In his book `Exit, Voice, and Loyalty'[1], A. O. Hirschman discusses his theory of exit, voice, and loyalty options in terms of consumers who feel that the products of a particular company are degenerating. Although his theory is stated in terms of businesses and the market, he also briefly discusses a further possible application to the dissatisfaction of interest group members with a group, specifically when the group in question is a political party. This is how I intend to apply the theory.
Hirschman's theory states that if a customer is dissatisfied with a particular product, they have the choice of either complaining to the firm - `voice' - or simply taking their custom elsewhere - `exit'. The voice option is broadly interpreted to include any effort that is an attempt to change a disagreeable state of affairs rather than escaping from it. The combination of voice and exit, Hirschman claims, provides a mechanism whereby firms can identify a problem and correct it before they lose too much custom. Similarly, falling membership rolls and vocal dissatisfaction of members could enable a political party to identify areas of discontent. The question is whether this reaction is occurring currently within the Labour Party, where members are dissatisfied with the leadership and the direction of the party[2]. Is the party indeed identifying such problems, and if so, is it then choosing to make suitable changes, as a business would have to in order to avoid loss of custom and subsequently of profit? In Chapter 4, I shall address the question of whether making changes in response to members' discontent is in fact sensible from the point of view of the party leadership, as well as examining the factors influencing members' choice of options; in the current chapter I shall look at the theoretical issues.
Crucially, the voice and exit options interact with each other, and also with a further factor/option, loyalty (discussed later). Exit is dependent on the ability and willingness of customers to take up the voice option. Customers may delay exit if they feel that voice is likely to be successful; that they can change the situation rather than having to find an alternative.
Relevant to this decision is the issue of whether there is any available alternative. A readily accessible alternative product (especially if it is of equal quality) will, Hirschman argues, be likely to increase the numbers of people who choose simply to exit; voice would cost too much in terms of time and effort. However, in a situation where there is no sufficiently similar alternative - for example, when the only alternative option is of poor quality - customers are more likely to attempt to rectify the situation. The customer with nowhere else to go is likely to be the most vocal, and thus is the most likely to affect quality. This seems likely to apply in particular in politics. If a member is unhappy with the direction in which a political party is going, is there always an alternative? In Chapter 4, I shall argue that in this country there is in fact no appropriate alternative to the Labour Party, due largely to the structure of the British political (specifically, voting) system. Certainly, it is clear that in general, someone who already feels strongly enough about politics to join a party (rather than simply voting in elections), and who also feels strongly enough, and is sufficiently dissatisfied, that she will choose between exit and voice options (rather than simply ignoring the problems), is unlikely to wish to absent herself entirely from active politics. Thus, it seems clear that the lack of a viable alternative will seriously affect the decision of members to exit.
A point made by Kitschelt is also relevant here. He argues that leaders of parties wish to remain leaders and thus will strive to minimise party disunity, to avoid activists leaving to go elsewhere.[3] In other words, they will be likely to pay attention to complaint. However, if there is no alternative for activists to exit to, will this still apply? If the lack of alternative makes activists significantly less likely to exit, this impetus for the leadership to minimise party disunity is lost, and they may instead choose to ignore the discontent (although there will be other factors affecting this decision, which I will consider in Chapter 4).
A further issue mentioned by Hirschman is which customers will choose which options. Ideally, a firm needs a "cushion" of less alert customers, who will provide sufficient time and money for the firm to make changes after the more alert customers either exit or complain. Most customers are of this type, and it seems likely that most members of any political party will be - most party members are not activists. It is, then likely that the customers who care the most, who would thus be the best agents of voice, are those who choose to exit first, as they are also the most likely to notice the changes. However, as mentioned above, this depends very much on the availability of alternatives. If there is no realistic alternative, these customers are more likely to stay and complain - because they care, it is worth their while to do so. This applies directly to political parties and their active members. The most active, and the most concerned, are the most likely to choose the voice option in the case where there is no alternative.
