Collective Bargaining Agreement Restrictiveness in Unionized Charter Schools

Abstract

Although charter schools are frequently afforded flexibility from many state laws that govern traditional public schools, a growing number of charter school teachers have now unionized and introduced collective bargaining to the charter sector. Using data from a detailed content analysis of teacher CBAs from California, we compare the restrictiveness of CBAs in 75 unionized charter bargaining units to the restrictiveness of CBAs in 31 nearest neighbor traditional public school district bargaining units. We find that independent charter CBAs are much more flexible than the CBAs of traditional public school districts, but charter school CBAs of bargaining units combined with traditional public school districts are comparably restrictive.

Keywords

charter schools collective bargaining teacher unions teacher contracts

Introduction

Despite the generally contentious relationship between charter school operators and teachers’ unions, these institutions share the belief that teachers’ autonomy to make educational decisions based upon the needs of young people is important to student learning (Kahlenberg, 2007; Lubienski, 2003; Wohlstetter et al., 2013). In their infancy, education reformers conceived of charter schools as a way to relieve education from school district rules and regulations by allowing groups of teachers, parents, or community members decision-making authority in school design (Chubb & Moe, 1988; Wohlstetter et al., 1995). Early supporters of the charter school movement favored placing teachers, in particular, at the nexus of innovation. Supporters believed that it would raise the profession’s status and student achievement by giving educators the managerial autonomy to design school and classroom structures and develop new and creative approaches to instruction (Budde, 1988; Malloy & Wohlstetter, 2003). Albert Shanker, former president of the American Federation of Teachers, was an early proponent of the charter model for this reason (Kahlenberg, 2007; Kahlenberg & Potter, 2015). Since teachers’ unions generally shared a concern about the centralization of education policy that could lead to strict managerial control and a lack of teacher input, a close relationship between charters and unions seemed likely (Murphy, 1992; Urban, 2004).

Although both of the large national unions accepted the charter school movement early (Vergari, 2007), the movement and teachers’ unions diverged eventually. State charter school laws often provided flexible unionization rules, such as those that govern exclusive union representation and collective bargaining. Rather than work in conjunction with union rules, charter schools were free from union influence, which created an “us versus them” attitude between teachers’ unions and the charter sector (Jochim & Lavery, 2021). Further, the centralization of teachers’ voice in managing charter schools did not materialize (Kahlenberg & Potter, 2015; Torres, 2014). Instead, large charter management organizations (CMOs), which now operate approximately one-third of charter schools, proliferated and thereby challenged teachers’ autonomy (Torres, 2014). Consequently, teachers’ unions have made restricting charter school growth a tenet of their national and state policy platforms routinely (e.g., Busser, 2017; Camera, 2017; Wiley, 2019). At the same time, in some states like Illinois, teachers’ unions have focused simultaneously on organizing unions in existing charter schools. Proponents of charter school unionization argue that restrictions in the growth of charter schools and the increased unionization of charter school teachers is a way to reinstate teacher autonomy and voice in the management-dominated charter space. However, opponents of charter school unionization argue that doing so will limit the schools’ competitive advantage—the freedom from state and school district regulations. This could lead subsequently to increased homogeneity in policies between charter schools and traditional public schools.

One potential channel for these homogeneous (or isomorphic) policy influences is through the collective bargaining process between teachers’ unions and charter management. In traditional public schools, collective bargaining agreements (CBAs) between teachers’ unions and school districts have received significant scholarly attention, as they are the primary way in which teachers’ working conditions are established within schools in 43 states (Goldhaber et al., 2013, 2014; Strunk, 2011; Strunk & McEachin, 2011; Strunk & Reardon, 2010). Research on traditional public school district CBAs often captures the strength of teacher working conditions in CBAs by measuring these agreements’ restrictiveness (the degree to which they limit managerial autonomy over teachers). Restrictiveness in traditional public school CBAs is associated with school district expenditures and student achievement in important ways (Marianno, 2021; Marianno et al., 2021; Strunk, 2011, 2012; Strunk & McEachin, 2011). Notwithstanding autonomy’s importance in the charter school model, less scholarly attention has been paid to charter school CBAs’ restrictiveness.

