How does teaching stack up to other occupations in terms of compensation? A recent analysis from the Economic Policy Institute (EPI), an organization with union ties, has gained attention for its findings on the growing teacher “wage gap.” Using data from the Bureau of Labor Statistics Current Population Survey (BLS-CPS), the EPI analysts report a 17 percent disparity between teachers’ weekly wages relative to other college-educated workers. When they add generous benefits on top—including health care and pensions—that gap shrinks to 11 percent. These differences in wages and total compensation, the authors find, are much wider than what teachers faced in mid-1990s. Based on their analysis, they suggest raising teacher wages and benefits across the board.
Do the EPI authors get it right? There are a few problems with their analysis: They chose a questionable comparison group by looking at other college-educated workers, and they don’t account for summers off. (Also see economist Michael Podgursky’s Flypaper article, which argues that BLS benefits data undervalue teacher pensions, leading EPI to overstate the gap in total compensation.)
Let’s start with the problem of EPI’s comparison group—workers holding a college degree. By using this group as a benchmark, the analysts don’t factor in the different labor market values of the skills and abilities associated with education majors versus other university degrees. As Andrew Biggs points out, “all bachelor’s degrees aren’t the same.” For instance, though they hold equivalent degrees, an engineer may possess skills that the broader labor market values differently than those obtained by most university-educated teachers. Unfortunately, this important nuance isn’t captured in the EPI analysis. Without attempts to control for the various skill levels among college-educated workers, wage comparisons become highly tenuous.
Turning now to the question of summer vacation: Comparisons of relative wages should account for the shorter work year and, consequently, the fewer number of annual work hours for teachers. The EPI analysts, however, evade this issue by using weekly wages. They write, “Attempts to compare the hourly pay of teachers and other professionals have resulted in considerable controversy by setting off an unproductive debate about the number of hours teachers work at home versus other professionals.” As a result, the authors “elect to use weekly wages to avoid measurement issues regarding differences in annual weeks worked (teachers’ traditional ‘summers off’) and the number of hours worked per week.”
Should they have avoided this issue and assumed that teachers work similar hours as their non-teaching counterparts?
Probably not. In an article recently published in Education Finance and Policy, Kristine West takes a closer look at teacher wages by relying on self-reported estimates of hours worked, including those worked over the summer. She uses the same BLS survey data on wages but supplements her analysis with data from the American Time Use Survey. This survey is administered to a subset of the BLS-CPS respondents, asking them to report how they’ve spent their time over the past twenty-four hours. Such “time diaries,” West maintains, “provide a clearer picture of hours of work for teachers and non-teachers than either administrative data or recall data from surveys.”
Averaged over the calendar year, West finds that teachers work five fewer hours per week than non-teachers (39.8 versus 34.5 hours). When basing pay comparisons on hours worked, she finds that elementary and middle school teachers are actually paid slightly higher wages relative to similar workers—evidence that contradicts EPI’s assessment. (That being said, her analysis reveals that high school teachers face a 7–14 percent wage gap.) West’s study does not include benefits—it looked at wages only—so it probably underestimates the relative advantage in total compensation that elementary and middle school teachers enjoy. Of course, teachers aren’t necessarily to blame for working fewer hours; summer vacation is a longstanding tradition, though one that is changing in some places. But West’s study does indicate that ignoring summer in pay comparisons, as EPI did, is a questionable and perhaps political choice.
EPI’s greatest shortcoming isn’t faulty measurement, but rather its policy conclusion that raising compensation for each and every teacher would greatly improve American education.
Merely boosting teacher salaries on the backs of taxpayers is unlikely to advance student outcomes. The reason is that today’s teacher compensation structures don’t link pay to increased achievement. In virtually every U.S. school district, a single-salary schedule based on degrees earned and seniority dictates wages. But research has consistently failed to uncover a relationship between degrees and teachers’ contribution to learning. Meanwhile, the link between experience and achievement largely dissipates after a few years in the classroom. Michael Podgursky explains the problem in the Handbook of the Economics of Education: “The single salary schedule suppresses differences between more effective and less effective teachers.” Given this compensation policy, there is no reason to believe that blanket pay raises will do much good for American students.
To be certain, top-performing teachers deserve greater pay, as do educators in high-need subject areas or those working in tougher conditions. Meanwhile, the compensation of poorly performing teachers is incongruent with their meager contributions to student learning. Instead of inefficient salary schedules, school leaders should have the flexibility to allocate pay strategically—just as managers do in many other parts of our economy. So yes, teacher compensation policies need a fundamental overhaul in the U.S. But not in the way that EPI and the teachers’ unions would have you believe.
