Why do only some people who support parties actually join them? Evidence from Britain

Abstract

What makes people join a political party is one of the most commonly studied questions in research on party members. Nearly all this research, however, is based on talking to people who have actually joined parties. This article simultaneously analyses surveys of members of political parties in Britain and surveys of non-member supporters of those same parties. This uniquely enables us to model the decision to join parties. The results suggest that most of the elements that constitute the influential ‘General Incentives Model’ are significant. But it also reveals that, while party supporters imagine that selective benefits, social norms and opposing rival parties’ policies are key factors in members’ decisions to join a party, those who actually do so are more likely to say they are motivated by attachments to their party’s values, policies and leaders, as well as by an altruistic desire to support democracy more generally.

Keywords:

• Party membership
• party supporters
• political parties
• joining parties
• General Incentives Model
• Britain

Since Patrick Seyd and Paul Whiteley first launched their seminal research programme on political party members (Seyd and Whiteley 1992; Whiteley et al. 1994, 2006), their ‘General Incentives Model’ (GIM) has become a widely accepted approach in comparative politics to understanding why some citizens in liberal democracies join political parties and become active within them (see van Haute and Gauja 2015). Indeed, one might almost refer to it as conventional wisdom or orthodoxy, and it has certainly spawned a host of emulations and adaptations to other settings by scholars around the democratic world and in Western Europe in particular (see, for instance, Clarke et al. 2000; Gallagher and Marsh 2002; Ridder et al. 2015; Spier and Klein 2015).

The standard way of testing the validity of the GIM approach has been to survey party members and ask them questions pertaining to the variables at the heart of the model. This is problematic. There are always limitations to any research undertaking, of course, but one in particular has constrained work on party members: namely that the survey approach has generated no variation on a key dependent variable – party membership itself. While it remains possible to examine interesting descriptive statistics on who the members are and what reasons they give for joining parties, surveys of party members which by definition exclude non-members render multivariate modelling of membership itself impossible. This article fills that gap. By combining recent, simultaneous surveys of members and non-member supporters of parties (i.e. voters who strongly identify with a party but do not go as far as to formally join it) into a single integrated dataset, we achieve variation on the key dependent variable, allowing us to properly explore the initial decision to join. We are thus able for the first time to model party membership properly: why do some people who are strongly supportive of particular parties choose to belong to them while others prefer not to?

The General Incentives Model

We start with a brief recap of the GIM – an approach

grounded in the assumption that participation occurs in response to different kinds of incentives … but it goes beyond a narrowly cast economic analysis of incentives to include emotional attachments to the party, moral concerns, and social norms, variables which lie outside the standard cost-benefit approach to decision-making. (Whiteley et al. 1994: 109)

To summarise the model (first set out in Seyd and Whiteley 1992: 112), it incorporates a combination of the following core General Incentives to join a party:

–	The respondent’s perception of the probability that participation in group activity through the party will achieve a desired collective outcome; in other words, the respondent’s sense of group efficacy.
	The respondent’s desired collective policy outcome, such as the introduction of a particular policy.
	The respondent’s assessment of the selective outcome benefits of activism; that is, material or career benefits.
	The respondent’s assessment of the selective process benefits of activism; that is, the intrinsic pleasure derived from involvement in political action.
	The respondent’s altruistic motivations for activism.
	The respondent’s perception of social norm incentives for activism; that is, the desire to conform with the behaviour and expectations of personal contacts;
	The respondent’s expressive or affective motivations for activism, such as the strength of commitment to or identification with a given party or leader.
	The respondent’s perception of the costs of activism; properly speaking, this is a disincentive.

In addition to these core features of the GIM, there are two others which are also added to the explanatory mix in some versions of the work of Seyd and Whiteley, although they are not unique in their approach:

• The respondent’s belief that individual acts can influence and have a real impact upon political decisions; that is, the respondent’s sense of personal efficacy.

• The respondent’s ideological motivations for activism.