There are reasons to believe that the presence of unions, and the CBAs they negotiate, could create increased isomorphic pressure on charter schools to implement policies applied to traditional public schools. For example, Lavery and Jochim (2022) and Jabbar et al. (2020) recent research suggested that charter school teachers unionize for many of the same reasons as traditional public school teachers. Consequently, the former may demand similar or identical protections in their CBAs as do those in traditional public school districts, which would lead to increasingly similar policies in both schools.

Even though charter school and traditional public school teachers unionize for similar reasons, very different policies could still exist within charter school CBAs. For example, a body of research has suggested that teachers in charter schools are quite different from their traditional public school peers (Ni, 2012; Cannata, 2011). Charter school teachers tend to be younger, less experienced, and hold fewer certifications than their public school counterparts. Further, evidence has indicated that they tend to identify with charters schools’ mission and vision (Cannata, 2011; Stuit & Smith, 2012). Charter school teachers work longer hours and more days as well, but feel a deeper sense of community and a greater degree of influence over school policies (Ni, 2017; Cannata, 2007). In short, because of the differences between the two types of teachers, charter school teachers may simply demand different protections than their public school counterparts during collective bargaining.

However, it is yet untested whether unionization in charter schools actually results in teacher working conditions that are more similar to those of their traditional public school counterparts. The specific constraints that unionization may place on charter school operations can be found in the CBAs established between teachers’ unions and charter school operators. Some scholars have used measures of contract restrictiveness to quantify and compare the relative constraints that the policies in the CBAs impose on schools (Marianno & Strunk, 2018; Strunk & Reardon, 2010). By employing a valid and reliable measure of the contracts’ restrictiveness, we can assess empirically the extent to which unionized charter school teachers adopt contracts that are similarly restrictive to those that cover traditional public school teachers. While our study cannot address whether these isomorphic pressures have a positive or negative effect on teachers or schools, we can assess the extent to which collective bargaining may constrain the initial concept of charter schools as laboratories of innovation in certain ways.

In this study, we drew on a sample of 75 unique CBAs from 188 unionized charter schools in California (86% of all unionized charter schools in the state and 14% of all charter schools in the state). Like most states, California does not require charter schools to engage in collective bargaining, although they can opt to do so when they seek a school authorizer’s approval for their charter (National Alliance for Public Charter Schools, 2022). Within this context, we asked: How do the policies contained in charter school CBAs compare to those in traditional public school districts? Following a growing body of literature on teacher CBAs in traditional public school districts (Goldhaber et al., 2013, 2014; Strunk, 2011; Strunk & McEachin, 2011; Strunk & Reardon, 2010), we generated measures of CBA restrictiveness from three types of bargaining units: independent charter school bargaining units (I-CU) formed for a single charter school or group of schools, traditional public school units (TPS-CU) that include charter school teachers and generally result from the conversion of a traditional public school into a charter, and traditional public school district, non-charter bargaining units (TPS-U). Similar to other research on the topic, we defined CBA restrictiveness as the degree to which a contract contains items that place boundaries on management’s role in teachers’ work (Strunk & Reardon, 2010). We compared I-CU, TPS-CU, and TPS-U agreement restrictiveness overall and in a number of different areas of the CBA. Subsequently, we compared these agreements in a number of the most restrictive bargaining agreement provisions that are found within teacher contracts in the state of California.

