Smart environments are ecosystems, which seamlessly embed IT assets into physical world's objects and hold promise for improving the services we receive from our social and economic ecosystems. The management of smart environment assets in multi-partner, dynamic collaboration scenarios where different sets of assets are owned and operated by different partners is a non-trivial problem, due to restrictive asset sharing policies applied by collaborating partners. In this work we formalize, evaluate and compare two asset sharing policies, investigating their impact on MSTA-P, a policy-regulated version of an existing asset-task assignment protocol. The first sharing policy is based on a traditional asset ownership model while the second is based on an edge model allowing asset sharing among collaborating partnes through cross-partner team formations. We find that while the traditional ownership model allows slightly better performance, the difference is only marginal, so a team-sharing model offers a viable alternative sharing approach. © 2014 IEEE.