These incentives have been prominent in descriptive data on the memberships of the Labour Party (Seyd and Whiteley 1992), the Conservatives (Whiteley et al. 1994) and the Liberal Democrats (Whiteley et al. 2006). Although some time has passed since this research was conducted, there is no obvious reason to suppose that the factors motivating people to join parties might be time-dependent, so we would broadly expect that the same factors will prove to be significant in explaining why some partisans decide to become members, whereas others remain non-member supporters. Seyd, Whiteley and their various collaborators were unable to move beyond descriptive findings – highly suggestive though they undoubtedly were – but our new dataset allows us to model the decision to join properly. How well does the GIM perform when tested in this way?

Data, measures and expectations

Overall, in our Party Members Project (PMP) dataset there are 12,079 respondents who, straight after the 2015 elections, identified strongly with one of the six political parties in our study, namely Labour, Conservative, Liberal Democrat, UKIP, Greens and the SNP. Among them, 5700 did go as far as formally joining a party, whereas 6379 did not.¹ Henceforth, we will refer to the former as party members and to the latter as party supporters.² Our aim is to identify the key drivers that lead party identifiers to become formal members of a party. Although we cannot investigate this in a causal way (as, theoretically, one could through a panel, for example), we can make some reasonable assumptions about what is likely to have led some respondents to join a party, and we can test these assumptions by looking at the key factors that help us predict which of the respondents in our dataset are party members and which are ‘only’ party supporters.

In addition to examining the impact of core GIM factors, we also look at whether standard socio-demographics have any influence on why people join parties, along with ideology and personal efficacy, which are all well-known predictors of ‘high-intensity’ forms of political participation such as party membership. We know from long-established research that resources are crucial when it comes to participation and so assume that this is also the case when it comes to party membership. These personal ‘resource’ attributes are central to the Civic Voluntarism approach, one of the most widely cited models of general political participation in political science. It originates with work on US participation by Verba and Nie (1972) and has since been applied in other countries, including Britain (Barnes and Kaase 1979; Parry et al. 1992; Verba et al. 1978, 1993). In a nutshell, this model claims that there are three reasons why people do not become politically active: ‘because they can’t; because they don’t want to; or because nobody asked. In other words, people may be inactive because they lack resources, because they lack psychological engagement with politics, or because they are outside of the recruitment networks that bring people into politics’ (Verba et al. 1995: 269). In particular, resources such as access to education and/or suitable jobs provide people with the chance to develop the organisational and communication skills that are relevant to political participation, as does being embedded in non-political social networks like churches. In truth, the GIM and Civic Voluntarism Model (CVM) approaches are not completely independent of one another, since the influence of the latter’s social networks is implicit in the former’s selective process and social norm effects. So, while it is not the aim of this article to test the CVM as such, in controlling for demographics and personal political efficacy, we are in effect recognising that GIM builds not only on rational choice individualism, but also incorporates key elements of the civic voluntarism approach in explaining the decision to join political parties.

On the basis of these prior findings, we therefore start by examining the impact of demographic resources, and expect partisans with higher levels of education and social class, as well as men, to be more likely to be members of a party; we do not, however, have any specific expectations about whether members are more likely to be younger or older than party supporters.