We found that 56% (42) of the unionized charter bargaining units are covered by a TPS-CU contract. For context, 48% of unionized charter schools nationally are covered by these contracts (National Alliance for Public Charter Schools, 2020). The remaining 44% (33) are I-CU contracts from independent charter school bargaining units. We found that I-CU agreements were much more flexible than TPS-U agreements (1.217 SD less restrictive than TPS-U agreements, p < .01), particularly in terms of retirement, non-teaching duties, transfers, and evaluation. However, I-CU agreements were equally restrictive as TPS-U agreements in certain areas of the contract (i.e., compensation, health benefits, union rights, grievances, leaves, workload, and class sizes). With respect to TPS-CU contracts, we found that these agreements were similarly as restrictive as TPS agreements overall and in all working condition areas except for teacher leave provisions.

Taken together, we provide evidence that contracts from I-CUs are not yet equal to the CBAs in traditional public school districts. We show further that the composition of the bargaining unit structure has important effects on the way that unionization influences charter school policies. Bargaining units that contain both TPS and charter school teachers result largely in CBAs with similar levels of restrictions on managerial discretion. However, there are clear differences in the contract restrictiveness between I-CU CBAs and neighboring TPS-U agreements. This indicates that notwithstanding the onset of unionization, managers of charter schools covered by I-CU agreements still have more autonomy in determining teachers’ working conditions than leaders in traditional public schools. It may also suggest that in certain key aspects of teacher work, charter school teachers prefer different working conditions. Alternatively, it may also mean that their unions are not yet sufficiently strong to negotiate for working conditions at levels similar to those in TPS-U agreements. Ultimately, our research provides an important starting point from which to assess the way that I-CU and TPS-CU agreements change over time relative to TPS-U agreements, and the implications of the way that unionization transforms charter schools and charter school teaching.

This paper proceeds as follows: First, we review the existing literature on teachers’ union and collective bargaining in traditional public schools, including the way that CBA restrictiveness is measured. Next, we review research on collective bargaining in charter schools. Then, we present institutional theory as a way to understand the potential pressures that CBAs exert on charter schools. Thereafter, we provide our data, methods, and results, and conclude with our work’s implications for researchers and policymakers.

Literature Review

Teachers’ Unions, Collective Bargaining, and Autonomy

Much of what we know about unionization’s effect on schools derives from research on traditional public school districts. Teachers’ unions and the CBAs they sign with traditional public school district management have long been acknowledged as a force to standardize teacher working conditions in education (Eberts & Stone, 1987). Negotiated at the district-level, these agreements create fairly uniform workplace rules for all teachers within a school district (e.g., all teachers work a specified number of hours each day; receive a specified number of sick days). Early public school teacher CBAs focused primarily on what are considered “bread-and-butter” working conditions, coupled with more professional contract provisions designed to give teachers greater autonomy (and administrators less) in their work (McDonnell & Pascal, 1988). Additional research has suggested that the content of traditional CBAs is now rather stable, with large changes to contracts made only when there is a precipitous event such as an economic downturn or large state policy changes (Cook et al., 2021; Cowen & Fowles, 2013; Strunk et al., 2022).

CBA Restrictiveness

Measuring the restrictiveness of the content of traditional public school district CBAs is one common approach to understanding what teacher contracts contain. This approach stems from the idea that bargaining agreement policies arise from bilateral negotiations between teachers’ unions and district management and establish boundaries or restrictions on the degree to which management can direct teachers’ work (Eberts, 1983; Eberts & Stone, 1984; Goldhaber et al., 2013, 2014; Strunk, 2011; Strunk & McEachin, 2011; Strunk & Reardon, 2010). However, while conceived typically as a restriction on management, Johnson and Kardos (2000) noted that these restrictions constrain teachers as well. Even if a teacher wanted to work additional hours to support students, these kinds of behaviors are considered often times a betrayal of unions’ collective action to create fair and equal working conditions for all union members.

Prior research has suggested that traditional public school district CBAs’ restrictiveness varies widely both across and within states based upon certain characteristics of unions and school districts (Eberts, 1983; Eberts & Stone, 1984; F. M. Hess & Kelly, 2006; Strunk et al., 2018, 2022). For example, CBAs in urban public school districts are often much lengthier and contain more restrictive working conditions than those in suburban and rural districts (e.g., Marianno et al., 2018). In addition, districts with a high proportion of underrepresented minority and low-income students have more restrictive contracts (Goldhaber et al., 2014). Bargaining agreements that share such geographic attributes as proximity or are negotiated under the same union leadership are similarly more restrictive than those from districts that are not located proximally or are negotiated by different union leadership (Goldhaber et al., 2014).

