AEI - Short Publications - Teacher Labor Agreements

Some four decades after the advent of collective bargaining in public education, the labor agreements it has produced--and their operational equivalents in non-collective-bargaining states--now touch virtually all aspects of school district operations, from how teachers are paid and assigned to how they can be evaluated, how or whether they can be disciplined or fired, when and where they complete professional development, and how much time off they are allowed for union activities.

Teachers unions, like all unions, fight for provisions that are favorable to union members--not necessarily to students. They consistently favor contract provisions that protect jobs, restrict demands placed on employees, and limit teacher accountability for student performance. But unlike, say, unions that protect auto workers or pilots, teachers unions are in the unique position of being able to help choose the management (e.g., the school boards or mayors) with whom they negotiate due to their influence in local elections. Teachers unions are also active in state politics, where they are frequently able to enshrine favorable provisions in state law (which, from a union perspective, is preferable to having to win the same concession in district after district).

The agreements that unions secure at the bargaining table typically include provisions that are at cross-purposes with the tenets of sound management and governance. But while this trend has been documented in recent research, little attention has been given to examining these constraints systematically or to measuring variability among different communities. Are there districts where principals are particularly constrained and others where they have greater leeway to lead? At least in some communities, labor agreements may not bear as much responsibility for enervated management as the more ardent union critics have suggested.

Districts serving the largest concentrations of poor and minority children are also those in which labor agreements are less flexible and more likely to hinder effective school leadership.

In The Leadership Limbo,[1] we set out to examine how much flexibility school leaders enjoy on key dimensions of management in America's fifty largest school districts. We culled data from the National Council on Teacher Quality's collective bargaining database and sorted them into twelve key components within three categories: teacher compensation, personnel policies, and work rules. We then graded each district on a scale of A to F based on the degree of flexibility it granted school leaders, allowing readers to identify and compare the most and least management-friendly contract language on a variety of important issues. Each district also received an overall grade point average (GPA) and rating, from "Highly Restrictive" to "Highly Flexible," allowing us to determine whether the actions (or inaction, as the case may be) of the district and school leaders can fairly be attributed to its labor agreement.

Compensation, personnel policies, and work rules are, of course, influenced by many factors other than local labor agreements. State laws and regulations must be taken into consideration, so we sought to isolate the degree of flexibility a labor agreement affords from the constraints imposed by state policy. It is also helpful to recall that many districts must contend with multiple contracts for various groups of employees, from janitorial staff to bus drivers to principals. Additionally, labor agreements are not self-implementing or automatic; district practice will reflect but not mirror the contract language. For all these reasons, the analysis presented here is best understood as revealing the outlines of formal district policy and not comprehensively portraying district practices.

We began with a simple question: how frequently do the labor agreements negotiated by school districts and teachers unions contain provisions that make it harder for public schools to be smart, flexible, high-performing organizations? There is no doubt that today's superintendents and principals bear greater responsibility for student achievement and school effectiveness than did their predecessors a generation ago; the aim of our study was to determine whether these new obligations have been accompanied by the authority necessary to lead successfully in a results-conscious era.

Our analysis unapologetically proceeds from the premise that such flexibility is desirable, indeed is a hallmark of sensible management practice in both the private and public sectors. More specifically, the following assumptions guided the report: 1) Individual schools should be organized as accountable, flexible educational organizations that empower school leaders with the requisite authority and tools; 2) Teacher compensation should acknowledge and promote professionalism by reflecting the scarcity and value of teachers' skills, the difficulty of their assignments, the extent of their responsibilities, and the caliber of their work; and 3) Administrators should be readily able to identify and support or terminate ineffective educators as appropriate. To be sure, there is reasonable concern as to whether today's principals possess the preparation and skills equal to the burdens implied by greater managerial flexibility. Promoting agile management, however, is crucial precisely because today's schools tend to reward cautious, passive management while repelling proactive problem-solvers.

The problem, then, is not with the existence of either unions or labor agreements per se, but with provisions that make it harder to operate good schools. It is important to note that every provision in today's labor agreements has been agreed to or adopted by school boards and district leaders; provisions are not unilaterally imposed by willful unions. So both sides are either jointly culpable or entitled to a share of the credit.

