So where is your data to show the level of risk that is mitigated by the formal final tests in the other examples you gave? How do you know it's not just self-perpetuating intuition?
Setting a standard and requiring it to be met does not inexorably entail a formal final test.
We all know people who've passed formal final tests but are nonetheless incompetent to fly/drive/sail/ride.
Let's reverse your argument. I reckon ATPL holders should be subject to another flight test before each and every flight in which the ATPL holder proposes to exercise the privileges of that licence. What is the safety objection to that requirement?
If you say the risk mitigated by that requirement is so vanishingly small as to not justify its cost and inconvenience, where is your data to prove that?
I note there are lots of people exercising the privileges of an ATPL who have never been the subject of a formal ATPL flight test. Why are they not considered 'unsafe'? Or is it that they are merely tolerated as being 'acceptably unsafe'? How 'unsafe' are they, compared with the pilots who have passed the formal ATPL test?