Further, research has found that CBAs’ restrictiveness is important in the way that school districts expend funds and how much students learn. CBA restrictiveness is associated positively with school district expenditures on students, instruction, and instruction support services, but not associated significantly with student achievement on standardized exams (Marianno et al., 2021). In addition, restrictive CBA content neither influences all students’ learning equally, nor do all areas of the CBA have a uniform effect on student achievement. For instance, Marianno (2021) found that increases in the restrictiveness of CBA class size policies are associated with a decrease in math achievement gaps between Black and White students. However, he found further that an increase in the restrictiveness of teacher leave policies is associated with an increase in math achievement gaps between Black and White students.

Collective Bargaining in Charter Schools

The number of charter schools in the United States continues to grow steadily, with over three million students enrolled annually or 7% of the United States student population (up from 1.6 million and 3% in 2009) (David & Hesla, 2018; Hussar et al., 2020; National Center for Education Statistics, 2022). In turn, teachers’ unions have made steady progress in unionizing charter schools (i.e., the number of unionized charter schools increased from 604 in 2009–10 to 782 in 2016–17), although the rate of charter school unionization has remained relatively stable over time, as 11.3% of charter schools are unionized (Jochim & Lavery, 2021).

Yet, although the number of unionized charter schools in the United States has increased, very few studies have focused on their CBAs. In one of the first studies on the topic, Price (2011) explored the restrictiveness of nine charter school CBAs using a method that F. Hess and Loup (2008) developed to score contracts on a 4.0 GPA scale. While F. Hess and Loup (2008) gave a sample of traditional public school contracts a C-minus for managerial flexibility, Price (2011) scored a sample of charter school CBAs a C-plus for managerial flexibility. He found that charter contracts had more flexible grievance procedures and workday schedules, but similar teacher compensation structures compared to traditional public school districts. Jochim and Lavery (2021) also analyzed nine charter school CBAs from 36 charter schools using F. Hess and Loup’s (2008) method. The CBAs in their sample scored an average of C on the scale. They found that the workday and teacher evaluation policies resembled nearby traditional public school districts, but the charter contracts had more flexibility with respect to compensation and grievances. Woo and Marianno (2021) conducted a descriptive examination of charter school CBAs in Chicago. Similar to previous researchers (e.g., Marianno & Strunk, 2018; Strunk & Reardon), they investigated high-profile provisions and found that charter school CBAs in Chicago, while significantly shorter, are similar to Chicago public schools’ CBAs.

Our study differs from past research on charter school contracts in several ways. First, we estimate formal measures of CBA restrictiveness based upon the Partial Independence Item Response (PIIR) approach used to study traditional public school district contracts (Strunk & Reardon, 2010). This allows us to incorporate (and classify the restrictiveness of) over 500 different contract items found in CBAs commonly, which is far more extensive than in past work. The restrictiveness of CBAs is a common critique leveled against the unionization of charter schools and focuses frequently on the way that restrictive CBA content could limit their competitive advantage in innovative personnel policies (i.e., staffing, salary, discipline, dismissal, hiring) (Jabbar, 2018). However, there is still very little empirical work on how much charter school CBAs borrow from, or look like, their traditional public school counterparts. Second, the content of charter school CBAs has not been explored systematically on a large scale, as within a single state, which makes it difficult to judge fully how charter CBA content varies across school contexts. While prior work has examined fewer than 10 CBAs, our work incorporates a much larger sample of 75 contracts, and thereby affords a more detailed look at the variation in charter school contracts’ restrictiveness. Third, prior studies have not used a systematic coding scheme to capture charter CBAs’ restrictiveness, as is used commonly in the study of traditional public school districts’ CBAs. This makes it more difficult to compare the restrictiveness of charter school CBAs to the restrictiveness of CBAs found within traditional public school districts. Finally, we take advantage of institutional theory to conceptualize some of the forces that could make charter school agreements similar and different from their traditional public school district counterparts. Consequently, this study resolves key shortcomings in prior research by collecting and analyzing charter school and traditional public school district CBAs in California systematically, and then grounding the analysis in theory to provide the most comprehensive exploration of charter school CBAs to date.