We do not presume that there is one "ideal" labor agreement. Various school districts, from Denver to New York City, have recently made important modifications to their longstanding labor policies. More significantly, Green Dot Public Schools have pioneered a "thin contract" that offers key worker protections while taking care not to restrict principal authority or school autonomy. In considering reforms to labor agreements, it is important to remember that there are several effective ways to establish flexibility for school leaders.

Although teacher labor agreements influence virtually all aspects of district operations, we focus here on three especially significant categories in which certain provisions tend to hinder management: compensation, personnel policies, and work rules.[2]

Compensation. As a general rule, one useful way to attract, retain, motivate, and manage talented people is to reward them monetarily for their efforts. But pay differentials in public education, like most unionized industries, are based mostly on seniority and formal credentials rather than performance.

Unfortunately, this approach has become untenable, as the teaching profession has changed dramatically over the last few decades. The likelihood that a female ranked in the top 10 percent of her high school cohort would become a teacher fell 50 percent between 1964 and 2000.[3] In areas like math, science, computer science, and special education, there are often too few qualified teachers to meet the demand, especially in urban schools.

Labor agreements that formalize inflexibility make it more difficult for principals and district leaders to attract skilled and experienced candidates or reward teachers for raising student achievement or taking challenging classroom or school assignments. Recent data suggest that just 5.5 percent of traditional public school districts use incentives (such as cash bonuses, salary increases, or extra salary steps) to foster excellent teaching.[4]

Personnel Policies. When it comes to assigning teachers to schools or within schools, much less removing weak educators from dysfunctional classrooms, managers find themselves hobbled by extensive labor rules. Language may flatly prohibit school leaders from making sensible decisions, but extensive grievance procedures and the compliance time they require may also deter forceful leadership and add up to management by paralysis.

In practice, there is a tendency to give senior teachers plum assignments--both the schools in which they work and their classes and duties within that school--with scant regard for their skills, student needs, or the implications for recruiting and retaining new teachers.

Tellingly, many teachers recognize how difficult it can be to purge ineffective practitioners, with 36 percent reporting that "between tenure and the documentation requirements, it's too hard for administrators to remove any but the very worst teachers," and just 14 percent stating that management's inability to remove bad teachers is not a problem.[5] Meanwhile, 32 percent report that administrators fail to exercise their responsibility for removing bad teachers properly, with the remainder of respondents being unsure.

Work Rules. In addition to provisions governing compensation and personnel assignment, labor agreements contain a host of highly specific work rules that govern the day-to-day activities of teachers and school leaders. Contracts routinely stipulate the number of students a teacher will instruct, the number of courses a teacher may have, the number of parent conferences a teacher will hold, how long faculty meetings may last, and whether school leaders may outsource nonteaching operations.

Labor agreement restrictions on professional development are especially outdated and impede efforts to treat teachers as serious, committed professionals. For instance, contract provisions specifying that limited professional development funds must be allocated on the basis of seniority make these dollars more of a subsidy for veteran faculty to accrue additional credit hours or to travel than a lever for school improvement.

Finally, generous "leave" provisions give principals little ability to make sure their teachers show up regularly. The U.S. Department of Education reports that in 1999-2000 (the most recent year for which data are available), 5.2 percent of teachers were absent on a given day--about triple the 1.7 percent absentee rate that the Bureau of Labor Statistics reports for all managerial and professional employment.[6]

Table 1 provides an overview of aggregate scores for the fifty school districts in this study. Even after grading district scores on a rather generous curve, just five districts managed to qualify as "Flexible." No district achieved the 2.75 GPA needed to qualify as Highly Flexible (the highest score was 2.62, earned by Guilford County, North Carolina). Eight districts were deemed "Somewhat Flexible," yielding a total of thirteen districts with labor policies judged Flexible or Somewhat Flexible.

The "Somewhat Restrictive" category includes twenty-two districts that compiled an aggregate GPA between 1.67 and 2.00. The seven districts graded "Restrictive" earned GPAs between 1.33 and 1.66, while the Highly Restrictive category included eight districts that earned GPAs below 1.33. Fresno Unified School District in California posted the lowest GPA of all the districts studied with a 1.04.

Table 1 makes clear that hardly any contracts enshrine a high degree of flexibility, but also that less than half are explicitly and egregiously restrictive. What to make of this apparent paradox? Consider first the finding that no more than five labor agreements in the nation's fifty largest school districts guarantee anything approaching reasonable managerial discretion on the crucial dimensions of compensation, personnel policies, and work rules. (Interestingly, the five districts that deserve a closer look from reformers--Guilford County, Austin Independent, Northside Independent, Dallas Independent, and Fairfax County--are all in states where collective bargaining is not practiced.)