We then go on to consider the impact of ideology and personal political efficacy, employing normalised scales that, unless otherwise stated, run from low (0) to high (10). Ideological incentives are measured by reference to various attitudinal scales which tap ideological dimensions widely recognised as salient features of contemporary British politics. The first is based on a well-known left–right scale drawing on a battery of questions that have routinely been asked of respondents to the British Election Study (and other surveys) since the 1990s (Heath et al 1993).³ The left–right additive scale runs from 0 (left-wing) to 10 (right-wing) and the scale items from which it is constructed produce a very high Cronbach’s alpha reliability coefficient (0.88 for party supporters; 0.91 for party members). Following the well-known ‘law’ of curvilinear disparity (May 1973), we would infer that party members are likely to be more radical than party supporters and that this ideological impulse is a motivation for joining parties. Thus, members of parties generally recognised as being on the left of British politics at the time the data was gathered in 2015 (Labour, Greens, Liberal Democrats and SNP) are expected to stand more to the left of the party average than party supporters, whereas members of parties on the right (Conservatives and UKIP), are expected to stand more to the right of the party average compared to party supporters. Although one of the two parties that we have designated ‘right-wing’ – UKIP – is actually only properly so described on questions of identity and social authority rather than on questions of political economy (Webb and Bale 2014), we still expect that members will be more likely to stand to the right of the party average compared to party supporters. We look at this through a variable measuring the extent to which party members and supporters stand at the extreme of their own party. After calculating the average left–right score for members and supporters of each party, we then calculate the difference between each individual respondent and this overall party left–right average; we are thus able to construct a single variable measuring proximity to the party’s ideological extremity, as it takes on positive values for positions to the left of the party average for parties on the left and to the right of the party average for parties on the right (Conservatives and UKIP). Negative values would indicate more centrist positions.

The second ideological indicator we focus on is an additive scale (also designed by Heath and his colleagues) to tap respondents’ positions on questions of social liberalism and authoritarianism (with 0 representing the liberal end of the scale and 10 the authoritarian end).⁴ This is also eminently reliable (alpha = 0.82 for party supporters and 0.84 for party members). In this case, although it is likely that members and supporters of parties on the left are more socially liberal than those on the right, we expect that, on the whole, party members will be more socially liberal than party supporters. This is because social liberalism is a value that places greater emphasis on democratic engagement as a civic right and a means of maximising liberty while enhancing political knowledge and personal development (Howarth 2007).

We also employ a measure of personal political efficacy. This is the respondent’s perception of the probability that their personal participation will achieve a desired collective outcome, and is not to be confused with the respondent’s sense of group efficacy. Although not one of the core general incentives for party membership (Whiteley et al. 1994: 96–7), we know from the CVM that an individual’s perception that political change is possible and that they can play a role in effecting such change is a driver of participation (Rosenston and Hansen 1993; Verba et al. 1995). The more individuals feel they are politically competent, the more they will be willing to participate in politics, and we believe this is particularly the case for high-level participation such as party membership. We therefore look, too, at the respondent’s perception of the probability that their personal participation will achieve a desired collective outcome. Political efficacy is measured by the degree of agreement or disagreement with two Likert-scale statements: ‘politicians don’t care what people like me think’ and ‘people like me can have a real influence on politics if they are prepared to get involved’. We expect party members to have a higher sense of personal efficacy than party supporters.

We now turn to core GIM predictors. Unless otherwise stated, responses for each of these predictors have been added together, normalised and coded into combined scales where 0 represents a low incentive and 10 represents a high incentive These start with a measure of group efficacy, which rests on the claim that party supporters are more likely to join a party when they also have a high perception of probable group influence through their membership. We measure this with a combined index created from the degree of agreement/disagreement with two Likert-scale statements: ‘when party members work together, they can really change the local community or country’, and ‘the party leadership doesn’t pay a lot of attention to ordinary party members’. We therefore expect party members to have a higher sense of group efficacy than party supporters.

A further factor derived from the GIM is the respondent’s expressive or affective motivations for activism. This is gauged through respondents’ strength of partisan identification, running from not very strong to very strong. We expect members to have higher perceived expressive motivations for activism than party supporters. The next GIM predictor included in our model is altruistic incentives for activism. This is measured by response to the Likert-scale statement: ‘Every citizen should get involved in politics if democracy is to work properly’. We expect members to be more likely to agree with this statement, and therefore to have higher altruistic incentives than party supporters.

