A 2009 study in the October issue of the Journal of Labor Economics finds that teacher unions have no effect on teacher pay.

Shocking! As the author, Michael Lovenheim of Stanford University, notes, these results aren’t expected. Unions negotiate for better work conditions and salary increases, so it’s a general assumption that school district unionization would increase teacher pay. Previous studies have shown unionization to increase teacher pay by as much as 12 percent. A paper by Caroline Hoxby, seen here at the Show-Me Institute talking about the effects of collective bargaining, found that majority teacher union membership results in a 5-percent pay boost for teachers (study available here).

Looking at the source of Lovenheim’s data, I wonder whether his findings are attributable to a lack of available details. He calculated monthly teacher pay by dividing a district’s monthly full-time instructional payroll by the number of full-time instructional staff. That pool includes not only teachers, he writes in the study’s appendix, but also educational support staff, school-level administrators, principals, and guidance counselors.

The things I saw while attending teacher-administration contract negotiations at the Lindbergh School District in Saint Louis lead me to believe that educational support staff, school-level administrators, principals, and guidance counselors don’t necessarily benefit from teacher union negotiations, even if the union purports to represent them. In fact, in the Lindbergh case, they lost out slightly. (Administrators and the teacher union ended up agreeing that counselors and librarians should contribute to the “substitute pool” when absent from school, even though substitutes aren’t hired when either counselors or librarians are absent.)

The average pay Lovenheim calculated could be obscuring teacher gains because of the varying salary trends among non-teacher staff. For an example of this in action, see the Miami-Dade School District. After 16 months of negotiations, the district awarded its teachers an average 1.8-percent raise, while “reducing the number of days support staff will work from 260 to 250.” For the unindoctrinated, school districts generally can’t cut pay; instead, they cut days.

Another issue with analyzing the effects of unionization on teacher compensation is that non-salary benefits, such as health, life, and vision insurance, as well as working conditions, are difficult information to find, and can be very hard to quantify. The lack of availability, Lovenheim writes, forced him to leave those measures of compensation out of his study. Non-salary benefits are not inconsequential, and can add up to thousands of dollars more in compensation. If unionization helps teachers negotiate for more non-salary benefits, that could amount to a large positive effect left unaccounted.

I’m not quite sure whether the following is a non-salary or a salary benefit. In any case, it’s my favorite teacher union contract provision. From the Mehlville School District’s 2009–2010 agreement with its teacher union:

Any savings exceeding $70,000 between now and August 1 resulting from teacher staff changes will be added to the base salary.

That surely is evidence of teacher union negotiating power, especially during a recession when school districts across the state are looking at ways to make cuts.