Preservation of management rights over class size requires diligence. More than a third of collective bargaining agreements in Michigan in 1998 established a maximum number of students for each class and provided for mandatory teacher salary bonuses any time this maximum was exceeded. Some contracts mandated that teachers be paid an additional $1 to $4 per day for each student over the maximum. Other contracts specified a $75 bonus per additional student per semester.

Negotiating smaller class sizes has proven to be a costly arrangement for school districts, especially those with growing student populations. Smaller classes mean that more teachers must be hired and put on the payroll, which increases education costs. Charles Rehmus and Evan Wilner concluded in "The Economic Results of Teacher Bargaining: Michigan’s First Two Years":

"Most teacher bargaining requests have included proposed limitations on class size. While school administrators and most school board members are sympathetic with the teacher preference for smaller classes, class size limitations have severe cost impact. A simple example makes the point. Reduction of average class size from 30 to a negotiated maximum of 25 students in a class would result in a [20]-percent increase in teacher salary costs."[259]

Establishing class-size requirements within a collective bargaining agreement restricts the school administration’s decision making about the most effective use of staff, space and scarce financial resources. There is also little evidence that supports the main justification for these proposals — namely, that smaller classes produce improvements in student performance. Education reformer Chester Finn explains the cycle:

"Parents take for granted that smaller classes mean better education. Teachers cheer any move to shrink their classroom populations. Unions get more members. Administrators get more staff, ... [yet] there’s no credible evidence that across-the-board reductions in class size boost pupil achievement."[260]

Finn goes on to cite University of Rochester economist Eric Hanushek’s study of the relationship between class size and student performance. Hanushek reportedly found that between 1950 and 1994 the student-to-teacher ratio dropped by 35 percent, from an average of 30 students per class to the current average of 22. At the same time, spending has increased to its highest level and student performance on standardized tests has not improved.[261] Hanushek concluded, "[T]here is little systematic gain from general reduction in class size."[262]

Jay Greene, chair of the Department of Education Reform at the University of Arkansas, found the relationship between class size reductions and increases in student performance to be a "myth."[263] While Greene notes that it may be possible to improve student performance through the reduction of class size, the evidence is at best mixed, and it is unlikely that such a program could be implemented on a large scale.[264]