change4Life One Year On report noted that a number of health charities, including Cancer Research UK, Diabetes UK and the British Heart Foundation ran their own campaigns in support of Change4Life. Businesses also supported the movement, for example by providing free gym access, discounted fruit and vegetables and low-cost bikes.[154] A number of witnesses agreed that the campaign had used partnership working effectively.[155] Tim Duffy, Chief Executive of M&C Saatchi, noted that the Change4Life campaign minimised conflict and Paul Kelly, Head of Corporate Affairs at Asda, said that the campaign worked because there was clarity around the role and responsibilities of all of the partners.[156]

5.18. Witnesses from businesses and the third sector observed, however, that partnerships worked most effectively where there was little or no conflict of interests or "internal conflict".[157] Tony Hawkhead, Chief Executive of Groundwork, agreed, suggesting that the Green Deal was a good example of effective partnership working because everybody involved got something out of it (see Box 9, page 37).[158] By contrast, voluntary agreements, which are established between the Government and businesses to change the way in which businesses operate without regulation, were cited as a particularly controversial form of partnership working because of potential conflicts of interest.

BOX 9

The Green Deal

The proposed Green Deal allows businesses to offer energy efficiency improvements to homes, community spaces and businesses at no upfront cost, and recoup payments through a charge in instalments on the energy bill. Mr Hawkhead argued that the Green Deal provides an example of a partnership where there is a clear role for Government, businesses and the third sector and no conflict of interest:

"The role for the Government ... is quite clearly setting a framework and creating a clear vision for how the Green Deal will work: negotiating with private financiers, setting out the legislation. Business's role ... will be quite clearly to install the home insulation … and probably to lead on some of the behaviour change work because they will be in there ... Where the third sector can come in there is the whole area around fuel poverty. The Green Deal will not work for fuel poverty, we will need to use the levy on our fuel bills to try and deal with that … That is where the trusting relationship that the third sector has uniquely in poorer communities can make the difference …"[159]

Voluntary agreements between Government and businesses

EFFECTIVENESS

5.19. Some witnesses from both the business sector and the Government favoured voluntary agreements. Officials from DH cited a report from the Organisation for Economic Co-operation and Development (OECD) which concluded that cooperation between governments and the food industry would be crucial if the problem of obesity were to be tackled successfully.[160] Mr Letwin argued that the Government should try to get businesses to work with them because "one of the very few pieces of extremely strong evidence … is that you can easily create regulations that people will observe in the letter but not in the spirit".[161] Voluntary agreements were not, he said, a means of "handing [businesses] the power". A number of witnesses also referred to work on salt reduction as an example of an effective voluntary agreement.[162] (We note, however, that the salt reduction campaign also publicly named and shamed products particularly high in salt and so its effectiveness cannot be attributed only to the voluntary agreement.)

BURDEN ON BUSINESSES

5.20. The Government also argued that voluntary agreements would be less burdensome on businesses than legislation. The evidence we received from businesses themselves was mixed on this point. Paul Kelly, Head of Corporate Affairs for Asda, agreed that Asda had had only positive experiences of voluntary agreements.[163] In contrast, Justin King, Chief Executive of Sainsbury's, said that voluntary agreements could be burdensome on businesses,[164] and that, furthermore, Sainsbury's were "not against legislation" and would in some instances "positively encourage it" if it was easy for businesses to work with.[165] Mr King emphasised that voluntary agreements tended to be short-term, making them more difficult for businesses to engage with properly, and that although legislation could be burdensome, it was more "consistent for everybody" and tended "to stand more the test of time than a voluntary agreement".[166] Dr Susan Jebb, Chair of the cross-Government expert advisory group on obesity, agreed that voluntary agreements could "be much more onerous" and said that it was an issue that the food network of the responsibility deal was thinking about.[167]

CONFLICT OF INTERESTS

5.21. Many witnesses expressed scepticism about the effectiveness of voluntary agreements because of the overriding commercial interests of businesses. Mr King, for example, said that they were the "refuge of scoundrels", tending to appeal to the lowest common denominator,[168] and were often overtaken by political events.[169] Others went further, suggesting that businesses would never be motivated to do anything which impacts their success. Professor Vivienne Nathanson, Head of Science and Ethics at the British Medical Association, for example, told us about her experience of the alcohol network of the Public Health Responsibility Deal (see Box 10, page 39):

170

Professor Nathanson said that the health organisations on the responsibility deal "believe that, inherently, voluntary agreements won't work and particularly in the alcohol sector. They may have more chance in the food sector—'may' being an important caveat there".[171]

BOX 10

The Public Health Responsibility Deal

The Public Health Responsibility Deal is an example of the Government pursuing voluntary agreements with businesses and other organisations to help achieve policy goals. The Deal was launched by DH to "[tap] into the potential for businesses and other organisations to improve public health and tackle health inequalities through their influence over food, alcohol, physical activity and health in the workplace".[172] It is overseen by a plenary group chaired by the Secretary of State for Health and includes five networks. Four work within a particular area of public health—food, alcohol, physical activity, health at work—and involve representatives of the Government, businesses and health non-governmental organisations (NGOs), who work together to establish "pledges for action".

