I have further explained, that I do not find his work particularly convincing because he does not compute anything, so one can not even say whether it was already in disagreement with observation if it was correct (and it seems to me the effect he is claiming exists should be substantial even in the asymptotically flat region).