More than three-quarters of contracts are restrictive more often than they are ambiguous, and hardly any contracts explicitly grant managers the tools to woo promising employees or to exercise judgment when assembling their faculty--prerogatives regarded as typical in many high-performing for-profit and nonprofit organizations. This is not the fault of unions alone--school boards and superintendents, as well as state officials, have rarely helped and sometimes hindered district efforts on this count.

The second finding, though more subtle, is equally important. There appears to be more than a grain of truth to union complaints that collective bargaining agreements and labor policies are unfair scapegoats for inept or lethargic management. While hardly any labor agreements are truly conducive to effective leadership, only nine of the fifty examined are egregiously restrictive (i.e., rated Highly Restrictive), and thirty--60 percent of the districts studied--are ambiguous on key indicators (i.e., rated Somewhat Flexible or Somewhat Restrictive). Most districts operate under labor agreements that are notably vague on just what management can and cannot do with compensation, personnel policies, and work rules, suggesting that inaction on the part of district leaders cannot be explained simply by citing formal labor agreement restrictions.

Doubtless, other constraints in practice, culture, or local politics are also operating--and in many cases are the product of union efforts--but our findings indicate that the impact of labor agreements on school and district leadership is less straightforward than many previous accounts suggest.

Let us turn to the average grades for each of the three areas of interest. The mean GPA for compensation was 2.01 (C), for personnel policies 1.95 (C-), and for work rules 1.44 (D+).

Compensation. In the area of compensation, four districts received a grade of B or higher, seven districts received Ds or lower, another eleven received grades of D+, and two districts received Fs. The existence of collective bargaining in states has often been blamed for inflexible compensation policies and practices, but eight districts in non-collective-bargaining states received a D+ or lower. Meanwhile, the nine highest-scoring districts included four from the union strongholds of Maryland and California. In short, these dynamics may be more complex than many conventional accounts suggest.

Personnel Policies. For personnel policies, five districts earned grades of B or higher when it came to providing school leaders with the ability to make quality-conscious determinations about teacher evaluations, transfers, and layoffs: Austin Independent, Chicago, Fairfax County, Guilford County, and Northside Independent. Notably, all but one of these districts are in states where collective bargaining is not practiced; as for Chicago, a state law has taken much discretion away from the district and its teachers union in the area of personnel policies. Another twelve districts earned grades of D+, including Los Angeles Unified, Cleveland, Denver, and San Diego Unified. Meanwhile, six districts received grades of D or lower.

Work Rules. When it comes to rules governing professional development, labor outsourcing, faculty meetings, and mandatory leave for union business, the average district operates under policies that are substantially more restrictive than those governing compensation or personnel policies. Just three districts, all in Texas (where collective bargaining is prohibited), received grades of B- or higher: Northside Independent, Austin Independent, and Houston Independent. Meanwhile, nineteen districts earned Ds or lower, including four districts that earned Fs: Los Angeles Unified, Fresno Unified, San Diego Unified, and Miami-Dade County. Miami-Dade County is the only case in which a district scored 0.00 in any of the three categories, making Miami-Dade County's policies on work rules the national standard for inflexibility.

The results within the three major categories show substantially more variance than the aggregate GPAs; however, districts that receive high scores in one category tend to offset that accomplishment by scoring poorly in the other two. (After all, no individual district's overall GPA exceeded a meager 2.62.) Such variation makes the search for an exemplary district or for "best practices" for contract negotiation a frustrating exercise to be pursued with great caution. Overall top-scorer Guilford County, for example, managed to muster only a D+ in the work rules category.

There were three policies on which districts did an especially poor job of ensuring managerial flexibility. First was the professional development component. Twenty-eight districts require teachers to be paid a stipend or salary credit for attending professional development outside the scheduled workday. It is difficult to think of other knowledge-based professions where additional compensation is a prerequisite for professional training.

Districts received almost equally dismal scores on extra pay for teachers in shortage subject areas--specifically science, math, English as a Second Language, and special education. Thirty-one districts--more than 60 percent--prohibit efforts to adjust pay to attract teachers to hard-
to-staff subjects.

