You are right about your comments on the carbon nanotubes. They are generally hydrophobic. I would say that the lack of adhesion (i.e. hydrophobicity) on the CNT walls is part of the reason for improved flow through them. The other reason is that the structure of the water also changes due to the size of the pore. Because the pore is so small, the water will form hydrogen bonds inside the tube, decreasing the disorder of the flow and therefore also the resistance. Both these effects play an important part.