We argue that voice is too simple a concept – there are different types of voice, collective (e.g. voting, group membership) and individual voice (e.g. complaining). We suggest that collective voice is harder to organise because of coordination problems whereas individual voice. We argue that individual voice does not trade-off with exit, collective voice does. Exit takes different forms too: moving providers within jurisdiction, moving jurisdiction, and exit to private services. We suggest there are different relationships between these variables. As a result of these characteristics, we find that many of the arguments both for and againsts choice need to be carefully considered, and we have reviewed these in the contest of the research evidence to show that the arguments for more choice need to be carefully considered and applicable to the context in which they occur. In the empirical part of our project, we test whether the possibility of exit makes voice less likely, which suggests that the greater complexity at the local level and new provision introducing more choice, allowing for greater exit opportunities, should cause the quality of local participation to suffer. We also test whether loyalty in the form of the level of social investment in a community or social capital should increase the probability of voice over exit. We used a panel survey of a representative group of 4000 internet users, asking questions about service satisfaction, use of private services, other exit opportunities, social investment into the local community, and the level of participation, which we repeated a year later. This research applies the ideas expressed in Albert Hirschman’s book, Exit, Voice and Loyalty. He argued that there is a trade-off between exit and voice, which may decrease efficiency because the voice of consumers keeps organisations responsive whereas exit leaves the poor performers behind. In public services, citizens may exit from public provision by moving across jurisdictions and/or moving to other private sector providers. Citizens may use collective voice through: voting and pressure group activity or private voice by personal complaints and comments to public officials. We will test whether the possibility of exit makes voice less likely, which suggests that the greater complexity at the local level and new provision introducing more choice, allowing for greater exit opportunities, should cause the quality of local participation to suffer. We also will test whether loyalty in the form of the level of social investment in a community or social capital should increase the probability of voice over exit. We will use a panel survey of a representative group of 2000 internet users, asking questions about service satisfaction, use of private services, other exit opportunities, social investment into the local community, and the level of participation.

Internet survey of a representative sample of users(individuals, internet users) from YouGov's panel. Aggregate data sources from Census, from Local Elections Unit, University of Plymouth, and from political data from local authority websites. Samples: 4067 respondents in Wave 1, 2610 for Wave 2. 741 variables (including 32 variables added from census and local authority data).