I have once, some time ago, received a rejection notice in which the journal editor made what I thought was a peculiar comment. The reviewers all found that the work was good, but one emitted a doubt that I had chosen the right journal for it, saying the paper might not appeal to the broad readership of the journal (it was a general physics journal). So far, nothing out of the ordinary.
However, the editor indicated that his judgment to reject the paper was based on the fact that very few of the citations in the manuscript actually referred to the journal I had submitted it to. (Like, 2 citations out of 35. Some of the other citations were to other general physics journals, some to more specific journals.)
Back then, this looked very weird to me. To some extent, it could be interpreted as a push by the editor to increase self-citation of his journal. It has never occurred to me since.
Is it common practice? When does it become ethically wrong?