American education reform in the 2010s centered largely on changing how teachers are evaluated and paid. Through Race to the Top and its state waiver program, the Obama administration successfully prodded 44 states to adopt new evaluation systems based, in part, on objective measures of student achievement. These states committed, at least on paper, to using teachers’ evaluation ratings for personnel decisions ranging from who receives tenure to who gets a bonus. In the meantime, a turbo-charged federal Teacher Incentive Fund program encouraged school districts to link educators’ compensation to their performance.
It is tempting to look back at that era and conclude that teacher-pay reform has failed—that we should move on to other strategies. Scores on the National Assessment of Educational Progress stagnated over the course of the decade, with gaps increasing between higher- and lower-performing students. A 2023 National Bureau of Economic Research working paper on the impact of state teacher-evaluation policies reports “precisely estimated null effects.” Policymakers, it seems, had already made the same estimation, for few are paying attention now to evaluation-and-pay reform.
Yet closer inspection uncovers a different story. Despite incorporating test-score evidence, the new state evaluation systems still failed at their most basic task of distinguishing the most- and least-effective educators. As Matt Kraft and Allison Gilmour report, the share of teachers rated ineffective in most settings barely budged—perhaps because the principals doing the rating knew they couldn’t fire low performers or even differentiate pay. Genuinely new approaches to evaluating teachers haven’t failed; they haven’t been widely tried.
Moreover, a growing body of evidence suggests that teacher evaluation-and-pay reform, when it is taken seriously and implemented well, produces gains. Education Next has previously reported on the consequences of the IMPACT evaluation-and-pay system implemented in Washington, D.C. under Michelle Rhee and her successor, Kaya Henderson (see “A Lasting Impact,” research, Fall 2017). In short, strong teachers improved their performance, ineffective teachers left the district, and student performance rose.
In this issue, Yale economist Barbara Biasi provides complementary evidence on the potential of performance-based pay based on Act 10, a 2011 Wisconsin law that limited the scope of collective bargaining to base pay (see “Wisconsin’s Act 10, Flexible Pay, and the Impact on Teacher Labor Markets,” features). As Biasi notes, this “allowed school districts to set pay more flexibly and without unions’ consent, in principle detaching compensation from seniority and credentials.” Act 10 also capped annual growth in base pay at the rate of inflation and required educators to pay more toward health care and pension costs. If you think that teachers should be paid both more and differently than they are now, Act 10 is not for you. But the law did give Wisconsin school districts unprecedented flexibility in setting teachers’ pay.
Not all districts took advantage. About half continued to use traditional step-and-lane salary schedules based on experience and graduate degrees. The other half, however, abandoned step-and-lane schedules and, in effect, allowed individual teachers to negotiate their pay. This natural experiment unfolded gradually across the state, due to differences in when pre-Act 10 collective-bargaining agreements expired, enabling Biasi to study the law’s effects.
She reports that, in districts adopting flexible-pay systems, teachers who were more effective in raising students’ test scores started to earn more than their peers—despite the fact that Wisconsin school districts at the time did not calculate value-added scores. (Apparently, administrators don’t need an algorithmic statewide teacher-evaluation system to identify their best performers.) These districts saw more weak teachers depart and experienced an influx of effective teachers, many of them poached from districts that stuck with seniority-based pay. Incumbent teachers in flexible-pay districts likewise improved their performance, and student achievement rose.
Act 10 did have unintended consequences. Districts serving poor students were less likely to adopt flexible pay systems. As a result, the personnel churn the law generated likely reduced these students’ access to effective teachers. A gender pay-gap emerged, as women proved less likely than male teachers to negotiate with male principals for higher salaries. The cap on growth in base pay may have kept districts from paying Wisconsin teachers more at a time when that would have been helpful.
Still, Biasi’s careful and creative research adds to the evidence that altering how teachers are evaluated and paid remains a powerful lever for improving student outcomes. It suggests that the Obama administration’s teacher-evaluation reform fell short at least in part because it wasn’t accompanied by a loosening of collective-bargaining restrictions. Act 10 reveals the value of first giving districts the flexibility needed to use what they already know about who their strongest performers are. States seeking to draw the right lessons from the past decade’s disappointments would do well to keep that in mind.
— Martin R. West