OUR WORK - The Collective Bargaining Agreement (CBA) Project

The CBA project is a multi-state, multi-study collaboration between researchers at the American Institutes for Research, Michigan State University, University of Washington and University of Nevada – Las Vegas that is funded by the Laura and John Arnold Foundation along with other funders.  This project aimed to better understand the content of CBAs in multiple states (California, Michigan and Washington), the degree to which these CBAs may restrict district administrators as they manage their workforces, and how CBAs have changed over time. In particular, we asked how recent changes to state laws implemented in Michigan and Washington that were intended to constrain collective bargaining and the degree to which CBAs might dictate local policy are associated with changes in CBAs.


Much of our work has focused on the ways CBAs and specific provisions within agreements affect schools, teachers, and students. In general, the research suggests:

  • There is considerable variation in the extent to which collective bargaining agreements (CBAs) in different districts “restrict” district management. For example, the following maps from California, Michigan, and Washington show the variation in contract restrictiveness (a measure derived from the individual provisions contained in CBAs) in the CBAs operating in different districts in these states, and this restrictiveness varies over time:

  • On average, CBAs do not appear to change dramatically in the degree to which they restrict administrators over the reform period.

  • However, just as there is substantial variation in the restrictiveness of CBAs within each state, there is also substantial variation in changes in the restrictiveness of CBAs over the time period in which many teacher policy reforms were enacted. This is shown in the map on the right from above.

  • Districts with larger enrollments, in urban locations, and with higher proportions of student in poverty have more restrictive contracts.

  • Larger discrepancies exist between urban and non-urban district contract restrictiveness in Michigan and Washington than in California.

  • Notwithstanding differences in overall contract restrictiveness, urban contracts look similar to their non-urban counterparts on certain "high profile" provisions. (for more on this see HIGH PROFILE PROVISIONS).

  • Few observable characteristics (i.e., district average performance, size, demographic makeup) predict changes in CBA restrictiveness.

  • One factor that does seem important in predicting the content and restrictiveness of CBAs is geography and organizational structure. In particular, collective bargaining outcomes appear to “spill over” into neighboring districts, particularly if union bargaining in those districts is supported by the same “uniserv council”.

  • Teachers are relatively responsive to the provisions in CBAs; for example, in Washington, some inequitable patterns in teacher transfers across different types of schools within districts can be explained by the provisions that regulate teacher transfers in those districts’ CBAs.

  • In California, contract restrictiveness is associated with average district performance on standardized tests; districts with more restrictive contracts fare worse on performance assessments.

  • In longitudinal analyses, however, it does not appear that changes in CBA restrictiveness cause decreases in student performance. The relationship between changes in CBA restrictiveness and student outcomes is generally negative, though not statistically significant.

Copyright © 2018 All rights reserved