The third component on which districts fared particularly poorly was mandatory leave for union-related activities, where they averaged a D. Twenty-four districts require that some teachers be permitted to skip instructional time to tend to union affairs, potentially depriving students of valuable instruction.

There were, however, areas where districts did a consistently better job of affording flexibility to school and district leaders. The component on which districts scored highest (averaging a B-) was adjusting teacher compensation to account for previous experience.

Districts also scored relatively well when it came to factoring student performance into teacher performance evaluations. Although only eight districts explicitly permitted it, earning As, the lack of explicit prohibitions on the practice bumped up the average score to C+.

Finally, districts posted their third-highest score on the question of hardship pay, with a mean grade of C. There was, however, substantial variation among districts in their ability to adjust compensation for teachers taking on more challenging work environments, with ten districts earning As because language or policy explicitly enables such adjustments and another ten earning Fs because they explicitly prohibit such practices.

To make some deeper observations about these labor agreements, we examined how district scores correlated with various geographic, demographic, and political characteristics. Table 2 lists aggregate district grades based on poverty, racial composition, and enrollment figures, as well as on whether the district is located in a state where teachers negotiate through collective bargaining. (In this study, districts in Georgia, North Carolina, Texas, and Virginia, as well as Jordan School District in Utah, do not negotiate via collective bargaining.)

First, a comparison of mean scores for the twenty-two districts in which a majority of students are eligible for free and reduced lunch with the twenty-eight districts in which fewer than 50 percent are eligible shows that high-poverty districts operate in somewhat more restrictive environments. High-poverty districts had a mean GPA of 1.65 and a mean rating of Restrictive, while districts with lower poverty had a mean GPA of 1.92 and a mean rating of Somewhat Restrictive. The results are similar when district scores are broken down by the size of their minority enrollments. The sixteen districts where 75 percent or more of students are Latino or African American had a mean GPA of 1.71, while the other thirty-four districts had a mean GPA of 1.85. In short, districts serving the largest concentrations of poor and minority children are also those in which labor agreements are less flexible and more likely to hinder effective school leadership. Comparing districts on the basis of size, however, yields no obvious correlations.

Finally, when we compare districts in collective bargaining states with those where labor agreements take the form of board policies, we find a significant difference: the sixteen districts in non-collective-bargaining states average a GPA of 2.21 and a Somewhat Flexible rating, while the thirty-four districts in states that practice collective bargaining have a GPA of 1.63 and a mean rating of Restrictive. There are two interpretations of this finding, both offering useful insights and deserving further attention from policymakers and reformers. On the one hand, in proximate terms, non-collective-bargaining districts appear to be about 25 percent more flexible than their counterparts. On the other hand, this difference accounts for less than one-third of the spread between the highest- and lowest-scoring districts when comparing aggregate grades, and districts in non-collective-bargaining states still have relatively restrictive labor policies.

It should also be noted that all five districts rated Flexible are in North Carolina, Texas, or Virginia--three states where collective bargaining is explicitly illegal. (Although teachers in Georgia do not practice collective bargaining, it is not prohibited by state law.) Of course, such a correlation cannot prove that collective bargaining per se causes restrictive management policy, but it should at least direct the attention of would-be reformers toward these non-collective-bargaining states.

Overall, the plain language of labor agreements generally appears less restrictive than the most ardent union critics have suggested. The problem is less that collective bargaining agreements prohibit leaders from acting and more that they are murky; they send mixed signals about the bounds of permissible action; or they conflict with federal, state, judicial, or district practices and policies. While leaders in some industries would interpret these ambiguities as green lights, in the bureaucratic halls of public education--where risk-averse principals, central office administrators, school boards, and superintendents are rewarded for loyalty and encouraged by the community to avoid unseemly conflict--regulatory obscurity usually equals inaction. In no small part, this culture of timidity is the handiwork of local teacher associations, which are enormously influential in school district affairs. Such groups frequently unseat board members who do not toe the line and turn to other levers like state and federal laws and regulations to make life complicated for obstreperous district leaders.

The problem is less that collective bargaining agreements prohibit leaders from acting and more that they are murky; they send mixed signals about the bounds of permissible actions; or they conflict with federal, state, judicial, or district practices and policies.

This state of affairs illuminates a crucial reality: labor agreements do not operate in a vacuum. It may well be that complementary forces--at least as much as the black-and-white text of labor agreements--are responsible for the failure of districts to move forward on competing for talent, removing mediocre teachers, considering student achievement data in teacher evaluation, or rethinking work arrangements.

