Since the intention is to have the same data at multiple locations, Peer-to-Peer replication would be the most appropriate. Peer-to-Peer is the only out-of-the-box technology that SQL Server provides to handle writes at multiple nodes, so it makes sense to use it. (For completeness of this answer, Peer-to-Peer does require Enterprise edition, while Merge replication does not.)
Once you get the security of the topology ironed out (I found this MSDN page very handy), it's actually not that difficult to set up. If you're considering Merge replication, chances are your schema is already is decent shape, so conforming to the requirements and restrictions for Peer-to-Peer shouldn't be a problem.
While I don't know your exact data requirements, it seems like writes would be nicely partitioned (i.e., data entered on one side probably never needs to be edited in multiple locations after that), meaning update conflicts would be extremely unlikely.