The aim of the study is twofold. First, to provide a picture of what happens when groups of citizens cooperate with municipalities and administrations to produce services essential to the community, i.e., elderly care or road maintenance. Second, to compare this picture with the picture of citizens’ involvement that the civil society theories describe. This is done by comparing four different cooperation projects. The empirical material has been gathered through four qualitative case studies – two elderly care cases and two road maintenance cases – and the analytical frame has been drawn mostly from organization theory, especially the resource dependence and the institutional perspectives. In the dissertation it is shown that in the projects with less complications the processes developed in a way that balanced, to some extent, the asymmetry in the dependence relation, i.e., the resources controlled by the groups became more interesting for the administrations and municipalities. These processes did also develop in a way that made it possible for the actors to come to an agreement of what problem the project was supposed to solve. These findings covariates with how interested the municipalities and the Road Administration organizations were to participate in the cooperation projects. It also covariates with the use of institutionalized cooperation forms. The short cut of an already defined and legitimated cooperation form implied that less transaction resources had to be invested in the cooperation itself – but as a result the actors did not communicate sufficiently and therefore did not develop a mutual understanding and trust. Another finding is that both the groups and the municipalities and administrations had pragmatic motives for their involvement in the cooperation projects, which led to an organizational form that was effective for the purpose of solving the identified problem with the elderly care/road maintenance, but not for the unintended consequences described by the civil society theories. As the group of citizens really involved was small, the consequences – greater solidarity and responsibility, and a decentralized democratic process, only comprised a few, mostly resourceful, citizens. Finally, the study shows that the groups’ contributions to the democratic process were limited by their involvement in actually solving the problem in question, i.e., to build and run an elderly home or to work with the improvement of the roads. The findings suggests that the picture of citizens’ involvement often put forward in the political debate in Sweden – as both a complement to the service provided by the public sector and a way to improve the democratic process – ought to be the subject of further research.