A school system run by Comptroller William Thompson would continue experimenting with teacher “merit pay,” he said yesterday in an exclusive interview with GothamSchools. But he said he wouldn’t expect such an experiment to yield much in the way of results.

His mixed message underscores the odd reality of performance pay plans. Though the plans enjoy increasing political support, no research studies have conclusively shown they improve student achievement.

“Would I continue merit pay? Yes,” Thompson said. “Should it make the difference? Hopefully not.”

Thompson told me he has supported merit pay as far back as his days as president of the Board of Education in the late 1990s. But he said giving some teachers more money for boosting student performance probably wouldn’t drive widespread improvement in the city schools.

He described a conversation with a retired teacher who argued that the possibility of an extra few thousand dollars on top of his $90,000 salary would not cause him to change his teaching strategies.

“If it creates a little extra incentive and a little extra recognition, that fine,” Thompson said. “But I wouldn’t see that as the way to be able to talk about better performance.”

I asked Thompson if it makes sense for the city to spend money on differentiating teacher pay if doing so shouldn’t be expected to improve student performance.

“There are a lot of those people who think that it may make a difference and number of those who think it won’t,” he said. “I think that what we’d like to see is to wait to see the outcomes and has it made a difference. I don’t think anybody has had enough time to be able to come to that conclusion.”

In fact, New York City does not have a true merit pay program right now. Such a program would provide financial incentives for individual teachers to do better. Instead, some schools are eligible for bonuses if their students’ performance on state tests improves. A team of teachers and administrators at participating schools then metes out the bonus money across the school’s teaching staff. After the program’s first year, there was no clear evidence that it boosted test scores.

That model is similar to the one introduced in the late 1990s by the Thompson-led Board of Education. That program, called Breakthrough for Learning, promised small bonuses to all teachers at schools in two districts if scores improved. At the time, Thompson called Breakthrough for Learning “an experiment worth having.”