“This is, of course, a very unrealistic assumption … these operations are often extremely costly … to prevent many transactions.”
That was him referring to the oft-straw man argument by opponents of Coasean insights about “transactions costs being zero.” Coase obviously knew that. And he obviously was making a different point. The Coase Theorem as you are told about it is not the Coase Theorem as is. In any case, Coase did not coin the term Coase Theorem either.
Reminded me of a line in Professor Landsburg’s Armchair Economist when he is talking about assumptions, not just in economics but in other fields as well (physicists assume no friction, etc.).
“Assumptions are tested not by their literal truth but by the quality of their
implications.”