Much ink has been spilled in the past week over what the pay for performance experiment in New York City's public school system means. Roland Fryer's finding that the NYC pay scheme didn't improve student achievement does not imply that differentiated pay for teachers doesn't work, however. In fact, I'm inclined to borrow a phrase from Chesterton: merit pay has not been tried and found wanting; it has been found difficult and not tried at all.
Merit pay trials in the US have mostly followed a familiar pattern: they're structured as one-time bonuses and are tied to some kind of objective measure like test scores or teacher value-added. This may look superficially similar to professionals' compensation on Wall Street and in the nation's top law firms, but crucial components are missing that make up true merit pay in the professional working world.
Permanent raises based on merit provide a more meaningful incentive than annual bonuses, though the latter are a helpful supplement. Performance-based raises tell professionals that they're hitting milestones on the way to full professional effectiveness (or not), and they communicate the worker's long-term value to an organization. One-time bonuses that sit on top of a never-changing salary schedule that undervalues newer employees' growth are bound to be ineffective by comparison.
Management needs discretion to use both objective and subjective measures of effectiveness to evaluate merit. This is a scary idea for many teachers, but it's a pro-teacher idea at its core. Every teacher benefits from working with colleagues who are team players and build up others by their efforts. Yet these team contributions are often easy to see and difficult to quantify. Managers have to be free to reward this behavior and shouldn't be bound exclusively to the kinds of objective measures that can pass academic peer review.
Finally, principals and superintendents have to be free to employ the ultimate merit pay tool in some cases: zero pay for zero performance (that is, the power to hire and fire). This is a pro-teacher idea as well, but what's good for the profession is not necessarily good for every teacher. Good work is not being rewarded if there is no absolute floor for performance.
I agree in principle with most of Steve Peha's points about the NYC study and pay for performance generally: teachers are not motivated by the same things that motivate investment bankers, teaching is a bad fit for the kinds of standalone bonuses we've used previously in the US, and educators are already working pretty hard. A real merit pay trial would not treat teachers like bankers; rather, it would reward long-term value and empower the kind of leadership that is critical to good instruction.
That's a hard sell to teachers, unions, and other stakeholders ? but it would make teaching more rewarding and more attractive to high achievers.
? Chris Tessone