There is also a fifth network on behaviour change, which

"... seeks to put behavioural science expertise at the disposal of the other networks, enabling them to push the boundaries of their work. The network is also exploring ways in which Responsibility Deal partners can help build the evidence-base for more ground-breaking future work to change behaviour in environments including the retail sector."[173]

Pledges under the deal were published on 15 March 2011 and six health organisations[174] did not sign up. They expressed particular concern with the alcohol network pledges, arguing that they: were too limited with little or no evidence of effectiveness; prioritised industry views; were not specific or measurable and did not indicate what would be a success. They also noted that there had been no commitment made on alternative actions if the pledges did not reduce levels of alcohol-related harm.[175]

5.22. Other witnesses echoed Professor Nathanson's view. Richard Wright, Director of Sensation, Perception and Behaviour at Unilever, told us that "the reality ... is that any business is in business to make money"[176] and that opportunities to influence behaviour will be taken if they are a means to selling more products.[177] Mr King said that decisions taken by Sainsbury's that might discourage consumption of unhealthy products, for example removing confectionery from their checkouts in some stores, were taken when they were what the customer wanted rather than on the basis of any judgement about improving the health of consumers.[178] Mr Letwin indicated a similar view when he said that working with businesses through voluntary agreements involved thinking about whether the agreement was "possibly in their commercial interest".[179]

EVALUATION, TIMELINES AND REGULATION

5.23. A number of witnesses were concerned that mechanisms should be in place to ensure that voluntary agreements were subject to rigorous evaluation, with clear outcome measures and timelines. Dr Jebb said that "it is ... vital that public health bodies and institutions are charged with monitoring and evaluating the success or otherwise of ... delivery".[180] Professor Lindsay Davies, President of the Faculty of Public Health, said that voluntary agreements were an "experiment" and so should not be allowed to "drift on and on and on as a substitute for ... taking harder action, because the obesity epidemic can't wait",[181] a point also made by Professor Nathanson.[182] Other witnesses emphasised the importance of timelines, arguing that if agreements could not be reached or were not effective, then the Government had to be prepared to regulate. Professor Nathanson said, for example, that the Government should be prepared to say "if we don't get a sufficiently challenging-to-the-industry ... agreement, then we would be prepared to regulate on the areas that the voluntary agreement should cover, as well as the areas that will only happen through regulation".[183] Mr Letwin acknowledged this point when he said "we may need to regulate [businesses] and not merely do deals with them".[184]

5.24. Anne Milton MP, Parliamentary Under Secretary of State for Public Health, provided some reassurance in relation to these concerns:

185

As we note in Box 10 (page 39), some organisations do not feel that these measures were present in the published Public Health Responsibility Deal. The pledge about reducing obesity made by the food network of the Deal does not set outcomes relating to changes in behaviour and does not reflect the available evidence about how to tackle the problem of obesity (see paragraph 7.20 below).

5.25. The involvement of other organisations to support the Government's behaviour change initiatives may provide valuable opportunities to improve the effectiveness of behaviour change interventions, in particular by allowing a range of messengers to be used to deliver them. We welcome the Government's intention to use such collaborations.

5.26. However, we have major doubts about the effectiveness of voluntary agreements with commercial organisations, in particular where there are potential conflicts of interest. Where voluntary agreements are made, we recommend that the following principles should be applied in order to ensure that they achieve their purpose:

The Government should specify clearly what they want businesses to do based on the evidence about how to change behaviour, and what steps they will take to achieve the same result if voluntary agreements are not forthcoming, or prove ineffective.
Voluntary agreements should be rigorously and independently evaluated against measurable and time-limited outcomes.

5.27. Given that these principles do not appear to have been applied consistently to the Public Health Responsibility Deal Network, we urge DH, in particular, to ensure that these principles are followed when negotiating further voluntary agreements. In relation to the current agreements, we recommend that DH should state for each pledge what outcomes are expected and when, and provide details of what steps they will take if the agreements are not effective at the end of the stated period.

The role of local authorities

5.28. Mr Letwin told us that the Government "are very determined to ... decentralise power and to leave local communities and local governments as free as possible to make their own decisions about how they do things".[186] In our two case study policy areas, the role of local authorities in delivering behaviour change interventions has been emphasised in recently published white papers.[187]

BENEFITS OF DECENTRALISATION

5.29. Witnesses described two major benefits of a local approach to changing behaviour. First, different local areas have different local needs and so interventions should reflect these differences.[188] In relation to sustainable transport, much of our evidence agreed that local authorities were best placed to design behaviour change interventions because they were most qualified to assess the need for, and implement, interventions.[189] Similarly, witnesses were mostly positive about the proposed reforms to public health which would move Directors of Public Health into local authorities, suggesting that they would provide opportunities to coordinate behaviour change activity across a range of areas.[190] Professor Dame Sally Davies noted that this shift should mean that local authorities will start to "look at all the things they do—education, planning, cycling paths, transport … through a health lens as well as through the cost lens and the service lens".[191] Dr Frank Atherton, Chairman of the Association of Public Health Directors, told us that "Directors of Public Health [were] ... universally welcoming the move into local authorities, because that's where the levers of change actually exist".[192]

