Browsing by Author "Ylikoski, Ilona"

Team problems are inherent in today’s world, as individuals often must collaborate in various settings from production to service provision. The need for collaboration originates from a group of agents’ pursuing a common goal. This dissertation is a literature overview of what kind of inefficiencies arise from these kinds of projects, and how to possibly overcome them. The focus is on games that consider the provision of public goods, or in other words, large-scale voluntary contribution projects. Examples of these kinds of projects are, for instance, library collections, public parks and environmental protection. What characterizes these projects, is that collaboration is not only a one-time effort but requires continuous work over multiple periods. This dynamism together with voluntary commitment, results in to the fact that the effort is not only underprovided but also distorted. Participants in these projects face the incentive to hold back their own contributions in order to free-ride on others’, but, at the same time, want to frontload. The previous means that the participants want to contribute to the project in order to encourage others to increase their future contributions. So, these kinds of projects are inefficient in the sense that the effort provision is distorted, and the projects tend to last longer than optimally. The previous happens as the participants fail to cooperate as efficiently as in the social optimum. In order to understand the underlying reasons behind this cooperative failure, I deconstruct a dynamic and voluntary game that considers the provision of a public good. In the game, a group of agents must exert costly effort over time to contribute to a project which is divided into sub-projects. The game ends once a prespecified threshold amount of effort is accumulated. Participants of the game can base their contribution decisions only on the commonly observed state of the project and how they expect others to behave. Individual contribution costs and levels are private information. The examined model presents that the underlying reason behind the inefficient outcome are the facts that agents’ contribution costs are hidden and that the agents move simultaneously. Social optimum would be achieved if group size could increase without bound. When group size cannot tend to social optimum, the agents must control the degree of dysfunctional behaviour by forming smaller groups and form arrangements to control agents’ incentives to deviate from cooperative behaviour. Examples of these kinds of systems are self-imposed sanctioning systems, sanctioning systems with a centralized authority, deadlines and infrequent monitoring. The strengths and weaknesses of these systems are briefly discussed in order to understand how feasible it is, in fact, to restore efficiency for dynamic games that rely on voluntary contributions. To further help the reader understand the deep-rooted role of free-riding and frontloading in these kinds of projects, few extensions to the model are covered from complementary contributions to uncertainty in outcome.