Association for Public Policy Analysis & Management

Failure of public service provision can drive users to exit or voice concerns about them, including by complaint, or even to take action to remedy defects through coproduction. However, a body of research in economic psychology emphasises how modes of interaction are contingent on whether people see themselves engaged in market or non-market activity. This body of literature assumes that users would be willing to coproduce failing governmental services that have clear collective benefits (for example by cleaning streets or parks, or participating in neighbourhood watch programs). Yet, if the same services would be provided by private companies, users would not anymore be willing to coproduce because of concerns that donated labour will be expropriated for private, instead of collective, benefit. This institutional effect will even be stronger for users who identify themselves as civic-republican citizens vis-à-vis public services. We hypothesise that the service being framed as being provided to citizens (versus customers) will make them draw more on citizenship concerns of rights and duties. In contrast, users identifying themselves as customers will behave more along a market-oriented logic and thus be less willing to coproduce. This has important implications because key public services nowadays are increasingly provided through non-governmental service providers. Some of these providers communicate with their service users referring to them as customers, while others treat them as citizens. We will investigate whether the publicness of services (i.e., their ownership status), and users’ role identities towards these services (i.e. citizen versus customer) affect their willingness to coproduce failing services.

Using a 2 x 2 factorial within and between-subjects design, we assess responses to failure in a tax funded service contingent on whether the service is framed as being provided to local ‘customers’ or to local ‘citizens’ (treatment 1), and whether delivery is provided directly by a local government or a private company as a contractor (treatment 2). The failures will use previously validated photographic descriptions of service failure. The research has been piloted using laboratory based convenience samples but will be supplemented, in summer 2015, using a large online representative panel sample in the US. The service areas will reflect three randomly ordered services that reflects actual variation in the way these services are provided in different local contexts: 1) cleanliness of streets, 2) local public security services and  3) local park play equipment.

We discuss the implications for the literatures on the potential for coproduction of public services and for the provision of services by local governments or contractors, both of which are subjects of much current debate.