3.6.5 Performance pay

Some countries or education systems link teacher rewards with performance, in the form of additional salary or bonuses (in the past and in some countries, often referred to as ‘merit pay’). Such plans may aim to attract and retain high-quality teachers, as well as to encourage motivation and effort to improve learning outcomes. As the evidence base is inconclusive (OECD, 2012), policy developers will want to carefully consider the arguments for and against performance-related schemes, the experiences of systems where they have been implemented, as well as their objectives and design, before incorporating them in a teacher policy.

The arguments in favour put forward research evidence that performance pay has been effective in raising teacher and student performance. The World Bank (2013: 34–35) cites a number of studies in high- and middle-income countries to support this finding; see also OECD (2012: 1, 4). Tying compensation directly to teacher assiduity and pupil outcomes is perceived as resulting in reduced teacher absenteeism and greater teacher instruction effort, indicators seen as major contributors to the learning progress. Such progress is almost always measured by standardized test results in the advocates’ studies. Other proponents suggest that, when applied within school-based programmes, performance-related incentives reinforce the teamwork ethos, and where based on teacher appraisal, motivation is reinforced by reward, thus combining both professional rewards (satisfaction with students’ progress) and financial ones (more income). These arguments and the evidence they cite strongly associate performance pay incentives with key education objectives.

The arguments against these schemes put forward research indicating there is little evidence of improvements in education or learning from performance-related pay (ILO, 2012: 166–168 cites studies and case studies from a range of countries – mostly high-income; see also OECD, 2012: 1, 4). Opponents argue that there is much evidence of negative impacts, such as reinforcing disparities among schools, thereby making it harder for disadvantaged schools to attract the teachers they need. Teachers working in such plans tend to ‘teach to the test’, to ensure that learners perform well on standardized tests, thus narrowing the skills and knowledge acquired. Performance pay schemes may fail to account for factors over which teachers have no control (such as poverty, language, parental engagement and the impact of previous teachers) that affect learning. Any such plan based on objective-setting and performance management is complex and requires time, skills and training that can overburden school managers. When based only on rewarding individuals, performance pay undermines teamwork as a factor in learning progress and good schools, as well as individual teacher motivation when a plan is perceived as unfair. These arguments are based on evidence that performance pay incentives fundamentally work against key education objectives.

When performance-related incentives are used more frequently, the evidence gathered in recent years suggests that the way in which they are applied makes a significant difference in achieving education goals. Key factors to consider are the methods of teacher appraisal as the basis for rewards; the size of the incentives and their financial sustainability over time; the close connection between expected behaviour and rewards; and the level of awards, either for the individual teacher, group of teachers or the school as a whole (see Harris, 2007; Ingvarson et al., 2007; World Bank, 2013, citing a range of studies). The overall salary system also plays a role. As an example, the high- and middle-income OECD member countries use a variety of reward plans for outstanding teacher performance, including positioning on the base salary scale and supplemental and incremental (step or grade) payments. Student performance is generally better when performance pay systems are in place in countries with comparatively low teacher salaries (less than 15% above GDP per capita) and lower in countries with relatively well-paid teachers (more than 15% above GDP per capita) (OECD, 2012: 2–3).

Human resource management policy suggests that a performance pay plan in teacher policy needs to answer the following questions:

• Whose performance is being assessed: individual teachers, a team or group of teachers, or all of the school’s employees?
• How performance is to be measured and evaluated: by outputs (the achievement of individual or group learning targets, for example) or inputs (teacher skills, knowledge and behaviour)?
• How performance is to be rewarded: assessment of individual or group performance, by whose judgement and on what criteria? (Kessler, 2005)

The administrative principles to follow to ensure successful performance pay include equity and transparency in application, diverse and relevant criteria, good communication, wide teacher understanding of the system and professional support to teachers whose performance does not meet the standards, up to the point where such support is no longer helpful (see also Section 3.9).

In sum, where the teaching and learning needs, management capacity and available resources suggest that performance pay could be useful and feasible, policy developers may wish to consider the following issues (ILO, 2012: 166, citing several pay specialists):

• Performance indicators: are they limited to very narrow, measurable indicators of student achievement by means of standardized tests in core subjects, or do they encompass a broader array of learning objectives, such as creativity or the capacity to reason or solve problems?
• Measuring progress: will success be determined by standardized tests or a more diversified array of measurements, such as student learning profiles and teacher evaluation results (peer, school supervisor or external assessors)? Is measurement based on progress from year to year in comparison with a desired benchmark or on a value-added definition? What weight is to be given to the contributions of teaching support staff in achieving learning improvements? Is the data used sufficient and reliable, particularly over time, as a basis for reward decisions?
• Adjusting for external factors: have the measurement instruments taken into consideration factors outside the school, such as poverty and disadvantaged learners, parent roles and differences in funding between schools or school systems?
• Eligibility and funding: will all teachers/staff be eligible? Will the plan focus on individual performance, whole-school performance or both? How large will the reward be in relation to other compensation? Are funds for the rewards sustainable over time?

A final crucial consideration for all teacher policy work is to enhance acceptance and cooperation and ensure the success of any performance pay plan by involving the teachers and their representative unions/associations in its design and application through social dialogue (ILO/ UNESCO, 1966: Art. 124). Many, if not most, schemes fail if they choose to ignore this principle.