This is a terrible metric for determining when to evaluate teacher performance. All you're doing is encouraging teachers to grade easier and punishing teachers who actually want their students to learn the material.
err, actually this is being done in many school systems. And it's not just the grades that the teachers are giving out. It is also the grades from standardized tests are being used. If a teacher is grading easier, the results will show in the standardized tests with discrepancy between the two.
As for Marlin, yes it takes time to weed out unproductive or bad teachers. But once a bad teacher has been established, if there is time to correct it then do so. If not, boot them. It's simple. As for new teachers, unless there is a specific complaint that needs investigating, there is NO REASON to fire them. Actually it is more cost effective to keep them on.
If once all the bad teachers are let go during a lay off for budget reasons, and more still need to be fired, THEN I support senority for determining who goes and who stays. If you are left with two acceptable performing teachers, it is harder to gauge how much more acceptable a new teacher is over an older one that is performing.
As for cameras, umm, you don't need a camera in every room either. A few per school hidden behind a "black box" that gets changed around is more than sufficient to accomplish this task. It would not be expensive either as I know many poor schools that have done this not only in class rooms but on busses as well. You are acting stupid because you are failing to use the logic I know you are capable of using to think this through. There are many other ways to measure a teachers performance, some more cost effective and faster than others, but the fact that you can't think of them doesn't mean squat unless you are in the field. Here is a list of some ways, but by no means is this list all inclusive, that teacher performance is tracked in different places.
1) Tracking student performance both in grades and standardized test
2) Tracking student and parental feedback
3) Random screening either through camera's or walk ins
4) Peer reviews from other teachers
5) Workshops and testing on teachers to grade their knowledge from time to time on how to teach
There are other ways and combined they give a more accurate picture.