Some school policymakers are promoting a new idea for improving the schools: merit pay plans that would tie teachers’ pay to the scores their students earn on standardized math and reading tests.
Advocates of this approach base their support on two assumptions: first, that merit pay is long-established and widespread in the private sector, and second, that students’ test scores are a reliable way to gauge how well teachers are doing their jobs. Both assumptions, according to a new research report issued today by the Economic Policy Institute, are faulty.
In Teachers, Performance Pay, and Accountability: What Education Should Learn from Other Sectors Scott J. Adams, John S. Heywood and Richard Rothstein examine the evidence that underlies these assumptions, concluding that the use of merit pay systems based on quantitative measures is fraught with perverse consequences that often thwart the larger goal of improving the quality of services and outcomes and that such systems are not widespread among private sector professionals.
As Daniel Koretz writes in his preface to this volume, “In large part because available numerical measures are necessarily incomplete, holding workers accountable for them—without countervailing measures of other kinds—often leads to serious distortions.”
In Part One of the study, entitled “Performance Pay in the U.S. Private Sector,” Adams and Heywood offer a detailed description of performance pay systems utilized by businesses and track the trends in their use. They find that, contrary to the claims of advocates of teacher merit pay, “relatively few private sector workers have pay that varies in a direct formulaic way with their productivity, and that the share of such workers is probably declining.” Haywood added, “Formulaic reward structures often reward only a few dimensions of productivity and run the risk of causing workers to abandon effort in the dimensions not rewarded.”
Their research shows that even though many workplaces pay “bonuses,” these are generally not regular performance-related pay of the kind that is being promoted for teachers. And even though the use of bonus pay has grown, that expansion has not been widespread but rather has been focused in certain occupations and industries. The authors describe this growth as “largely a non-union, male phenomenon concentrated among managers and professionals and in finance, insurance, and real estate.” Performance pay now covers only about one in seven workers and represents only a small portion of their compensation.
In Part Two, Richard Rothstein explores “The Perils of Quantitative Performance Accountability” in the field of education, as well as a broad range of other areas extensively studied and documented by social scientists and management theorists. Rothstein’s work shows how even the best-intentioned attempts to create systems for measuring performance often subvert the goals and values of the firm or organization being measured.
Rothstein paints a vivid picture of the perverse consequences created when numbers-based accountability measures encounter the human talent for gaming the system. He draws upon familiar examples such as body counts employed by the military during the Vietnam War, ticket quotas and crime clearance rates used by law enforcement agencies, TV sweeps week, best-seller lists, and college rankings, as well as examining the impact of health care report cards on health care delivery.
Rothstein does not conclude schools and teachers cannot or should not be held accountable; rather, he urges that any accountability system must be built on the extensive experience and research inside and outside of education and on an informed assessment of the gains and losses inherent in any system.
As he writes in his conclusion: “In education, most policy makers who now promote performance incentives and accountability, and scholars who analyze them, seem mostly oblivious to the extensive literature in economics and management theory documenting the inevitable corruption of quantitative indicators and the perverse consequences of performance incentives that rely on such indicators. Of course, ignorant of this literature, many proponents of performance incentives are unable to engage in careful deliberation about whether, in particular cases, the benefits are worth the price.”