Joining a party may also be the result of more narrowly self-centred drives, of course. GIM theory suggests that some might be motivated to formally join a party because of rewards or gratifications that flow only to members. This is the case with the respondent’s assessment of the selective outcome benefits of activism. Such benefits are typically related to career or material ambitions. In other words, membership is seen as a necessary path to a particular outcome that will benefit them personally – for instance, a career in electoral politics. It is measured by looking at the degree of agreement with the statement ‘a person like me could do a good job of being a local councillor or MP’. In addition, the respondent’s assessment of the selective process benefits of activism might also be a significant factor; here, the gratification restricted to the members is achieved through the process of participation, such as enjoyment of interaction with other (like-minded) people. Here it is measured by looking at the degree of agreement with the statement ‘being a party member is a good way to meet people’. We expect members to be more likely to agree with these statements and therefore to have higher perceived selective outcome and selective process benefits of activism than party supporters.

Joining a party also comes down to social norms, or, in other words, the pressure to conform to the influence of other people in being willing to join a party (Whiteley 1995). This is something that could be compared to what, in the CVM, has been described as the network and mobilisation factor (Verba et al. 1995). In other words, those who are not so embedded in wider political action networks are less likely to be mobilised into party membership. We therefore operationalise social norm pressure with a dichotomous variable indicating whether the partisan is already a member of other non-political or semi-political organisations, such as trade unions, Greenpeace (for environmentalists) or Saga (for retirees and seniors). We expect party members to be more likely than party supporters to be members of other civil society organisations.

Last but not least, partisans might also be disincentivised to join a party by the perceived costs of activism. This is measured by a simple Likert-scale question about whether respondents agree or disagree with the idea that: ‘party activism often takes time away from one’s family’. Once again, the index runs from low costs (0) to high costs (10). Since, in the case of costs of activism, we are dealing with disincentives to becoming a party member, rather than incentives, we expect members to perceive the costs of activism as lower than do party supporters.

Finally, it should be noted that, due to an absence of equivalent indicators in the party members and party supporters survey, we cannot here include a measure for collective policy incentives. We would argue in any case that collective policy incentives often closely parallel general ideological positions, so the lack of these indicators in our model should not represent a threat to the validity of the results.⁵

Results

We start with a review of the descriptive statistics in Table 1. How do members and supporters compare and contrast in terms of these demographic and general incentives factors? The first thing that strikes one in reviewing the table is that the differences between members and supporters are significant for all variables, demographic and attitudinal, although the mean age difference (51 for party members to 52 for party supporters) seems too small to warrant serious consideration as an important explanation of why people might join parties. The members are notably more male, middle class and educated than supporters, however, showing that socio-demographic resources could be an important factor in party membership. As expected, party members are also less moderate than party supporters: in left-of-centre parties members are more left-wing than supporters, whereas they are more right-wing in right-of-centre parties. This is apparent in both the basic left–right index scores and the ‘left–right extremism’ scores in Table 1.⁶ However, and once again as expected, members are more socially liberal than supporters in both left- and right-wing parties. Members also have a decidedly greater sense of personal political efficacy than do supporters. All of this meets our hypothetical expectations.

Table 1. Social and political characteristics of British political party members and supporters, 2015.

Attribute	Supporters	Members	N	Significance
Mean age	52	51	12,069	*
% male	48.0	65.0	12,072	***
% graduates	32.1	46.2	11,780	***
% ABC1	59.1	77.6	11,914	***
Mean left–right
Left parties	1.99	1.75	7,660	***
Conservative Party	5.08	6.34	2,264	***
UKIP	2.62	3.59	1,790	***
Mean left–right extremism
Left parties	–0.09	0.08	7,660	***
Conservative Party	–0.62	0.64	2,264	***
UKIP	–0.41	0.56	1,790	***
Mean liberty–authority
Left parties	5.40	3.93	7,317	***
Conservative Party	7.39	6.58	2,232	***
UKIP	7.97	7.54	1,790	***
All parties	6.22	4.99	11,339	***
Personal efficacy	4.18	6.28	11,758	***
Group efficacy	5.13	7.33	10,855	***
Expressive incentives	6.34	8.28	12,053	***
Altruism	5.80	7.07	11,649	***
Selective process	5.76	7.15	11,076	***
Selective outcome	5.11	6.54	11,130	***
% Social norms	29.6	48.2	12,071	***
Cost of activism	6.87	6.44	11,001	***

Source: PMP; * p < 0.05; ** p < 0.01; *** p < 0.001.