5.30. Secondly, some witnesses suggested that the devolution to local authorities of responsibility for designing and implementing interventions would provide an opportunity to help build the evidence-base for the effectiveness of population level interventions.[193] Mr Letwin said that, in principle, the decentralisation of power should provide a "rich field" for evidence generation—although he acknowledged that it would be important to ensure that mechanisms were in place to take advantage of it.[194]

POSSIBLE PROBLEMS

5.31. Some witnesses expressed doubts about decentralisation however. They questioned whether there were the requisite levels of skill in designing and evaluating interventions at a local level, or adequate mechanisms in place for the dissemination of knowledge, to allow the Government to make the best use of what is learnt about the effectiveness of interventions. In relation to the use of evidence, Paul Sacher, Chief Research and Development Office for MEND Central, and Zoe Hellman, Company Dietician for Weight Watchers, said that commissioners of weight management programmes within Primary Care Trusts did not review evidence accurately and appeared not to understand the most important measures of effectiveness.[195] A similar view was reflected in the evidence we received in relation to our case study on reducing car use to limit carbon emissions (see paragraphs 7.42-43 below). In addition, witnesses suggested that local authorities may not have the range of skills and resources necessary to interpret the available evidence.[196]

5.32. Witnesses also expressed concern that devolving responsibility for behaviour change interventions in some policy areas might have a detrimental effect on evaluation.[197] The National Obesity Observatory noted, for example, in relation to the commissioning of weight loss interventions, that:

198

5.33. The Government acknowledged the importance of effective evaluation of local initiatives. Mr Letwin told us that the Government had discussed setting up some research apparatus at low cost to "investigate what had been done by one local government in one place, and enable it to be evaluated and transmitted to other local governments in other places".[199] Mr Dowie said the DfT recognised that it had "a responsibility through the ... sustainable transport fund to ensure there is a proper evaluation framework in place".[200] To that end, the department has "published impact evaluation guidance aimed at scheme promoters and evaluation practitioners to help them choose an evaluation approach which is best suited to their evidence needs and helps them design an evaluation which enables the observed impacts to be attributed to the scheme". They have also developed a framework "for evaluating schemes aimed at encouraging sustainable and active travel behaviours".[201]

5.34. In relation to sharing knowledge at a local level, witnesses who had designed and commissioned local interventions told us that there was no mechanism for the broader dissemination of the lessons learnt from their behaviour change programmes.[202] Robin Gargrave from Central YMCA said that while there was some informal knowledge exchange among local organisations, there were no "national data that shows the direction of travel, and also indicates what's working, what's not working and why". He added: "anything that the Government can do to help facilitate that would be most welcome".[203] The DfT stated in its Sustainable Transport white paper however that it would be "stepping back from monitoring" and that "the Local Government Association and local authorities themselves will be responsible for spreading best practice, sharing what works and developing a framework that improves capability across the local transport spectrum".[204]

5.35. Finally, a number of witnesses cautioned that localism should not detract from the important role central government still had to play. Lynn Sloman, Director of Transport for Quality of Life, said for example that the DfT should continue to foster experimentation and that it should "provide a consistent, long-term direction of travel so that the local authorities and everybody else knows where they are".[205] This was supported by other witnesses in relation to public health.[206]

5.36. Although decentralising responsibility may provide a useful opportunity to tailor local behaviour change initiatives and to help build the evidence-base for applied behaviour change research at the population level, steps should be taken to ensure that interventions are evidence-based and properly evaluated. To this end, we recommend that the Government:

produce guidance for local authorities on how to use evidence effectively to design, commission and evaluate interventions and on the need to involve experts in the design and evaluation process (see paragraphs 4.25 and 6.3), and provide advice on how to best use the tendering process to ensure value for money;
take steps to ensure that evaluation of interventions, including data collection and reporting of behaviour change outcomes, across local areas is of sufficiently high quality to allow comparisons and analysis;
take steps to ensure that what is learnt by a local government in one place can be readily transmitted to other local governments; and
provide funding only for those schemes which are based on sound evidence. Demonstration of rigorous evaluation and contribution to the evidence-base should be a requirement for future funding for behaviour change interventions.

118 BC 114. Back

119 Q 703. Back

120 Q 706. Back

121 Ibid. Back

122 Q 703. Back

123 Q 19. Back

124 Q 8. Back

125 Q 703. Back

126 Q 703. Back

127 Q 153. There has of course been legislation to encourage people to stop smoking, such as the Health Act 2006 which banned smoking in enclosed public places. This legislation appears to have had an effect on the numbers of individuals who smoke, and consequently who smoke in their homes. Professor Britton was instead referring to a direct ban on smoking in homes. Back

128 We were given the example of the relatively small changes made to letters sent out by Her Majesty's Revenue and Customs (HMRC) which seem to have had a substantial impact on the levels of response to tax collection letters. We note however that HMRC cannot be sure that the increased response was wholly a result of changes to letters, as other changes to the tax collection process were made simultaneously (BC 114). Back

129 Q 129. Back