Why try to increase volunteering?

The idea of co-production is that public services and citizens contribute jointly to deliver positive social outcomes and sympathetic environments in which people can have creative, productive and fulfilling lives (see, for example, Boyle and Harris 2009). Underlying recent interest in co-production across the political spectrum is a belief that 60 years of state provision of welfare, however well-intentioned, has eaten away at citizens’ capacity and desire for mutuality and for self-help. These are not new ideas. William Beveridge, seen by many as the founder of the British welfare state, argued in 1948 for the importance of the ‘mutual aid motive in action’ and the ‘philanthropic motive in action’ (Beveridge 1948). He was ‘keen to recognize that the state could and should have a vital and proactive role in developing policy frameworks to nurture both solidaristic and sympathetic human motivations and their capacities for expression’ (Kendall 2009: 3). In Beveridge's view, the state's role was to exercise self-restraint, and voluntarism would balance the power and dominance of the state.

There have been well-documented shifts in the types of civic activity in which people engage; preferences for new individualized and consumer-based expressions of solidarity and philanthropy have arguably overtaken traditional forms such as volunteering. So, new activities such as boycotting or ‘buycotting’ consumer goods on political or ethical grounds, or wearing badges and wristbands for causes, are on the rise internationally (Micheletti 2010). However, there remains a firm interest in getting citizens to participate in co-production through active volunteering. For many policy outcomes, action by the state cannot substitute for civic acts. For example, local efforts by public sector bodies to clear up unsightly bulky refuse (like mattresses and fridges) are not a substitute for citizens choosing not to dispose of unwanted items by dumping, and preferably by citizens also assisting with the process through clean-up days and similar exercises. Volunteering is also not easily replaced by other civic acts, such as ethical buying, which may be driven by similar sympathetic human motivations, and towards similar solidaristic ends, but which do not offer the potential for co-produced outcomes.

In this chapter we review the evidence on promoting volunteering. The chapter contains summaries of the various options that have been tried and a discussion of what a nudge strategy could offer. It examines a particular case and reports from a design experiment in an English local authority which attempted to facilitate volunteering by asking citizens who complain to a local authority telephone call centre to do other civic-minded acts. What do these findings imply for the challenge of promoting volunteering? By changing the choice architecture can you turn complainers to volunteers?

What do we already know? How much volunteering already exists?

There have been numerous policy interventions by public institutions, in the UK and internationally, to encourage volunteering. In the ten years from 1997/1998 to 2007/2008, UK central government investment in adult volunteering increased five-fold (Das-Gupta 2008). Youth volunteering policy has seen marketing of the V brand, and the creation of a national youth citizenship scheme. Beyond the UK, 2011 was the United Nations’ and the European Union's International Year of Volunteering.

Is Britain different from elsewhere? In the United States, data from the US Bureau of Labor Statistics show that 26.8 per cent of the population volunteered through, or for, an organization at least once between September 2008 and September 2009 (Corporation for National and Community Service 2010). These figures were down slightly from 27.4 per cent in 2002 and highs of 28.8 per cent between 2003 and 2005, but an increase from 26.2 per cent in 2007 and 26.4 per cent in 2008 (Corporation for National and Community Service 2006). Despite these dips, the overall rate of volunteering in the United States is similar to that in the United Kingdom, where 28 per cent of people were engaged in regular formal volunteering in 2009, with no change between 2001 and 2009 (Department for Communities and Local Government 2009). Occasional formal volunteering showed an increase from 39 per cent in 2001 to 43 per cent in 2009. There were decreases in occasional informal volunteering (from 67 per cent in 2001 to 57 per cent in 2009) and regular informal volunteering (from 37 per cent in 2003 to 33 per cent in 2009) (Department for Communities and Local Government 2009). It is important to be cautious about these results as they are based on survey reports and may not be accurate measures of actual behaviour.

Reliable and consistent trend data in nearly all countries are not longitudinal and therefore do not show how many drop out and how many join in. Given a wide range in the lengths of time spent volunteering, these figures may represent many more individuals contributing over the period than is apparent. Volunteering levels of between 26 and 57 per cent indicate an arguably healthy base of citizen activity, but with potential to drive up levels particularly on regular formal volunteering.