When we look at the core GIM incentives measures, it is apparent that members outscore supporters on incentives such as group efficacy, expressive ideals, altruistic concerns for the wellbeing of the political system, and on the selective outcome and process incentives of membership. That is, members have a far more prominent expressive belief in the ability of particular parties, as well as in the ability of party members, to effect political change within a community; they are also much likelier to rationalise their membership in terms of a high-minded desire to contribute to the healthy functioning of democracy. At the same time, however, members are also more likely to think that being a party member is a good way to meet like-minded people as well as a way to potentially fulfil an ambition (latent or manifest) to become a politician. Moreover, members are more likely to respond to social norms as they are part of a wider network of civil society organisations. Finally, however, it is supporters who perceive the cost of party activism (in terms of the time they presume it takes up) to be higher than members, who know how much time it involves. Again, all of these descriptive findings conform with our expectations.

Thus far, emulating the approach of previous researchers by examining descriptive statistics seems to confirm the continuing validity of the GIM today. But how do these explanations fare in the context of a multivariate model? This is where this article can add real value to existing research. Table 2 reports the results of a logistic regression model in which the dependent variable is party member (1)/non-member supporter (0). The first model (M1) looks at whether socio-demographic resources, personal political efficacy and ideology help explain which partisans are more likely to have become members of a party. The second model (M2) introduces the core GIM factors.

Table 2. Logistic regression of party membership, 2015.

	M1		M2
	(N = 8714)		(N = 8714)
	B	SE	B	SE
Male	0.63	0.05***	0.81	0.06***
Graduate	0.18	0.06***	0.30	0.07***
Age	0.03	0.02*	0.04	0.02*
Social grade	0.52	0.06***	0.63	0.07***
Socially liberal/authoritarian (0–10)	−0.18	0.01***	−0.20	0.02***
Left–right extremism (−8.5/7)	0.16	0.02***	0.06	0.02***
Personal efficacy (0–10)	0.54	0.01***	0.26	0.02***
Group efficacy (0–10)			0.48	0.02***
Expressive incentive (Very strong vs. Fairly strong party id)			1.45	0.06***
Altruism (0–10)			0.10	0.01***
Selective outcome (0–10)			0.05	0.01***
Selective process (0–10)			0.13	0.02***
Social norms (member of other organisations vs. non-member)			0.83	0.06***
Cost of activism			−0.10	0.02***
Constant	−2.26	0.18***	−8.97	0.32***
Pseudo-R2	25.6		44.5

Source: PMP; * p < 0.05; ** p < 0.01; *** p < 0.001.

Note: A control for parties has been included in the two models but is not shown here in order to simplify presentation.

We see from Model 1 that something that was apparent in Table 1 is confirmed: all of the predictors have a statistically significant bearing on the decision to become a party member, and in the expected direction. This offers broad confirmation of the importance of personal resources, political efficacy and ideology. Holding all other factors constant, party members are significantly more likely to be male, middle class, educated and older than supporters. Party members also have a significantly stronger sense of political efficacy than non-members; quite whether this ‘can-do’ mentality is something that drives them to join parties or whether involvement in party activity has enhanced their sense of political efficacy is a causal puzzle that has long been discussed in the wider literature of political participation (e.g. Finkel 1985) and one that, sadly, our data cannot unravel; that said, both are plausible explanations – and, indeed, elements of both may be behind this finding. Moreover, those who hold more extreme positions compared to their party average (on the left for left-wing parties and on the right for right-wing parties) have higher probabilities of being party members rather than party supporters. Finally, M1 confirms that party members, regardless of the ideological inclination of their party, are, on average, more socially liberal than party supporters.