Institutional Theory as a Conceptual Framework to Understand Charter School Collective Bargaining

To understand the “… complex relationship between decentralized schools and their environment” better, Huerta and Zuckerman (2009, p. 414) applied institutional theory to charter schools based upon Meyer and Rowan’s (1977, 1978) initial application of the theory to traditional public schools. Institutional theory is a helpful lens through which to understand charter school policymaking, as it emphasizes organizations’ internal struggle to obtain legitimacy in their institutional environment through conformity to normative rules (i.e., the way that schooling is supposed to be conducted), rather than through rational decision making that adheres to their goals (i.e., the mission of the school) and pursues technical efficiency (i.e., instructional effectiveness).

Flexibility in many of the state and district regulations that govern their traditional public school counterparts is critical to the theory of action of charter schools (Ableidinger & Hassel, 2010). By taking advantage of their policy autonomy, charter schools could feasibly serve as organizations that challenge the traditional norms of schooling and find new ways to improve educational effectiveness. A close emphasis on goal driven effectiveness is what Huerta and Zuckerman (2009) referred to as their “decentralized frame” (p. 415). However, as they noted, charter schools are not insulated from the wider institutional environment of schooling. For example, they are subject to state accountability pressures to improve student performance and must attract and retain students and teachers at their school (Henig et al., 2005; Lubienski & Lee, 2016; Renzulli et al., 2015). These pressures could lead charter schools to tend toward a bureaucratic frame, or toward a subjection to institutional rules that define what legitimate schooling looks like. Charter schools are caught between these two frames, in which market forces and their school’s local demands push them toward a decentralized frame, while their need for legitimacy coupled with accountability and regulatory pressures pull them toward a bureaucratic frame.

Recent evidence has emerged that suggests that innovation in the charter sector is becoming increasingly rare (Lubienski & Lee, 2016; Renzulli et al., 2015). Henig et al. (2005) found that despite differences in charter school organizers, charter school behaviors often converged. Lubienski and Lee (2016) found that charter school mission statements appear to have become more similar over time. Similarly, Renzulli et al. (2015) found that while some specialized charter schools appear to be becoming more distinct, the number of these schools has declined over time. These studies all suggest to some degree that charter schools are experiencing isomorphic pressures that make them less distinct from one another over time.

We argue that the charter school collective bargaining process could also be consistent with either the decentralized frame or the bureaucratic frame. In short, charter schools could be hubs for innovation in union contracts, consistent with the decentralized frame, or charter schools and unions could create contracts that contain many of the same working condition policies found in traditional public schools, and thus shift charter schools toward the bureaucratic frame. One factor at play is the process by which bargaining becomes institutionalized within charter schools. In California, the setting for this study, teachers in new charter schools can follow the process that traditional public school teachers use to organize a union and bargaining unit to engage in collective bargaining. Traditionally, these bargaining units are local affiliates within the National Education Association (NEA) or the American Federation of Teachers (AFT) and they are covered under a unique charter contract negotiated specifically for that school or group of schools (i.e., an independent charter unit or I-CU). Alternatively, some charter schools, particularly those that are converting from a traditional public school, join an existing bargaining unit of traditional public school teachers and thereby, are covered by the same agreement with only slight or sometimes no modifications (i.e., a traditional public school charter unit or TPS-CU).

