Lately we have seen increased interest in paying teachers based upon "merit" rather than according to contracts. The theory is that merit pay would allow good teachers to rise to the top and bad ones to leave. The second appeal is based on belief that you get paid for the value of your contribution. Both arguments have their flaws.

News of bonuses paid to Wall Street execs responsible for the economic collapse undermines the argument that merit pay allows the cream to rise to the top. Also, it is certain that the size of the Wall Street bonuses dwarf what might be paid to teachers. Would adding a few thousand dollars to the checks of some really inspire the others to do better? The argument that somehow this will help the good teachers rise to the top and the bad ones go away is not sound either. In most states teachers are protected, not just by teacher contracts, but by state laws. Plus, bad teachers can already be dismissed.

The problem isn't with the F-grade teacher, it is with the D+ or C- teacher. Merit pay might punish them by paying them less, but that would probably embitter them rather than encouraging them to leave.

The current push for merit pay is now being led by President Obama. His new budget calls for a dramatic increase in funding for "merit" pay initiatives.

Why? It allows him to position himself as a "reformer" who is willing get out of the box and take on the teachers unions.

The problem now is the belief that merit pay and student achievement, meaning test scores in today's climate, are somehow linked. The push to get the best teachers in the worst schools could well penalize the best teachers who will have more trouble showing higher test scores. It also presupposes that current tests are fine-tuned enough that results are meaningful and that human nature will not create the temptation to cheat among those who are already proven to be less than stellar.

Simply looking to teacher evaluations to set the pay doesn't work either. Not all principals are equal and one may rate one teacher much higher or lower than another. Is this fair and does this really tells us who should be rewarded?

If we really wanted to shake the schoolhouse up, we should try differentiated staffing. We will probably never get 3 million grade-A teachers in the classroom. How might we expose all children to good teachers? First, identify the best through a process of supervisory, peer and parent review. These "master teachers" would be in charge of organizing the learning process and could receive a substantially higher salary. We could then develop a cadre of teachers who are regular teachers, who work under the direction of the master teachers. All students would be exposed to the master teachers and the work would be carried out by the regular teachers.

To aid them, we could then develop a group of "interns" from the ranks of aspiring teachers or teacher helpers. They would support the work of the regular teachers, act as tutors for individual help and do the more menial tasks of the classroom. School systems also could reduce the ratio of adults to students for about the same money now being spent while rewarding those who are the best and creating a career ladder at the same time. This plan would also probably make teachers unions unhappy but it has the merit of actually exposing all children to great teachers.

President Obama has many good ideas and I applaud him for trying to get "out of the box" with his thinking, but I would respectfully suggest that taking on a warmed-over idea of paying teachers for test scores puts him in the wrong box.


Paul Houston is a former superintendent of schools and executive director of the American Association of School Administrators.