A factor having a further impact on the interaction between voice and exit options is loyalty. Hirschman points out that loyalty will automatically tend to increase the influence of voice: the customer will prefer to expend extra energy on voicing their opinion, rather than take their custom elsewhere. It can neutralise the tendency of the most quality-conscious customers, in an environment with available alternatives, to be the first to exit.
It is important to note that loyalty can be seen in two different lights: as a third alternative to voice and exit, or as a factor affecting the choice between voice and exit. Hirschman discusses it purely in the second sense; he seems not to see it as an alternative option. This could be because, depending on how `loyalty' as an alternative is precisely defined, loyalty and voice may not be mutually exclusive. Indeed, if `loyalty' refers simply to `not exiting', then loyalty is a prerequisite for voice. However, loyalty can also be defined as failure either to voice or to exit, in which case it fills the gap left by the other options. Loyalty as a factor affecting the decision between exit and voice can be seen as a post-hoc equation filler. It explains the fact that customers/interest group members may continue to choose the voice option long after the exit option would appear to be the more rational choice. Once loyalty is considered, the rational decision alters. I shall consider loyalty in both of these potential lights: as an alternative, and as a factor.
When considering loyalty as a factor, it might at first appear that it will have less effect in an environment (such as the one we are considering) with no real alternatives. However, it is in fact also likely to increase the pressure to utilise the voice option even in such an environment - the consumer feels a certain identity with the product and wishes to help save it. This factor is likely to apply with particular force in political parties. Party members who feel strongly about alterations in party policy are likely to be those who feel most strongly about the party itself: those with a strong sense of loyalty. This strong feeling about the party is likely to increase the impetus to protest about policies with which they disagree; they wish to rescue `their' party. Political parties can inspire a sense of ownership in this sense to a much greater degree than would necessarily be commensurate with the actual impact which an individual member may be able to make.
When considering loyalty as an option, it has two forms - loyalty with, and loyalty without, voice. Clearly, any member choosing voice is also choosing loyalty, at least to the extent that they are remaining within the party. The question is, is there a point at which the costs of choosing voice outweigh the benefits, and if so, will members then opt for exit, or for silent loyalty? This will depend on how they analyse the benefits and costs of loyalty and of exit (considered in Chapter 4). I shall interpret loyalty in this context, unless specified otherwise, to be silent loyalty.
A vital factor when considering Hirschman's theory in political terms is that party members are involved in both the supply and demand sides of the organisation. This means that, even if they choose to leave, the quality of the organisation will continue to affect them. One factor in this is loyalty, as discussed above - customers who leave continue to care about a product, or members continue to care about a political party. More important with regard to political parties, particularly when they are (as Labour currently is) in government, is the fact that the organisation continues to affect their lives even after departure. Labour's policies affect all citizens, whether or not they are party members (even in opposition, this would be true to some extent; when in power, it is obviously true to a very significant extent). This means that full exit is impossible, and further increases the likelihood of members choosing to voice their complaints rather than leaving.
The decision taken by a customer on whether to voice or exit thus involves several related factors: the chance of the voice option being successful; the availability of an alternative; and the cost of the voice option weighed against these factors. If you feel that your complaint is unlikely to make a difference, you may instead choose to cut your losses.
In order to apply Hirschman's theory to the issue of Labour Party members disagreements with the leadership, there are several questions which need to be answered. What are the currently available voice options? How valuable are these options in actual fact, i.e. how much attention is the leadership likely to pay to them? Are there any viable alternatives to the party for dissatisfied members? And finally, what are the other costs and benefits of each of the options? This includes considering the attitude of Labour Party members in terms both of loyalty, and of whether they continue to care about the party after leaving. I will consider these questions in the following chapters, after assessing in the rest of this chapter the importance of the members to the leadership, which will have a very significant effect on the decisions of both sets of actors.
