Merit-pay plans for teachers may be growing more popular with politicians, but a report released today argues that such compensation plans are rarely used in the private sector and can sometimes bring about unintended negative consequences.
During the 2008 presidential campaign, both President Barack Obama and his rival, U.S. Sen. John McCain, endorsed the idea of performance-incentive plans that would tie teachers’ pay to their students’ scores on standardized tests. Mr. Obama’s proposed fiscal 2010 budget, in fact, calls for boosting spending on the Teacher Incentive Fund, a program that awards grants to school districts to devise performance-pay programs, to $517.3 million, up from $97.3 million in the current year.
But in “Teachers, Performance Pay, and Accountability,” the report published today by the Washington-based Economic Policy Institute, researchers point out that such pay plans are less common in the private sector than their proponents sometimes claim. According to the report, only one in seven workers in the private sector is covered by bonus or merit-pay plans, and most of those workers are in the real estate, finance, and insurance fields.
“There has been modest growth in the U.S. of bonuses,” said John S. Heywood, the University of Wisconsin-Milwaukee economist who co-wrote that part of the study, “but a lot of bonuses may be poorly related to individual performance.”
Workers might get a bonus, for instance, if their entire department or organization has a profitable year or because employers use them as a substitute for paying health benefits, according to Mr. Heywood.
But compensation plans that use formulas or indicators to reward employees on the basis of their productivity—which are among the kinds of programs that growing numbers of policymakers have in mind for teachers—are less common, and may even be declining, according to Mr. Heywood and Scott J. Adams, the other author on that section of the report. Mr. Adams is an associate professor of economics at UW-Milwaukee.
‘More Harm Than Good’
The two researchers based their conclusions on recent data from several large-scale or nationally representative surveys, including the National Compensation Survey, the National Longitudinal Survey of Youth, the Panel Study of Income Dynamics, and the National Study of the Changing Workforce.
The study finds, for example, that in the 2005 National Compensation Survey, only 6 percent of workers received regular, output-based payments in the strictest sense. Even the growing number of bonuses that are not linked to specific measures of productivity account for just 2 percent to 3 percent of overall pay, the report says.
Moreover, the report says, research on the private sectors’ experiences with pay-for-performance schemes suggests that they sometimes yield unproductive results.
“A general lesson from this part of the economy is that when you have jobs where it’s very hard to identify all the dimensions of productivity, and when it’s hard to measure all the individual contributions of productivity, formulaic pay plans tend to be suspect and to do more harm than good,” said Mr. Heywood.
Part of the problem, explains EPI economist Richard Rothstein in the second half of the report, is that pay-for-performance plans based on narrow indicators often lead to unintended, negative consequences—sometimes because workers game the system or sometimes because the measures themselves induce perverse incentives.
Average Performance Improves
For instance, he writes, when the Soviet Union, before its collapse, set targets for the number of shoes its factories should produce, manufacturers responded by making larger numbers of smaller shoes. They may have saved on leather, but the shoes were too small to be of any use to consumers.
Likewise, in the United States, researchers have documented instances in which the advent of health-care “report cards,” which report mortality rates on a hospital-by-hospital basis, led some providers to decline to treat more difficult, severely ill patients. Mr. Rothstein says similar consequences have resulted when police departments set ticket quotas or when television stations conduct promotions that artificially boost their ratings during “sweeps weeks,” so that they could set their advertising rates higher.
“And, of course, we saw what happened with the stock market collapse,” Mr. Rothstein added.
Yet, by the same token, Mr. Rothstein also concludes, studies show that accountability systems based on narrow performance measures can nonetheless improve average overall performance, even as they result in some perverse consequences.
In education, though, Mr. Rothstein maintains “most policymakers 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.”
While he did not have time to read the report or make a judgment on it, Michael J. Podgursky, an economics professor from the University of Missouri, Columbia, who has studied teacher pay-for-performance plans, noted that lessons for education from such efforts in the private sector may be limited, because most such workers do not have tenure, as teachers do. Private-sector employers, particularly those whose workers are not members of unions, may not need to spur motivation when employees know they can be fired at will.
The EPI report is the first of three on teacher merit-pay programs that are intended to add context to current debates over such pay-for-performance plans.
A version of this article appeared in the May 20, 2009 edition of Education Week