Last spring, Milwaukee Public Schools committed to reducing class sizes, setting new guidance that schools should include no more than 22 students in kindergarten classrooms, increasing up to 32 by middle school. Earlier, the Milwaukee Journal Sentinel ran its own analysis, finding that the majority of the district’s classrooms were already well within district recommendations. The average kindergarten classroom size was 22, for example.
At the same time, the Wisconsin Department of Public Instruction’s database reports that MPS has only seven students for every staffer. What do we make of this discrepancy? How can we explain the large gaps between class sizes and student-teacher ratios?
There will, of course, be variability due to basic enrollment differences. The Journal Sentinel’s own reporting, for example, found a spread of kindergarten classrooms ranging from as few as nine students and as many as 34. When I was a teacher, my classes varied because of scheduling constraints, students testing into or out of various classes, and enrollment differences across grades.
But how can there be one staff member for every seven students and yet classes as large as 34?
Simply put, there are way more administrators, instructional aides, janitorial workers, and other support staff than teachers. Consider the growth of students, teachers and other staff since 1959, according to data from the National Center for Education Statistics:
| Increase in… | |
| Student enrollment | 47% |
| Teachers | 139% |
| Principals and assistant principals | 210% |
| Instruction coordinators | 631% |
| Guidance counselors | 779% |
| Instructional aides | 1,476% |
In six decades, staff growth has bloated far beyond actual student growth. To put the numbers another way, in 1949, teachers were 70.3 percent of all school staff. In 2022, they were only 47.5 percent. Herein lies the explanation for the discrepancy: Districts hire lots of new people, while classes themselves remain the same size.
Any functional system requires non-instructional staff such as cafeteria workers, janitors, administrators, special education aides, and front office staff. But currently, the system resembles a construction site where one man digs with a shovel and five others mill about, shouting unsolicited advice.
If we line up staffing trends alongside two other trends, salary growth and academic outcomes, the correlations (or lack thereof) introduce compelling questions about the past six decades of policy decisions.
Regarding pay, in the same time that staff rosters ballooned, teacher salaries increased 38 percent in constant dollars (and per pupil expenditure over 300 percent). Instead of investing in teacher salaries, schools expanded personnel rolls. We could have instead kept staffing rosters lower — even increased classroom sizes slightly — and doubled or even tripled teacher pay.
After all, we spend over $1 trillion annually on public education in America, or about $316,000 per teacher. Much of that has to go toward fixed costs such as real estate, building maintenance, and necessary support staff. But there’s plenty of money in the system to improve teacher salaries. We just choose to spend it elsewhere.
Comparing staff growth to long-term educational trends raises even more questions. A comprehensive recent paper comparing four different measures of academic achievement makes the case that since the 1950s, academic scores have increased, but only marginally and largely due to improvements in health and nutrition. All those new positions for what?
In fact, class size reduction policies can actually backfire. For example, in 1996, California introduced a billion-dollar class size reduction initiative. A nice idea, but policymakers cannot Thanos-snap their fingers and improve teacher-to-student ratios. The same number of students show up the next day, and districts have to hire lots of new people to fit the class size caps. California had to hire 25,000 new teachers.
Follow-up studies found that this inflow of new teachers came largely to high-poverty schools, while experienced instructors seized newly created positions in affluent districts. This reality “partially or, in some cases, fully offset the benefits of smaller classes.”
In my time in schools, I’ve observed lots of teachers — some of them truly incredible and others shockingly incompetent. A group of 30 students would learn far more in one class run by a master teacher than were those same 30 students split between two mediocre educators. Class size reduction — hiring more people — is simply the wrong lever to improve outcomes.
To put the problem bluntly, we have too many people working in schools. We have allocated increasing resources toward quantity, not quality, of staff. But that doesn’t mean that a principal, district administrator, or state policymaker should willy nilly halve their teaching workforce tomorrow.
Aaron Smith of the Reason Foundation has proposed a suite of policy ideas to reset schools’ incentives so they make wiser funding decisions. In particular, he recommends reallocating state categorical grants that earmark money for specific non-teacher positions. Putting the money directly into the state funding formula allows principals to hire a librarian or provide a heftier pay raise as their school’s needs demand.
Behind that simple discrepancy of large classes despite small staff-to-student ratios are years of misallocated funds and thousands of individual decisions that led to bloated staffing. Right-sizing public school personnel rosters would not only save the taxpayer money but could increase teacher pay and by extension, perhaps even student outcomes as well.
Daniel Buck is a research fellow and the director of the Conservative Education Reform Network at the American Enterprise Institute. He lives in Wisconsin.
Any use or reproduction of Badger Institute articles or photographs requires prior written permission. To request permission to post articles on a website or print copies for distribution, contact Badger Institute Marketing Director Matt Erdman at matt@badgerinstitute.org.
