Voluntary Pooled Public Knowledge Goods and Coalition Formation

In this paper we develop a theoretical model of the mechanisms behind the voluntary provision of public knowledge goods in coalitions in presence of social preferences. The model builds on the large empirical literature on voluntary production of pooled public knowledge goods, such as source code in communities of software developers or data voluntarily provided to open access data repositories. This literature shows that the provision of public goods is strongly dependent on the presence of social preferences such as group identity and social approval of individual pro-social attitudes. To integrate these effects in standard public good theory this paper builds a private-collective model of public good provision, where contribution to public knowledge goods generates both public and exclusive private benefits for the members of the coalition only. The analysis shows that, when the private benefit is important, the effect of the social preferences on the coalition formation is ambiguous. In particular, in the latter case, in presence of strong individual reputational effects, the public knowledge goods will be more difficult to produce. The comparison of the predictions of the theoretical model with the stylized facts of large scale surveys of Free/Libre/Open-Source (FLOSS) software developers confirms the results of the model.