Despite working longer hours and experiencing higher levels of stress than professionals in other industries, teachers earn significantly less than similarly educated workers—a gap that’s grown even wider in recent years and that contributes to the high attrition rates that have plagued the profession for decades. Even more concerning are the effects of such attrition on students. Research has shown that, when schools experience high teacher turnover rates, it’s not just students with new teachers who experience the impacts; all students suffer academically. A new working paper by researcher Prasiddha Shakya of Colorado State University explores one possible approach to rectifying this problem by examining the impact of New Jersey’s $50,000 salary floor on student achievement, graduation rates, and college enrollment rates.
New Jersey’s “50K the First Day” campaign, a bargaining negotiation led by the New Jersey Education Association (NJEA), set a minimum salary of $50,000 for new teachers at the first step of the salary schedule. Most Garden State districts adopted this salary floor between 2010 and 2015. To determine its effects, analyst Prasiddha Shakya collected data from 2003–2019 from three main sources: the New Jersey Department of Education (NJDOE) website (for enrollment, demographics, academic outcomes, and expenditures), a formal NJDOE data request (for teacher names, job categories, experience levels, and educational attainment), and NJEA (for information on when each district adopted the $50,000 salary floor).
He then used a staggered differences-in-differences design that compares outcomes in “treated” districts that had adopted the minimum salary (the treatment group) to “not-yet-treated" districts (the control group) during any given year in the study period. A comparison of early adopters (pre-2010), intermediate adopters (2011–2015), and late adopters (2016 and later) showed that schools falling in these categories were similar enough to make the treatment and control groups comparable. (“Never-treated” districts were found to be different and were excluded.)
Two models were constructed to examine different aspects of the salary campaign. The first estimated the impact of implementing the minimum salary on teacher salaries, controlling for teacher experience level. The second model estimated the effects of salary increases on district and school-level outcomes, including spending, school composition, and student outcomes (such as, the percentages of fourth graders, eighth graders, and high schoolers proficient in English language arts and math, graduation rates, and college enrollment rates).
The results of the analysis showed that the campaign increased the salaries of teachers of all experience levels by an average of $1,500 without significantly increasing total per-pupil spending in treated districts; rather than re-allocating funds from other components of education, “treated” districts hired fewer new teachers. These findings confirm that the effects of the campaign on student achievement were not confounded by differential impacts on teachers by experience level or the re-allocation of funds from other aspects of education. In addition, as a result of hiring fewer teachers, “treated schools” saw a small increase in student-teacher ratios (of less than one additional student per teacher); however, if this increase were to affect student outcomes in the study, Shakya argues it likely would introduce a negative bias, making any positive effects even more notable.
The analysis of student outcomes revealed modest and sustained increases in both math and ELA proficiency rates for fourth graders (by 0.055 SD and 0.045 SD, respectively) and high schoolers (by 0.06 SD in both subjects). Shakya did not, however, find any significant impacts for eighth graders. The minimum salary campaign also had a positive impact on high school graduation rates (which increased by 0.06 SD, on average). Effects on college enrollment rates were also slightly positive, but much smaller.
There are some limitations to these findings. Most notably, the teacher labor market in New Jersey does not necessarily reflect labor markets in other states. For example, New Jersey teacher salaries were already, on average, significantly higher than the national average when the reform was implemented. And while the author was able to rule out some potential causal mechanisms, there is not enough information to identify any specific mechanism by which the reform impacted student achievement.
Nonetheless, the study may have some important implications for teacher pay policy in the United States. First, the analysis of expenditures indicated that schools did not re-allocate funds from other components of education, but rather slowed their hiring post-policy, to account for the cost of increased salaries, the study shows that such salary increases need not come at the cost of other essential aspects of classroom instruction. More importantly, the study reveals that—so long as they are implemented across all teacher experience levels—even modest salary increases of $1,500 can positively impact student performance, indicating that campaigns like “$50K the First Day” that set a statewide salary floor could be a viable approach for policymakers in addressing the negative impacts that low teacher salaries can have on student achievement in the United States.
SOURCE: Prasiddha Shakya, “What Happens When We Pay Our Teachers More? Evidence from New Jersey Public Schools,” Annenberg Institute at Brown University (December 2024).