If we turn to Model 2 (M2), we see that the addition of the core GIM factors does not alter the direction of any the demographic or ideological predictors in M1. In addition to Model 1, we also see confirmation that those who join parties rather than remain mere supporters are significantly more likely to score highly in terms of group efficacy, expressive orientations towards parties, altruistic motivations for political engagement, selective outcome and process motivations; they are also more likely to be involved in other civil society organisations. But, as expected, members are less likely than supporters to regard the opportunity cost of the time involved in membership as being high.

In short, the factors that clearly distinguish members from supporters, and may go furthest in explaining why they bother to join up, are being male, better educated and coming from higher up the social hierarchy (i.e. having more resources), being socially liberal and ideologically more radical (for parties on the left), having a strong sense of personal and collective political efficacy, strong expressive belief in a particular party and altruistic belief in the wider importance of political participation, a perception of the value of selective process and outcome benefits, involvement in other civil society organisations, and not being overly deterred by the time commitment of being a member. This amounts to clear confirmation of the continuing relevance of the GIM explanation of party membership in the twenty-first century.

Differing narratives of party membership: members versus supporters

While we now have convincing confirmation of the general incentives explanation for party membership, we can extend our knowledge by further investigating the different narratives that members and supporters develop about their reasons for joining. Do non-joiners have realistic expectations of why people decide to join a party? Is it possible that they maintain rather different beliefs to those held by members, which then serve to disincentivise them from joining? One way of shedding light on this possibility is by asking supporters a series of analogous questions about why people might join parties. The two models presented so far (M1 and M2) are based solely on identical questions that have been asked of both members and supporters. However, we also asked the latter a number of questions that varied slightly from those asked of members. Party members were asked questions relating to the importance of a series of motivations for joining a party, for which they were able to draw directly on their own experience of membership in answering. However, for obvious reasons, non-member supporters could not draw on that experience. Instead, we asked supporters questions with analogous, but different, wordings. Thus, whereas party members were asked ‘How important were the following reasons for joining the party?’, supporters were asked: ‘The following are reasons people who join political parties sometimes give for joining them. Thinking about the type of people who join political parties and why you think they do so, please rate the importance of each one’. Asking supporters questions about why they think some people join parties, rather than why actually they did join parties, allows us to compare whether supporters’ perceptions are similar to members’ reported motivations. This is interesting because it sheds light on whether supporters have what, from the point of view of party members, might be inaccurate ideas about party membership; should this prove to be the case, it might point to further reasons why they have elected not to join a party despite thinking a great deal of it.

There are six possible motivations that we can address in this way: expressive incentives (‘An attachment to the party’s principles’ and ‘Belief in the party’s leadership’), collective incentives (‘To support the party’s general policies or a specific policy that mattered greatly to me’ and ‘To oppose the policies of a rival party, or the power of a social or economic group, such as big business or unions’), altruism (‘To support the democratic process’ and ‘To promote the interests of the nation’), social norms (‘The influence of family, friends or colleagues’), selective process (‘Being able to engage in activities in which you would be mixing with other like-minded individuals’) and selective outcome (‘To enhance their career’).

The results reported in Table 3 are highly suggestive. In a number of cases, they show that party members score more highly in response to the ‘direct’ questions put to them than supporters do to the analogous, counterpart questions. That is, members see expressive incentives, collective positive incentives and altruistic motivations as being more important reasons for party membership than supporters imagine them to be. However, the reverse is true in respect of several other factors. Supporters are slightly more likely to imagine that opposing a rival party’s policies is important than members actually feel to be the case. More obviously, though, supporters clearly regard selective process and (particularly) outcome incentives, as well as the pressure of social norms, as likely to be of greater importance than members actually acknowledge.

Table 3. Comparing perceptions of membership incentives though direct and analogous questions.

