Sherman has a second, and even weaker, argument for opposing student pay for performance:
[Another problem] is that these programs are not finely calibrated. Whether they reward status achievement (straight As or a certain score on standardized tests) or some sort of growth/effort, there are going to be some rewarded students who did not work hard for the reward and other unrewarded students who probably deserve it.
It appears that Sherman is purposely ignoring the fact that this goes on every single day in a heterogeneously grouped classroom, i.e., the typical American classroom. The instruction in these classrooms are not finely calibrated for low-performers (or high performers). Low performers receive a daily stream of negative feedback due to their low performance relative to the higher-performing students, especially the ones who perform well with seemingly little effort while the low-performing student struggles mightily and and achieves little. Given these conditions it is unlikely that the low-performing student is going to be too upset with yet an other program which shows he is incompetent.
I'll take this argument more seriously when Sherman uses the same reasoning to denounce heterogeneously grouped classrooms.
Sherman also questions rewarding students due to status because some rewarded students may not have "work[ed] hard for the reward." The answer to this questions depends on the preexisting motivation state of the students. If these students were already motivated to do the work an external motivation system is superfluous and should not have been implemented. There is, however, no problem with providing incentives to unmotivated students who are capable of performing the work:
If students are able to perform well on an activity but are not motivated they are perfectly logical and their lack of motivation is based on evidence. They have evidence that their work doesn't make a difference. If they work hard to finish a task, they discover that the teacher was not impressed and acted as if this effort was not adequate. The evidence these students need is that the material is important; the teacher responds to it as if it is important; and when students do it well, the teacher is very impressed with their importance. (emphasis original)
Fixing Motivation Problems, Engelman and Crawford, DI News Fall 2007.
Sherman also questions the failure to reward students with growth/effort because the "unrewarded students ... probably deserve it." The answer to this question depends upon whether the student was actually capable of reaching the goal in the first place. If the student was capable of achieving the goal, yet failed to achieve, this subpar behavior should not be rewarded because rewarding this behavior will merely reinforce the student's not working hard which is something you definitely don't want to do. If, however, the student isn't capable of reaching the goal in the first place, then we are in the realm of shaping behavior which brings us into the realm of differential reinforcement and shifting criterion of enforcement and is outside the realm of the kinds of motivation programs under discussion.
I think it's time that Sherman graduated from the potty training manual to a real primer on operant psychology for the classroom. Let me suggest Applied Psychology for Teachers: A Behavioral Cognitive Approach.
Update: Daniel Willingham has a good discussion of the behavioral research behind these pay for performance programs.