The system depends on some endorsers being willing to review works by people unknown to them. Otherwise, it's nothing but an old boys' network where one cannot get an endorsement for a new field unless you know someone who is an endorser in that field. I've needed that done for me, and I've done it for others.But we're drifting. Do you think it's the responsbility of endorsers to review works by people unknown to him that do not have greenlighted institutional affilaitions? I don't. I get a few every year, and I send them straight to the trash. Partly because I am an experimeter and these are always theories.
And an endorsement review is not near as involved as a real peer review. When I do a real peer review, I spend many hours on the paper, reading it very carefully, reading the references, considering alternate approaches, making sure the discussion and conclusions are well supported by the results, etc. For an endorsement, I spend an hour or two. I read the paper and if it's not complete garbage, I grant the endorsement.