Research Highlights Featured Chart

September 1, 2026

Teacher pay and student achievement

Large stipends drew effective educators into Dallas's lowest-performing schools.

Tyler Smith

Source: nruboc

Despite extensive efforts, education policy in the United States has enjoyed limited success in raising achievement at the lowest-performing schools. Attracting and retaining highly effective teachers poses a particular challenge, but districts have very rarely used financial incentives to elevate the quality of instruction in the lowest achieving and often most difficult to staff schools. 

In a paper in the American Economic Journal: Economic Policy, authors Andrew Morgan, Minh Nguyen, Eric Hanushek, Ben Ost, and Steven Rivkin evaluate a Dallas program that confronted this problem by paying substantial effectiveness-based stipends to attract the most effective educators to its lowest-achieving schools.

The Accelerating Campus Excellence (ACE) program, launched in Dallas in 2016, offered teachers in the poorest performing schools annual stipends ranging from $6,000 to $10,000 based on prior-year performance ratings.  Principals also received stipends of $13,000. The program required all existing teachers at ACE schools to reapply for their jobs, and approximately 80 percent of them—and every principal—were replaced in the first ACE year. Importantly, high performing teachers moved to these challenging schools, and student performance immediately rose.

Figure 4 from the paper illustrates the authors' central findings.

Figure 4 from Morgan et al. (2026)

The chart plots standardized math and reading scores from 2012 to 2019, tracing four groups: the initial rollout of four ACE schools in 2016 (labeled ACE 1), the second wave of five schools added in 2018 (labeled ACE 2), and each wave's control group, defined as the lowest 15 percent of nontreated schools by prior achievement.

Before treatment, ACE schools and their controls follow similar trends, suggesting a like-for-like comparison. In each wave's first treated year, the ACE lines rise sharply and diverge from their controls, with math gains exceeding 0.4 standard deviations, and 0.3 for reading. The improvement brought the previously lowest-performing schools close to the district average.

The figure also shows what happened when the incentives ended. In 2019, Dallas eliminated stipends at three of the four original ACE schools, and the ACE 1 scores fell for both math and reading. This reversal coincided with an exodus of highly rated teachers, and it suggests that sustained gains are dependent on continued high compensation.

The authors estimate that ACE increased average test scores by 0.313 standard deviations per $1,000 per student, making it more cost effective than class-size reduction or high-dosage tutoring under standard assumptions.