	Supporters	Members	Significance
Expressive incentives
Attachment to party’s principles	8.24	9.21	***
Belief in party’s leadership	7.73	7.89	***
Collective incentives
Support policies	8.30	9.20	***
Oppose policies	7.82	7.72	***
Altruism
Support democratic process	7.17	8.56	***
Promote national interest	6.88	8.88	***
Selective process	7.55	5.59	***
Selective outcome	6.49	2.06	***
Social norms	5.71	2.98	***

Source: PMP; * p < 0.05; ** p < 0.01; *** p < 0.001. N = 12,069.

Note: Highlighted incentives are those which supporters score more highly than members.

Thus, party supporters believe that party members must derive intrinsic pleasure from involvement in party life, and career or material benefits, and that they must be embedded in a social network that draws them into the party. Actual members are far less likely to believe that this is the case – especially with respect to selective outcomes or social norms.⁷ Hence, we have evidence of the different understandings or narratives of what party membership actually entails: while supporters imagine – perhaps somewhat cynically – that opposing rivals’ policies, meeting like-minded people, and selective outcomes are key factors in the decision to join, party members are more likely to believe that they are motivated by rather high-minded attachments to party principles, policies and leaders, and by a desire to support the democratic process and the national interest.

Of course, whether this means that members really are more altruistic and principled than supporters, or that they are merely in denial about the true significance of the potential material or career benefits they hope (or hoped) to gain from party membership, is ultimately impossible to tell; it is known that ‘[c]onscious deliberation and rumination is … the rationalization of multiple unconscious processes that recruit reasons to justify and explains beliefs, attitudes and actions’ (Lodge and Taber 2013: 22), so we must regard these results with a general caveat concerning the subjective and often cognitively biased nature of attitudes. Even so, the findings do seem to provide a strong clue as to the gap in perception that distinguishes members from supporters. In other words, if supporters do not believe they themselves are driven by the same motives as party members, and/or if they find those motives somehow off-putting, they are probably less likely to become members themselves.

Conclusion

In this article, we have sought to test the widely cited General Incentives Model of party membership using a unique dataset that combines both party members and non-member supporters to explore what distinguishes the two sets of partisans when it comes to the decision to join the party. We have seen that the GIM largely stands up well: all of the factors that constitute the model turn out to be significant and to impact on the decision to join in ways that the GIM would predict. A strong sense of personal and collective political efficacy, an expressive belief in a party and an altruistic commitment to the wider importance of political participation, a sense of the value of selective process and outcome benefits, involvement in a wider set of civil society organisations that bring social norm pressures with them, and not being deterred by the implicit time commitments, all impel people to join up. Being more ideologically radical than the average (strong) party supporter in left–right terms also makes people more likely to join parties, as does being more socially liberal, which is in keeping with the emphasis such individuals place on democratic engagement. Thus, the general array of factors that might incentivise an individual to join a political party, in conjunction with the social and psychological resources that have widely been held to facilitate political participation more generally, can be confirmed by our study as being directly important.

Through comparison of responses to direct and analogous questions put respectively to members and non-member supporters, we have also identified the impact of the different narratives of party membership maintained by these two groups. While supporters imagine that selective benefits, social norms and opposing rival party policies are key factors in the decision to join, those who actually join a party are more likely to believe that they are motivated by principled attachments to its values, policies and leaders, and by an altruistic desire to support the democratic process and the national interest.

This has practical implications. By respectfully and sensitively challenging supporters’ narratives about members – perhaps by encouraging activists to talk more about the realities of membership wherever possible – political parties may be able to convert more supporters into members. Research on the campaigning contributions of both groups in the era of the ‘multi-speed membership party’ (Scarrow 2015) suggests that doing so may not be as essential in terms of electioneering as is commonly imagined (Webb et al. 2017). However, it could still be extremely valuable in terms of, for instance, party legitimacy, candidate recruitment and organisational presence (Kölln 2015) – particularly when, outside of the UK anyway, membership in many political parties still seems to be dropping like a stone (van Biezen et al. 2012).