While a handful of districts have clearly done more to create flexible management environments than others--including Guilford County, Austin Independent School District, and Northside Independent School District--none of them could reasonably be deemed a shining example of doing it right. Moreover, districts that offer flexibility in one area have seemingly taken pains not to overdo it, and they routinely exhibit restrictive policies in another area.

The lesson that thoughtful reformers should take from these findings is not a need for new, one-size-fits-all solutions, like statewide performance-pay systems that force districts to deliver bonuses based strictly on test results. This would invite implementation debacles and likely hurl us from one era of compliant management into another. Rather, there is a need to move on multiple fronts to promote flexibility for district and school leaders--and to ensure that leaders are prepared for their roles and held accountable for using their authority effectively. A few recommendations from our research follow.

First, superintendents and school boards need to negotiate better agreements. In many districts, this means removing strictures that inhibit effective management, while in others it means working to win explicit contract recognition of managerial discretion. Obviously, efforts to upend established agreements will be arduous and unpleasant. Board members and superintendents can hope to succeed only if they have staunch backing from reformers, community leaders, and parents.

Second, we must promote transparency of what labor agreements say, build awareness of the problems that restrictive provisions cause, and highlight examples of flexible language that superintendents and school boards should pursue in new labor policies.
Third, while addressing formal policies is vital, critics may overstate the degree to which labor agreements themselves hamstring large districts. Most agreements are ambiguous (or silent) on key questions of compensation, personnel policies, and work rules. Superintendents must push principals to lead more aggressively with the authority they have, and boards must ask superintendents to lead more creatively within the parameters of existing agreements.

Finally, it is essential that advocates, policymakers, and funders keep pressing National Education Association (NEA) and American Federation of Teachers (AFT) locals to embrace the kind of rethinking that the United Auto Workers (UAW) accepted in 2007 when it overturned the 1957 Treaty of Detroit legacy and finally accepted a changed world for labor. NEA and AFT are among the few unions still adamantly opposed to the kind of operational agility that UAW and General Motors brought to American auto manufacturing with the first Saturn plant two decades ago.

The challenge for reformers and policymakers is to create labor agreements under which healthy management-employee relations can thrive. This will require policies that ensure that dynamic leadership does not depend upon or devolve into exhausting bureaucratic maneuvering. Ultimately, in a nation dotted with thriving public and private enterprises of all kinds, it does not seem too much to hope that we can empower leadership to transform even our most troubled schools into professional, talent-friendly communities.

Frederick M. Hess (rhess@aei.org) is a resident scholar and the director of education policy studies at AEI. Coby Loup (cloup@edexcellence.net) is a policy analyst at the Thomas B. Fordham Institute.

AEI research associate Rosemary Kendrick worked with Messrs. Hess and Loup to produce this Education Outlook.

1. Frederick M. Hess and Coby Loup, The Leadership Limbo: Teacher Labor Agreements in America's Fifty Largest School Districts (Washington, DC: Thomas B. Fordham Institute, February 2008), available at www.aei.org/publication27672/.
2. Each of the three was noted as an area of prime concern in Frederick M. Hess and Martin R. West, A Better Bargain: Overhauling Teacher Collective Bargaining for the 21st Century (Cambridge, MA: Program on Education Policy and Governance at Harvard University, 2006), available at www.aei.org/publication24167/.
3. Sean P. Corcoran, William N. Evans, and Robert S. Schwab, "Changing Labor Market Opportunities for Women and the Quality of Teachers, 1957-2000," The American Economic Review 94, no. 2 (May 2004): 230-35.
4. Michael Podgursky, "Personnel Policy in Traditional Public, Charter, and Private Schools," NCSC Review 1, no. 1 (2003): 10-13; Julie Kowal, Emily Ayscue Hassel, and Bryan C. Hassel, Teacher Compensation in Charter and Public Schools: Snapshots and Lessons of District Public Schools (Washington, DC: Center for American Progress, 2007).
5. Public Agenda, Stand by Me: What Teachers Really Think About Unions, Merit Pay and Other Professional Matters (New York: Public Agenda, 2003), 21.
6. Michael Podgursky, "Fringe Benefits," Education Next 3, no. 3 (Summer 2003): 71-76.