An important aspect of the rational choice assessment of membership and leadership behaviour is the actual value of the members to the party. Financially, there is of course benefit to having individual members, as they all pay dues. However, this is not a major source of funding - most of the Labour Party's funding comes from trade unions and individual donations. This is therefore unlikely to be a major consideration for the leadership in deciding on responses to dissatisfied members - there is always potential to make the money up from elsewhere. A further use for rank-and-file members is in terms of local government - the party needs people in local constituency organisations who are prepared to be councillors, local school governors, etc. However, again, this is unlikely to be a major concern - the strong likelihood is that there will be more than enough New Labour local members to fill these posts, and these would clearly be preferable to the dissatisfied members. Against this point, it is worth noting that what might be described as the `New Labour' members (those who joined the party since 1994) appear to be notably less active, according to Whiteley & Seyd's 1997 survey, than longer-term members, at least in terms of constituency campaigning and attendance at party meetings.[4] This lack of interest seems likely to extend to involvement in local government, which can occupy significant amounts of time.
There are also other benefits to having members. Clearly, Labour must recognise at least some of these, since they are committed to extending their individual membership to 500,000 by 2001. Seyd identifies three further reasons for needing members.[5] Firstly, it enables the leadership to respond to the political disaffection of the public by demonstrating a commitment to openness and participation at a grassroots level. Secondly, it is a means for the leadership to demonstrate to the public that they are in control of their party; and finally, it enables them to bypass the activists. This final reason, however, indicates that it is docile members, who agree with New Labour, that Labour are looking for: not the sort of members who might vocally disagree with them. It appears that New Labour are actively attempting an increase in docile members - giving them greater financial benefits but less problems with attempted interference. Thus, the value of disaffected members may be lowered significantly. Members leaving because they lack influence would also help in showing party control. However, it would tend to count against the image of participation, if an exodus becomes publicly obvious.
The most important area of potential value of rank-and-file members is electoral campaigns. Many commentators have argued that in the current era of highly centralised, media-orientated campaigning, the role of the local constituency organisation (and thus the local activists) has decreased to the point of being unnecessary. In the Nuffield election studies, local constituency campaigning is frequently described as a `ritual', and the dominant view is that the strength or intensity of a local campaign makes little or no difference to the result of the election, although it is conceded that it can make a difference in a marginal contest. In the 1997 study, Curtis & Steed comment that `The 1997 election does not seem to support claims made that local campaigning can make a difference... The Labour Party targeted 90, mostly marginal Conservative constituencies... Yet... the performance in these constituencies... was very similar to that in other Conservative/Labour contests.'[6] In opposition to this, Whiteley & Seyd point out that, according to their survey data, only 9% of party members did any work at all in another constituency, and only 3% worked `most' or `all' of the time in another constituency.[7] This implies that targeting did not in fact work, and that the reason that there were good results for Labour across all types of constituency was simply that there was a lot of work done in all types of constituency! Denver and Hands also point out that `to a large extent [the conventional wisdom] derives from the... Nuffield studies..., although they rely, in the main, on impressionistic evidence.'.[8]
There is a significant body of evidence against the conventional opinion. Seyd and Whiteley, looking at the 1992 election, were able to show that membership levels in constituencies had a statistically significant impact on the vote. Indeed, a 25% higher membership level in just 12 key marginals in 1992 would have removed the Conservative majority.[9] Denver and Hands, also examining the 1992 election, also show that for Labour, there is a significant positive relationship between the intensity of the constituency campaigns and the change in its share of the vote.[10] Indeed, their figures given the (hypothetical) results that in constituencies where the Labour campaign was well above average strength, Labour received 7% more of the vote than they would have done had the campaign been well below average strength. They comment `there would clearly be a substantial penalty to be paid if the strength of local campaigning were reduced'.[11] Seyd et al point out (again, referring to the 1992 campaign): `Particularly in close-fought contests, an effective and energetic local campaign can make the difference between winning and losing. Political parties ignore the work of their local members at their electoral peril.'[12] In support of this, Holt and Turner's study of a marginal London constituency during the 1964 General Election led them to the conclusion that `our research strongly suggests that organisation was an important influence in the 1964 election in Baron's Court'.[13] Seyd also notes that, with strict local campaign spending limits, and in the absence of large-scale state funding, the voluntary work of local activists is important in campaigns, particularly for local government, which can often be closer-run than national campaigns.[14] A further point worth bearing in mind is that even if the campaigning at a local level were to have little effect, ceasing, or even winding-down, campaigning and leaving the field open for other parties would be likely to have a serious negative effect.
