If performance-linked pay models are not tweaked, teachers will focus only on the brighter students.
The Indian economy grew at a better-than-expected rate of 6.7 per cent in 2008-09, partly due to the implementation of the Sixth Pay Commission’s recommendations. The payout in the form of increased salaries and pensions to central government employees contributed around 0.4 per cent to GDP growth. The larger question which, however, remains unanswered, is whether a higher compensation would provide the incentive for attracting and retaining good teachers in our universities and schools? Promoting good teachers is vital because it plays a critical role in the process of human capital formation, which in turn is the key to long term growth and prosperity.
India has a considerable shortages of skilled labour in a number of fields. Although we have managed to rapidly increase access to education, the quality of it remains low, even at the school-level. For example, a 2007 nationwide survey carried out by NGO Pratham found that though 97 percent of children of school-going age are in schools, only 33 percent of third-graders could manage basic arithmetic. The main policy challenge facing India is how to improve the quality of education.
An important determinant of the quality of education is the quality of teachers, especially their motivation. Increasing motivation of teachers and consequently enhancing their capabilities, therefore, are central to any systematic attempt to improve learning outcomes. Now, with a significant increase in salaries, would teaching once again be considered an attractive enough career option? There is little doubt that raising the average pay of teachers is a good first step. It is equally important to put in place however, an effective performance-linked-pay system that would ensure building in accountability and reward effectiveness, and not simply seniority. But how does one go about designing such a system, especially at the school-level?
One possibility is to reward teachers for improving students’ performance. The most common elements in performance-based compensation schemes for teachers that are in practice across the world include measuring of student achievement, measuring of parent satisfaction, and teacher evaluations. Among these, the measuring of student achievement is the most important. However, schemes that equate better examination scores of students with high teacher performance can lead teachers to focus only on students who are likely to improve test scores significantly with some extra effort by teachers. As a result, academically weak students would be left behind even more and the teachers who are assigned to these students would have little motivation to perform.
In a recent paper Barlevy and Neal* have developed a performance-pay scheme that is fairer and more transparent than those used currently in the developed world. The authors have come up with a merit-pay system, supported by vigorous mathematical proofs, that compares the progress of students with the same level of baseline achievement. That is, students are divided into relevant peer groups, based on their initial test scores. Teachers would be paid a relative performance pay when a student achieves a higher percentile score within his peer group. This way, no teacher is placed at a systematic advantage or disadvantage because of the type of students s/he teaches. Further, there is little incentive for teachers to skew their efforts towards a specific set of students. Such a system would also greatly improve the degree to which teachers are made accountable to their clients (children and parents) and their managers (head teachers), just as in the corporate world.
Such a policy would no doubt stir up the debate about segregating students on the basis of their performance. Of course, this segregation need not be made explicit to students or parents. And even if this is done, would that really be a bad thing? Research suggests that students benefit from being taught in more homogenous peer groups. Greater homogeneity allows teachers to tailor their teaching to students’ needs.
In sum, it is a tough job to come up with an objective and a fair measure of evaluating teachers, but it is possible. Performance-linked rewards alone will not be sufficient to improve the quality of school education in the absence of good physical infrastructure, a lack of social recognition for the profession and flawed government policies in the education sector. They would, nonetheless, be a crucial step towards accomplishing that goal.
The author is Senior Economist, CRISIL
*Barlevy, G and Neal, D. ‘Pay for Percentile’ March 2009,
http://economics.uchicago.edu/pdf/neal_042009.pdf