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the Economic and Social Research Council (ESRC) [grant number ES/M007537/1]. The authors would like to thank ESRC for this grant.

Notes on contributors

Monica Poletti

Monica Poletti is an ESRC postdoctoral fellow in Politics at Queen Mary, University of London on the Party Members Project (PMP). Her research focuses on political behaviour and parties, and has covered a range of topics including party membership, social/political participation, electoral politics and populism. She is also part of the COST-Action project The True European Voter. [[email protected]]

Paul Webb

Paul Webb is Professor of Politics at the University of Sussex, researching party and electoral politics. A Fellow of the Academy of Social Sciences, his most recent book is Organizing Political Parties: Representation, Participation and Power (Oxford University Press, 2017), which he co-edited with Susan Scarrow and Thomas Poguntke. He is co-editor of the journal Party Politics. [[email protected]]

Tim Bale

Tim Bale is Professor of Politics at Queen Mary, University of London. His research focuses on political parties, and has covered a range of topics, including coalitions, membership, links with interest groups, and how parties on both the left and right have coped with the challenge posed by mass migration and the populist radical right. His latest, co-edited book is Left-of-Centre Parties and Trade Unions in the Twenty-First Century, and he is the author of a widely used textbook, European Politics: A Comparative Introduction. He tweets @ProfTimBale and his writing for the media is collected at proftimbale.com. [[email protected]]

Notes

1 YouGov recruited the survey respondents from a panel of around 300,000 volunteers who are paid a small reward for completing a survey. Upon joining the YouGov panel volunteers complete a survey asking a broad range of demographic questions which are subsequently used to recruit respondents matching desired demographic quotas for surveys. Potential respondents for the party member survey were identified from questions asking respondents if they were members of any of a list of large membership organisations, including the political parties. Results reported in this article are not weighted in any way since there are no known official population parameters for the various party memberships. However, previous YouGov party membership surveys using unweighted data have generated predictions for party leadership contests that came very close to (that is within 1% of) the final official outcome, which gives us confidence in the quality of the data.

2 Note that ‘supporters’ in this article should not be confused with those who are officially ‘registered supporters’ of the Labour Party. This is a new category of party adherent that was introduced under Ed Miliband’s leadership; while it is possible that some of those who fall into our party supporters category actually are registered Labour supporters, the overwhelming majority will not be – not least because not all British parties have similar schemes. We have therefore not sought to treat officially registered supporters as a separate category for the purposes of our analysis.

3 The individual items on which these scales are based are as follows: Please tell us whether you agree or disagree with the following statements: Government should redistribute income from the better-off to those who are less well-off; Big business benefits owners at the expense of workers; Ordinary people do not get their fair share of the nation’s wealth; There is one law for the rich and one for the poor; Management will always try to get the better of employees if it gets the chance. Respondents could select from the following options in answering each of these questions: (1) Strongly agree; (2) Tend to agree; (3) Neither agree nor disagree; (4) Tend to disagree (5) Strongly disagree; (6) Don’t know. Don’t knows are excluded from analysis, and all left–right item responses are coded so that 1 is the most left-wing option, and 5 the most right-wing option. These scores have then been normalised to a scale running from 0 (left) to right (10).

4 The individual items from which the liberty–authority scale is constructed (in similar fashion to the left–right scale) are as follows: Young people today don’t have enough respect for traditional values; People who break the law should be given stiffer sentences; For some crimes the death penalty is the most appropriate sentence; Schools should teach children to obey authority; Censorship of films and magazines is necessary to uphold moral standards. 0 = liberal, 10 = authoritarian.

5 Note that we are, however, able to consider the impact of collective policy incentives in the ‘different narratives’ section of this article (see Table 3).

6 Note that the overall left–right combined mean scores for members and supporters are as follows: Conservative 5.7, Labour 1.5, Liberal Democrat 2.99, UKIP 3.03, Greens 1.46, SNP 1.69.

7 This is so even though political careerists may have become proportionately more significant in the context of long-term membership decline.