The literature – how can volunteering be increased?

It is not immediately clear what else, if anything, could be done to raise volunteering rates. However, there are some clues in the literature. Verba, Schlozman and Brady's study of facilitators of civic voluntarism outlines a civic voluntarism model, which sets out three drivers to stimulate volunteering: capacity; motivation and mobilization (being asked) (Verba, Schlozman and Brady 1995). In this model, mobilizing is effective where: ‘people are resource-rich, have plenty of free time and have a strong sense of efficacy’ or interest already (Pattie and Seyd 2003: 446). Mobilization is predicated on sufficient capacity and motivation.

How could these ideas be translated into possible nudges to increase volunteering? The presumptions in the civic voluntarism model offered some interesting areas to test using the idea of nudge. Was there an untapped pool of unasked, skilled and keen people who could be more effectively mobilized than through previous interventions? Could the predicates be reversed or bypassed? Was it necessary to build up or rely on pre-existing capacity and motivation, or could mobilization through a nudge develop capacity and motivation? Mobilizing is a classic nudge strategy, as it presumes citizens might already be able and willing but not activated, or could become more able and willing if activated.

Someone or something needs to be the mobilizer and do the asking. In the original civic voluntarism model, mobilization is done through ‘networks of recruitment’ (Verba 1995: 3) and interpersonal relationships, primarily informal recruitment through friends, acquaintances and colleagues, but also political parties. The workplace, churches and voluntary associations act as the ‘loc[i] of recruitment’ (Verba 1995: 144). Verba et al. discuss the ‘non-political secondary institutions of adult life – the workplace, voluntary associations, or church’ (Verba 1995: 369). These institutions are crucial, mostly insofar as the settings in which mobilization happens are where colleagues gather and have conversations and where ‘psychological engagement’ (motivation) is cultivated through debate and other cues and hooks. These settings are also where transferable skills (capacity) are acquired. Formal requests by institutions play only a small role. Later work by Lowndes, Pratchett and Stoker extends the model and sets out an enhanced formal role for institutions as mobilizers in their own right, and as mobilizers of a wider group of citizens than the institutions’ members or employees (Lowndes, Pratchett and Stoker 2006). They propose the CLEAR framework (see Table 4.1): Can do – that is, have the resources and knowledge to participate; Like to – that is, have a sense of attachment that reinforces participation; Enabled to – that is, are provided with the opportunity for participation; Asked to – that is, are mobilized by official bodies or voluntary groups; and Responded to – that is, see evidence that their views have been considered.

Existing practice by local government institutions as official bodies suggests a limited and not fully effective mobilizing role. Local government advertises local events and activities, and most contribute funding to Volunteer Bureaux. Data produced for 2007/ 2008 shows that 3 per cent of adults found out about volunteering opportunities from Volunteer Bureaux, under 10 per cent through advertising and over half by word of mouth (Agur and Low 2009: 166). At the same time, each local government body conducts thousands of direct transactions with citizens and service users, but does not use these contact points as mobilizing opportunities. Moreover, a high proportion of the transactions are initiated by the citizen, many because something has gone wrong with public services. The citizen or service user contacts a local authority with a one-way request for the public body to solve a problem, for example a person rings a customer contact centre direct to request that street cleaners come to their neighbourhood and remove litter from streets.

A civic request is not made of residents during these routine service interactions. There may well be a reciprocal civic request being made of the same citizens by the same institution, for example, to join in a neighbourhood clean-up day. But the institution's request is often made by a different department, at a different time, through an indirect route (for example, posters at a local school), and usually generates extremely low returns. When a citizen calls a local authority contact centre, the presumption by the authority appears to be that their behaviour is not civic. Yet the person being treated as a customer may be acting as a good citizen, hoping their phone call will lead to neighbourhood improvements, cleaner streets or better neighbourhood relations. Even if a call is not orientated towards improving public goods, or influencing the institution more generally, there is still a possibility that the caller would be willing to consider making this sort of contribution, if they were asked. Therefore, the potential for co-production is often lost.