It appears, then, that there is significant evidence that the work of the members in electoral campaigns is important to the leadership of the party. If large numbers of members were to choose to leave, this could have a serious impact on the party's electoral chances. It must be borne in mind, though, that this effect is only evident in marginal constituencies. Seats where Labour has a large majority are unlikely to be affected by the strength (or lack of it) of the local campaign. However, despite Labour's current overall large majority, a significant number of these seats were won on the strength of abnormally large swings, and also have quite small majorities. 65 of the seats won by Labour had majorities of less than 5,000.[15] The results in these seats might well be affected by local campaigns. However, there are also other factors affecting this potential source of membership leverage, most notably the level of credibility of this threat given by the issue of alternatives. If there is no available alternative, and if a member by leaving runs the risk of allowing the Conservatives back into power, is it really sensible for them to leave, after all? This will depend on their judgement of the preferability of a Conservative government against a New Labour government, and it seems likely that a New Labour government, whilst not ideal, would be judged preferable. As Tony Blair has repeatedly said to dissatisfied activists: `The choice is not between a New Labour government and the Labour government of your dreams, it is between a New Labour government and a Tory government.' This is a factor which the leadership will take into their calculations when deciding whether or not to react to the dissatisfaction. I will address these, and other factors, more specifically in Chapter 4.
| [1] | A O Hirschman Exit, Voice, and Loyalty |
|---|---|
| [2] | The Seyd & Whiteley 1997 report on Labour's membership showed that 2/3 favour redistribution of wealth, 2/3 do not want production to be left to the free market, 3/4 favour increased public spending, and 1/2 want nationalisation: all of which opinions seem to be out of step with Labour's current policy direction. (source: Tribune 10.4.1998) |
| [3] | H Kitschelt The Transformation of European Social Democracy (Cambridge: CUP 1994) |
| [4] | P Whiteley & P Seyd `Labour's Grassroots Campaign in 1997' in Denver, Fisher, Cowley, Pattie (eds.) British Elections and Parties Review 8 (London & Portland OR: Frank Cass, 1998): p201 |
| [5] | P Seyd `News Parties/New Politics?' Party Politics 5, 1999: p385 |
| [6] | A O Hirschman The British General Election of 1997 (Macmillan, 1997): p312 |
| [7] | P Whiteley & P Seyd `Labour's Grassroots Campaign in 1997': p197 |
| [8] | D Denver & G Hands Modern Constitutency Electioneering (London & Portland OR: Frank Cass, 1997): p267 |
| [9] | P Seyd & P Whiteley `Labour's Vote & Local Activism', Parliamentary Affairs45: p589ff. |
| [10] | D Denver & G Hands `Constituency Campaigning', Parliamentary Affairs 45: p543 |
| [11] | D Denver & G Hands Modern Constituency Electioneering: p302-3 |
| [12] | Pattie, Whieley, Johnston, & Seyd `Measuring Local Campaign Effects', Political Studies 42: p479 |
| [13] | R T Holt & J E Turner Political Parties in Action (New York: The Free Press, 1968): p298 |
| [14] | P Seyd `New Parties/New Politics?': p385ff. |
| [15] | Source: http://www.election.demon.co.uk |