The engagement of its citizens is the lifeblood of a democracy. In the UK, engagement with formal electoral politics is on the slide, turn-out is low,  party membership has fallen, and trust in politicians is lower even than in estate agents, bankers and journalists. Political engagement is also profoundly unequal. It is more common among the well-educated and well-remunerated, and among older adults. It has been argued that volunteering is, itself, an alternate form of civic engagement; a direct substitute for electoral engagement. It has also been treated as part of a complex “syndrome” of social capital – albeit one in which causal connections are hard to discern. In this world view, volunteering promotes generalised social trust by facilitating face-to-face contact with various dissimilar others. Any policy which increases rates of volunteering should also therefore increase levels of trust and levels of political engagement. However, volunteering is as unevenly spread as other types of civic engagement: it, too, is more common among the middle classes and those with higher education.

In UK public policy, governments of all stripes have promoted volunteering as a way of improving levels of civic and political engagement and also as a good in itself. Volunteering has been used as a part of a battery of measures to promote wider civic and political engagement, particularly among otherwise disengaged young people, for example as part of National Citizen Service. Policies like this, which promote volunteering as a way of shoring up and equalising civic engagement, rest on an argument from social capital which has been extensively criticised both in empirical and theoretical terms.

TSRC research student Vicki Bolton took various approaches to the study of volunteering in her doctoral research. Key conclusions from her thesis, which used a range of longitudinal and cross-sectional survey datasets to explore the connections between volunteering and civic engagement, are as follows:

Young adult volunteers are more likely both to volunteer and to vote in their 50s than individuals who did not volunteer in early-adulthood. This is in keeping with the literature on the formative nature of volunteering, which shows an increase in future community service, voter-registration, greater intent to vote, and interest in politics associated with voluntary association membership and community service in high school. However, the effect was strongly differentiated by social class. Individuals born into homes supported by a professional or managerial worker were more likely to volunteer, and those middle-class volunteers were also more likely to remain civically engaged in middle-age than their working class volunteer counterparts.

The class effect was more pronounced for later volunteering than for later voting. The probability of volunteering at age 50 was roughly equal for 23 year old volunteers and non-volunteers from a social class V background. However, for those from social class I households, the probability of volunteering at age 50 was more than doubled by volunteering at age 23. Nevertheless, volunteering at age 23 had a positive effect on voting at age 50 for all but those with the most working class backgrounds (for whom the effect was actually negative).

The effect of volunteering on political engagement was investigated over the short run using data from the British Household Panel Study that tracks its respondents continuously from 1991 onwards. Much of the literature focuses on the formative effects of adolescent and early-adult volunteering, rather than effects for all adults over a shorter period. Volunteering had a positive effect on political interest, once individuals’ backgrounds were appropriately accounted for. Although the effect size was reduced by using these strong controls, it remained statistically significant at the 1% level. Since both volunteering and political engagement are strongly predicted by social class, upbringing and education, a smaller effect size is to be expected as likely confounders are accounted for.

There were statistically significant indirect relationships, mediated through generalised social trust, between volunteer hours and time spent in meetings of a ‘political’ kind, and between volunteer hours and political ‘contacting’ activity. Volunteering is associated with increased levels of social trust, and that increase in trust is associated with an increase in political activity: this is the thesis suggested in the social capital literature, and particularly by Robert Putnam. However, the size of these indirect relationships via social trust was small compared to the overall effect size: most of the relationship between volunteering and political acts is explained by something other than trust. This finding contrasts with the theoretical literature, much of which expects a somewhat more important role for trust, but does fit with existing empirical evidence.

The role of institutional trust was also investigated. Institutional trust did not have a mediating role between volunteer hours and either ‘meeting’ or ‘contacting’ forms of political engagement. In the literature, the role of volunteering in promoting institutional trust, and of institutional trust in promoting political engagement, is contested: some authors predict that institutional trusters will be more engaged, while others contend the opposite. Some predict that volunteering should promote institutional trust, and others that the knowledge acquired through volunteering could diminish that trust. For this population, volunteering is associated with slightly increased levels of institutional trust, while institutional trust is associated with reduced meeting and contacting behaviour. However, institutional trust is not a mediator between volunteering hours and political activities.

The research presented here shows that volunteering is formative of future civic and political engagement; that volunteering also influences political engagement in the shorter term, and that trust is a mediator in the relationship. These findings are not, in themselves, novel: but the data and methods used here allow them to be stated in more confidence than from previous research. However, all the effects described above are strictly limited, either in size or scope. The formative nature of early life volunteering is constrained by social class. Individuals born into professional or managerial class homes see a larger civic effect from their voluntary work than those from manual or working class backgrounds. The short-run effect of volunteering on political engagement applies only to political interest, and not to formal electoral engagement. And the role of generalised social trust is rather small, compared to the overall size of the relationship found between volunteering and political activity in the Citizenship Survey data.

As with broader reviews of the impact of volunteering, therefore, we should be cautious in over-claiming for the wider benefits it has for individuals and communities.

Public policy is being made on the basis of the broad social capital view of volunteering and political engagement which is explored critically here. This research shows that, although volunteering exerts some civic leverage over certain types of people and certain types of political engagement, it shouldn’t be relied on as a method to bring currently disengaged voters back into the electoral fold, to raise the general level of engagement with the electoral process, or to reduce disparities in engagement.

Firstly, the effects of volunteering on civic and political engagement are strongest for those who are already most likely to be engaged. Early-adult volunteers from manual work backgrounds are no more likely to be civically engaged in middle-age than those who did not volunteer in early adulthood: they are no more likely to volunteer and less likely to vote. It seems likely that, if it had an effect at all, improving rates of volunteering in early adulthood would produce a civic and political benefit mostly among the middle classes, whose rates of engagement are already higher. This should be part of the continued conversation about inequality in civic and political participation. Those with least power, who are most reliant on the state and its institutions, are the least likely to be active participants in the systems which govern the state and its institutions. Volunteering policy is not likely to be an effective tool in tackling this inequality.

Secondly, volunteering can affect political interest, but does not influence party support. Political interest is a motivator in formal electoral engagement, but does not guarantee it: indeed, political interest and other types of engagement are seen to be jointly caused, rather than causal. Politically interested individuals may be merely ‘spectators’ rather than active participants. Further work would be required before volunteering could be suggested as a tool to increase political engagement via political interest.