What is the intervention? The nudge experiment

When citizens do get mobilized through being upset about the streets not being cleaned or bins not been emptied, the public authority can use such a situation to nudge citizens into more civic-minded behaviour. The nudge experiment was undertaken by the University of Manchester working in partnership with a local authority. The idea was to see if voluntary activity could be mobilized more effectively. The partners agreed to change the choice architecture: the default settings of the institution – the local council in this case – were changed from one that assumed that citizens have a largely passive rule to one where they have a fuller relationship with public authorities and their communities, and would want to become volunteers. The original default setting was that routine service transactions were with service users as passive complainers and consumers. This was altered to a new default setting that any interaction was with potentially active citizens. This also required changes in the organization's cultural default settings. Seeing the callers as complaining customers suggested that the council anticipated customer hostility to being asked to volunteer. Regarding people as both consumers and potential co-producers required the council to presume people would be comfortable with being asked.

To test this out, the experiment was focused on promoting voluntary activity in two neighbourhoods in a Blackburn. Callers reporting a problem or making a query to a local authority customer contact centre were asked if they wanted to get more involved in the neighbourhood. Typical reasons for calling were complaints about environmental services. After the query or complaint had been dealt with, citizens from those neighbourhoods who telephoned the contact centre were asked: ‘We are currently promoting civic awareness in [your neighbourhood] and are looking for people to get involved in improving the area. We want to encourage people to take action on community issues in the area. Would you be interested in finding out more?'

Citizens who were identified as interested were split into two groups. Half of the potential volunteers were allocated randomly to an intervention group which was then encouraged – using a variety of approaches – to take further steps such as joining a local group, becoming volunteers or changing their environmental behaviour. The other half of the volunteers were allocated to a comparison group who were sent an information pack on opportunities for civic participation in the neighbourhood. The information pack contained information that was already publicly available and marketed to residents, and, as such, was similar to a placebo treatment.

There were two waves of the experiment. In the first, existing opportunities to participate were unchanged; what was new was the proactive approach by the contact centre, combined with a local neighbourhood officer, providing information and encouragement to the intervention group. In the second, the follow-up contact with the neighbourhood officer was supplemented by the creation of new volunteering opportunities for the intervention group. All participants were interviewed at the start of the project, and again eight weeks later, and were asked about civic activity and attitudes. These new volunteering activities were then offered to the comparison group after the data collection had been completed.

The first phase occurred in April–July 2008 in one neighbourhood. During this first phase, the intervention was tweaked in response to the research findings. Using the lessons from the first wave, the experiment was repeated in a second neighbourhood. The second phase took place between February and August 2009 in a different neighbourhood. A steering group for the project included the manager of the contact centre, the head of customer services and three members of the council's neighbourhood team, a housing association manager, a policy manager and the researchers. The steering group met regularly to reflect on the emerging findings from the research and decide whether the intervention should be tweaked to improve its effectiveness. The aim was to identify the most effective and appropriate way to design the intervention, both in the contact centre and in the neighbourhood.

The actual research took place in Blackburn in northwest England. The first neighbourhood was a residential area, relatively deprived in national terms, but fairly affluent compared to surrounding areas of the town. The housing was mostly owner-occupied or privately rented. The area had a high ethnic minority population, mostly of Pakistani and Bangladeshi family heritage. The area was chosen as a suitable for this experiment because it had an active and welcoming community association together with a range of other potential activities. The second neighbourhood was less affluent, within the lowest 5 per cent of neighbourhoods in England, and was among the most deprived in the borough. It was a predominantly white, former council estate, now managed by a housing association. Most of the properties were social housing, with pockets of owner-occupation. Over 40 per cent of households claimed state benefits. The experiment included all telephone callers from the neighbourhood to the council's contact centre complaining about cleansing, environment or neighbourhood services. During the second wave, calls about council tax and housing benefits were also included.

