Its worth listening to the recent Background Briefing about the issues of paying teachers for results. It does indeed beg all sorts of questions, not the least of which is the degree of objectivity that might be brought to assessing the idea of "success".
Does Mathematics success mean that a Year 12 student can accurately reproduce the Sum and Difference Formulae (this is the point at which Maths and I parted company...why learn stuff that you can reference more easily in a book) or that they are able to amortise a housing loan, or accurately cost a building project?
More seriously is the issue of how teachers start to "Teach the Test". We all know the problem. If your pay begins to depend on how your charges perform under a certain style of assessment then your method begins to reflect the process which will maximise the results under that method of assessment. Certain schools have masterd this art for years.
More seriously are other incontrovertible truths like parental educational levels influence performance at least as much as teachers. Middle class children with well-educated parents are likely to do well in our system. There are exceptions, bright working class kids who float to the top, but having worked in a number of educational environments where kids do not start with the advantage of having tertiary educated parents it is all too evident that these kids begin at something of a disadvantage.
Some might say this doesn't matter...they are the cannon fodder. But I would want to reward the teachers who struggle with the most disadvantaged and get them to take a few steps forward rather than the "successful" socially advantaged kids who have everything goign for them.
It is a very dangerous game that we play!!