REL 2005 Policy Forum
Teacher Compensation Research and Policy Overview

Dr. Anthony Milanowski - Consortium for Policy Research in Education (CPRE)
University of Wisconsin-Madison

Dr. Anthony Milanowski described research on teacher compensation and used examples to illustrate results from state and local education efforts implemented in the Unites States. His message was clear: we must see pay strategically and ensure we think of teacher compensation as a system, i.e., with pay and supports for the whole teacher system. He stressed that how a teacher compensation program is administered is a key factor to its success. Additionally, he emphasized that teacher compensation programs should be aimed at need; therefore, assessments of issues such as teacher recruitment and retention; teacher skill levels; and alignment between school and/or district priorities, goals, and strategies and what is expected of teachers are necessary.

Milanowski commented on teacher union issues that must be considered when thinking about teacher compensation reform, including the use of incentives and/or implementing knowledge and skill-based pay systems. He suggested public education agencies work with unions and teacher associations when constructing compensation policies, but recognizes this is not always possible and covert tactics may be necessary.

He described the use of incentives for teachers as one tactic, but called attention to the need to evaluate the implementation and impact of incentives. One question about incentives that we should ask is, “Does the incentive move people into the field versus keep them in the field?” In regard to incentives, Milanowski described a number of different types and ways states, districts, and schools have implemented them. Many are direct monetary funds such as signing bonuses, additions to base pay, and individual pay awards for performance. Incentives may also include loan forgiveness, housing assistance, and extra retirement credits. Some incentives are based on teacher knowledge and skills, some are tied to recruitment and retention efforts, particularly in high-need schools, i.e., those with geographic, subject area, and/or school demographic challenges. Incentives may be used to improve teacher skills or as a means of motivation. Research provides some guidance on the use and impact of incentives.

Milanowski discussed that incentives for teaching in high-need schools often are in the form of bonuses and have been used in New York, Nevada, and a number of local jurisdictions such as Charlotte-Mecklenburg’s Equity Plus Program in North Carolina. Econometric studies suggest relatively large financial incentives would be needed to influence teacher choice. However, there are no large scale studies of targeted incentives, only anecdotal evidence from some districts that these incentives are effective. For additional research and information see, “Making Schools Work: Improving Performance and Controlling Costs” by Eric Hanushek et al. at
and Charlotte-Mecklenburg program at (link no longer functioning as of 1/2011)

Incentives specific to teachers in schools with shortages in academic subjects, such as math, science and special education, often provide incentives in the form of a salary add-on. This has been instituted in North Carolina and local districts. Some evidence has shown that math and science teachers have better paying alternatives outside of education than do other teachers which speaks to the amount needed for the incentive. For research see, “An Exploration of the Pay Levels Needed to Attract Students with Mathematics, Science and Technology Skills to a Career in K-12 Teaching” by Anthony Milanowski at

Milanowski also described incentives for teachers seeking or having National Board Certification (NBC). Incentives in the form of pay bonuses and application fee and tuition reimbursement have been implemented in many states. Research indicates that incentives raise the rate of NBC participation; however, there is mixed evidence on whether NBC improves teachers’ skills. Milanowski raised the question about NBC, “Are we rewarding teachers who are already good or are we rewarding teachers to improve?” He noted some research that has shown a connection between increased student performance and teachers with NBC. However, he cautions us to ask, “Are we allocating teachers with NBC where they are needed?” For research see, “Can Teacher Quality Be Effectively Assessed?” by Dan Goldhaber and Emily Anthony at and “A Distinction That Matters: Why National Teacher Certification Makes a Difference” by Lloyd Bond et. al at

In relation to professional development incentives for teachers, they are often in the form of subsidized tuition and necessary time off for participation. These type of incentives have been put into practice in Iowa, Delaware, and local jurisdictions. Limited research on the effect of this incentive for teachers. For research see, “Helping Teachers Teach Well: Transforming Professional Development” by Thomas B. Corcoran at

Milanowski also focused on knowledge and skill-based teacher pay systems. These systems include efforts for demonstrating competencies in the classroom. A number of programs that were implemented were terminated; however, there are a currently some who are experiencing positive results, such as in Philadelphia and Kyrene, Arizona. To effectively use a performance pay model, he recommended instituting rigorous assessments of teacher practice measured by multiple classroom observations and lines of evidence, including explicit standards, various practice levels, and behavioral rating scales. He emphasized the importance of training for evaluators. Additionally he noted that hiring evaluators specifically for this purpose is costly but worthwhile. Research conducted by CPRE on the effects of evaluation can be found at In his research, Milanowski also found that teachers ask how they can improve, wanting feedback and coaching. He suggests ensuring knowledge and skill-based pay systems incorporate immediate feedback for teachers on process and “not results only.” Further, he learned that teachers are less enthusiastic about moving up a compensation career ladder because they believe they can’t do it. He proposes we “up-skill our teaching workforce,” using higher level teaching skills with multiple supports to enhance those skills and aligned professional development.

Another form of performance pay are school-based awards, often offered as a bonus to all teachers (and others) in a school when that school achieves pre-established performance goals. Milanowski described research showing that these programs often help focus staff attention on performance goals; however, they are not highly motivational, may increase teacher turnover, and can act as a deterrent for new teachers. Milanowski suggests for a program to be successful we must first ask, “Do you have the school/district/statewide support to do this type of bonus program?” For this research go to,

Milanowski stated that merit pay, i.e., teacher pay increases based on a principal’s subjective evaluation, led to many equity and personnel problems but still exists in some schools today. Current pay for performance models use a value-added approach, i.e., teacher pay based on their individual student’s achievement over time. Research has shown both positive and negative effects of using a merit pay approach and additional research is needed. He noted the following research concerns:

  • It is hard to determine the middle ground teachers (those not the very best or very worst)
  • It is difficult to control for what students go to what teachers
  • Not all teachers teach tested subjects.

In conclusion, Milanowski pointed out that some incentive programs and knowledge and skill-based pay systems can work; however, changing a teacher’s pay is not an end unto itself nor is it a simple solution. We must use changes in compensation to support other reform strategies that impact instruction. He added that distrust of past efforts is so strong that making changes to pay structures are difficult. Further, Milanowski emphasized that administrators must also be included in compensation system reform. He suggests cost-benefit studies be conducted to financially justify the implementation or continuation of compensation programs.

Dr. Anthony Milanowski's Presentation (PPT)