The institution offering the education must be able to test if their education is effective, education doesn't work well without tight feedback loops. Neither teachers nor students are drones whose behavior can be measured by some perfect third party process or human(s).

That said what passes for education at most places works best when you do start treating teachers and students as drones. The logical conclusion of that is something like Khan Academy with spaced repetition, and the savings in time and money worldwide would be tremendous if that system were to be embraced by the US department of education. What passes for education is definitely worthwhile for some things, but it's not really in the traditional spirit of education.

Of course the education institutions need to be able to get feedback on their teaching if they are to be effective (the tighter the loop the better), they should just not be charge of determining if it is effective.

In the UK we have examination boards for school level exams, I'd love to see that extended to degree qualifications as well.

Didn't you have a major problem with schools choosing the examination board that marked the easiest?

Yes, and we also have grade inflation - we used to mark on an A-E scale but now we have an A* too. BBC article, with interesting graphs: http://www.bbc.co.uk/news/education-11012369

Yes, but I don't think that is too difficult to resolve.

Yes don't let the schools choose who is going to be assessing if their teaching is effective.