The results

There were two critical successes that suggest the potential for a nudge strategy to change complainers into volunteers. The first was that citizens welcomed the change in the choice architecture. From the start of the research, there had been some concern from local authority staff that there would be an adverse reaction from citizens to this change of approach. In particular, the contact centre managers were worried that people phoning to report problems or make complaints might be irritated by being invited to be proactive on neighbourhood issues; those reporting a problem with a local service would be angry at being asked to take action themselves. Members of staff were not convinced that people would welcome a change in the nature of the relationship.

These concerns over the change in the default setting were not borne out by citizens’ responses. The research tested the assumptions about citizens’ preferences. The doubts of members of staff proved to be unfounded among those citizens who took part. People were happy to be mobilized by public institutions. Citizens were generally supportive, with 92 per cent across both intervention and comparison groups agreeing that the council should encourage callers to get more involved. One person spoke about the importance of the council working in partnership with members of the public:

They [the council] are restricted in what they can do. They should work with the people to get to the bottom of it rather than tell people to ring somewhere else. We raise the same issue repeatedly and no records are kept. They probably live in the community and can benefit. We need to step back and see each others’ perspectives.

The research also looked at people's motives for wanting to get involved, and found that they supported the change in default setting. People's motives did appear to be broadly concerned with public goods such as community safety or environmental conditions, and therefore could be classified as civic: most people wanted to make a difference in the area. Seven of the 30 participants in the first iteration were motivated by a principled feeling that everyone should do their bit, which is an explicitly stated belief in co-production: for example, ‘I think I should be not just complaining, actually doing something’ and ‘If you're not making an effort you can't complain’. Where people were concerned, it was scepticism about how far the institution was making a genuine change: ‘Worried it's just a token gesture’. Others were sceptical about to what extent the project could overcome barriers to participation, and argued that other citizens would not respond: ‘They are flogging a dead horse’ and ‘it will fall on deaf ears’.

The second critical success was that the initial nudge did start to mobilize people towards volunteering. The request by the contact centre generated additional interest from citizens who had not previously been involved in volunteering, and was successful in attracting a cross-section of people in the neighbourhood. In neighbourhood one, 30 callers were recruited, including Asian women, younger people and a large proportion of people in work. Five people had not undertaken any civic activity in the past year and a further 11 had only done limited, one-off activity. In neighbourhood two the profile was different, reflecting differences in the local population. Of the 33 callers recruited, all were white, and compared to the callers in the first location, participants tended to be slightly older. Only a quarter were in work and a third were sick or disabled or caring for a sick or disabled relative. Seventeen had not taken part in any civic activity in the past year and a further nine had only done limited, one-off activity.

However, the initial surge in interest was not translated into activity. The comparison group performed as well as the intervention group in carrying out voluntary activity, and neither group showed massive increases. The nudge was in two iterations: changing the way the authority mobilized through the contact centre, with a light-touch follow-up using existing volunteering opportunities; then a second iteration which also created new volunteering opportunities. The fundamental initial shift in the default setting at the contact centre created a different citizen response. But the experiment failed to capitalize on the initial expression of interest in the follow-up intervention. There were several reasons for this. One obvious argument is that when put on the spot, people gave socially acceptable responses without being genuinely committed to considering civic action. Therefore, interest tails off. While this is always possible, there are more cogent arguments that may explain the drop-off in citizen interest.

The neighbourhood officers did not make contact with eight of the seventeen people in the intervention group, meaning that the intervention was flawed. More importantly, in the follow-up interviews people fed back that the voluntary opportunities offered in the first wave were not appropriate ones. For example, the types of voluntary contributions in which people expressed interest included making themselves available to present their views, or having a more in-depth dialogue with services, in order that services could better tailor, or adapt, their responses. This could include expanding the presence of police community support officers or improving parking or helping to improve the appearance of the area by reducing the amount of litter or rubbish dumping. It could include citizens helping to improve things as individuals, for example by improving their own front gardens, or by running practical community projects to help each other, such as neighbourhood watch schemes (‘We should be vigilant and help each other’) and projects to reduce the isolation of elderly and housebound people.

