Went into the data and took the top 1000 individuals with self cite percentages over 40%, then sorted by institution. Nearly every major institution had individuals in this group: Johns Hopkins (4), Cal Tech (4), Georgia Tech (2), MIT (5), each of the Max Planck Institute campuses (3-7), Moscow State (7), Penn State (6), Stanford (1), Utrecht (2), University of Zurich (4), ETH Zurich (1), DLR (3), Imperial College London (3), University of Tokyo (2), Princeton (5), Kyoto University (4)...
I feel like if this problem were very concerning we'd see the distribution concentrated at certain institutions but I'm not sure there's one with over 10 researchers at them. We hear a lot about questionable Chinese journals, but the highest institution in this list is the Chinese Academy of Sciences with 3 individuals.
I think the more likely case is there are a few bad apples, some bad practices we can't ever fully get rid of, and that some research lends itself more to self-citation.
I feel like if this problem were very concerning we'd see the distribution concentrated at certain institutions but I'm not sure there's one with over 10 researchers at them. We hear a lot about questionable Chinese journals, but the highest institution in this list is the Chinese Academy of Sciences with 3 individuals.
I think the more likely case is there are a few bad apples, some bad practices we can't ever fully get rid of, and that some research lends itself more to self-citation.