1: \begin{abstract}

2:   We consider {\em network contribution games}, where each agent in a

3:   network has a budget of effort that he can contribute to different

4:   collaborative projects or relationships. Depending on the

5:   contribution of the involved agents a relationship will flourish or

6:   drown, and to measure the success we use a reward function for each

7:   relationship. Every agent is trying to maximize the reward from all

8:   relationships that it is involved in. We consider pairwise

9:   equilibria of this game, and characterize the existence,

10:   computational complexity, and quality of equilibrium based on the

11:   types of reward functions involved. When all reward functions are

12:   concave, we prove that

13:   % a pairwise equilibrium always exists, and

14:   the price of anarchy is at most 2. For convex functions the same

15:   only holds under some special but very natural conditions. Another

16:   special case extensively treated are minimum effort games, where the

17:   reward of a relationship depends only on the minimum effort of any

18:   of the participants. In these games, we can show existence of

19:   pairwise equilibrium and a price of anarchy of 2 for concave

20:   functions and special classes of games with convex

21:   functions. Finally, we show tight bounds for approximate equilibria

22:   and convergence of dynamics in these games.

23: \end{abstract}