No one explicitly mentioned wanting to attend meetings or join existing community groups. Indeed, some participants had previously been put off volunteering through their experience or knowledge of these existing opportunities. However, in the first wave, the options offered to interested citizens were largely about attending public meetings or joining local community associations to help with their limited range of activities, in tightly prescribed roles. There were few options that fitted people's preferences. For example, no support was offered for garden tidying, and there were no creative ways for people to have dialogue with service providers other than the conventional routes of public meetings or individual complaints. There was some assistance offered to those wanting to set up mutual aid and self-help community projects, but it was to transfer the job of mobilizing on to the citizen, so people were given information packs for neighbourhood watch schemes, but were not given the names of other residents who had also expressed an interest. Residents said they were uncomfortable with the level of administration involved in setting up a scheme.

The second wave of the experiment was adapted to address this, using the things citizens had said they wanted to do in order to develop new options. For those wanting to use their views more constructively to improve services, the authority then started to set up the public sector equivalent of a mystery shopping exercise. It had previously used mystery shopping with paid staff, but wanted to extend this to citizens. People are offered training before using a script to make a series of requests of the authority, and the response is recorded using a set of criteria. Tests are made using different scenarios, from different types of citizen, with different staff and departments. Feedback is then given on how well the authority performed and on areas for improvement. However, the process of creating the scheme was a long and involved one, requiring agreement at a senior level on the timing of the mystery shopping exercise, its parameters, the script used and feedback mechanisms, as well as organizing a relatively expensive specialist training package for the volunteers which took place over an induction period.

For those wanting to offer help to neighbours, the authority facilitated a good neighbour scheme through which residents could offer social support to isolated older people by means of a good morning phone call. This involved many weeks of work by the council to identify vulnerable older people who may need help, advertise the service to them and set up training in befriending for the volunteers.

Therefore, although progress was made on creating new volunteering opportunities, the delays in doing this meant that momentum was lost, and citizens’ initial fears that the institutional change would not be deep enough seemed to be confirmed. The nudge did not go far enough in changing the choice architecture. Feedback from citizens suggests that mobilization would have resulted in more activity had the changes been more extensive.

Discussion – could we have improved the nudge?

Could the nudge have been improved? What would have made it a better nudge? Arguably, the experiment needed to go further by changing all aspects of the choice architecture, and doing this more quickly. The change in default setting did not extend quickly enough to volunteering options. The original intention was to offer a creative menu of voluntary options based on individuals’ skills and interests. However, the institution reverted to a default setting previously hidden, which was co-production options that suited the institution and its skill set, rather than options tailored to the citizen. Participants in the first-wave intervention were offered an arguably uninspiring menu of involvement in existing neighbourhood groups and forums – easy for the authority to understand, based on a established repertoire of engagement skills, known entities, low supervision and transaction costs – though citizens would have preferred stronger voice mechanisms, less group-reliant activities requiring additional organization, higher monitoring and transaction costs, and innovative thinking. In the language of the CLEAR model, people were not ‘enabled to’; the choices we tilted people towards were not attractive or tailored enough. By the second wave, the authority had started to develop new options, but this was too slow for many of the people we had mobilized.

Another issue was that the experiment focused on changing default settings, but neglected other elements of nudging, in particular the incentives, structuring complex choices, giving feedback and expecting error. If volunteering had been incentivized, for example through Timebanking (a scheme where volunteers earn time credits to spend later on, see http://www.timebanking.org) or childcare offers, this may have increased activity. There was a gap in helping practitioners and citizens structure complex choices. Both parties know they wanted volunteering to happen, but neither was clear how best this could be done. Both had identified problems they wanted to tackle, but were unsure how to go about this. We started to help people structure these choices, first by getting local authority staff to conduct a skills audit with citizens, and then holding a workshop discussion with staff to take account of these skills and interests in developing a more extensive menu of options for citizens. Nudge includes the idea of feedback, which was used in the experiment but too late to have any real feedback effects. The authority did give feedback to people in the experiment (or ‘responded to them’ in CLEAR terminology) about how many other citizens had participated, and thanked them in an effort to validate citizens’ efforts in coming forward. We have seen and can understand some of the gaps in the experiment in terms of nudge ideas. But as well as offering a way of interpreting our results, nudge is a potential tool for strengthening the activity. The likelihood is that a stronger nudge may have been able to extend the results further.