There is a greater opportunity for innovation within the I-CUs than in TPS-CUs that employ both charter and traditional public school teachers. Within I-CUs, management and teachers’ unions are free to incorporate charter teachers’ voices in the bargaining process to take into account the unique needs of students when teachers’ working conditions are established (Freeman & Medoff, 1984). In this way, their working conditions are democratized in a similar way as Albert Shanker envisioned union involvement in charter schools originally (Kahlenberg, 2007; Kahlenberg & Potter, 2015; Peterson, 2010), which could shift charter schools toward the decentralized frame.

Alternatively, charter and union leaders in I-CUs could adhere to a nearly identical collective bargaining process and focus as that found in traditional public schools. Because local affiliates of the NEA and AFT negotiate many charter school CBAs, they could adopt provisions and language from those in traditional public schools. Further, teachers in charter schools could demand similar protections as those found in traditional public schools in an effort to garner legitimacy in their schools. Jabbar et al. (2020) and Lavery and Jochim’s (2022) recent work suggested that charter school teachers unionize for many of the same reasons as traditional public school teachers (e.g., improving teacher retention and school stability). It may be that the integration of collective bargaining into charter schools leads to the creation of union contracts that contain many of the same working condition policies found in traditional public schools, and thus shifts charter schools toward the bureaucratic frame.

Methods

To explore variation in the policies contained in charter school and traditional public school district contracts, we draw upon data from a self-collected database of contracts from unionized charter schools and their nearest neighbor traditional public school district within California. In what follows, we describe our coding procedure and the way that we generated measures of CBA restrictiveness.

California Charter School CBAs

During the 2018-19 school year, California had more charter school campuses open than any other state (1,351) (White et al., 2020). Among all students enrolled in public schools, California also had the third highest percentage (10.6%) of children enrolled in charter schools (behind Arizona at 18% and Florida at 11%) (White et al., 2020). Further, California represents a key focal point in the charter unionization movement. Approximately 10% of all charter schools in the United States are unionized (718 schools), while 14% of California charter schools have unions (National Alliance for Public Charter Schools, 2020), and 34% of all unionized charter schools in the United States are located in California. Although California is a hub for the unionization of charter schools, it is important to recognize that the vast majority of charter schools (86%) in the state are not subject to CBAs’ terms and conditions.

We obtained data on the union status of charter schools in California from the National Alliance for Public Charter Schools during the 2018-19 school year. We requested the CBAs from all of the 247 unionized charter schools in California. We confirmed that 26 charter schools had closed or were not actually unionized. From the remaining 220 schools, we obtained the CBAs for 188 schools (85% response rate). These 188 schools constitute 75 unique bargaining units, each with its own CBA (i.e., some charter management organizations negotiate one contract for all of their schools). Supplemental Appendix Table 1 presents t-tests of the differences in average school characteristics for schools that responded to our request relative to those that did not. Non-responders tend to have lower enrollment; however, they have similar percentages of FRL and minority students. Further, they are less likely to be located in urban areas and more likely to be located in suburban areas.

Types of Contracts Analyzed

A traditional public school charter union (TPS-CU) bargaining unit, or bargaining unit that combines traditional public and charter school teachers because a school was converted from a traditional public school to a charter school is the most common charter school bargaining unit classification in California. These schools are represented by the same union and covered under the same CBA as other teachers in the school district, with several select sections of the contract that alter certain workplace rules for the charter campuses. We refer to the resulting contracts from these bargaining units as TPS-CU contracts. As shown in Table 1, TPS-CU contracts cover 56% (n = 42) of the unionized charter bargaining units. By comparison to national statistics, 48% of unionized charter schools are covered by a TPS-CU contract, suggesting that TPS charters are slightly more common in California than nationally (National Alliance for Public Charter Schools, 2020).

Table 1.

Average Characteristics of IC, TPS-C, and TPS Bargaining Units.

	Independent Charter Units (I-CU)	Traditional Public School Charter Units (TPS-CU)	Traditional Public School Units (TPS-U)
School characteristics
% Urban	33.70 (26.20)	38.44 (44.32)	39.03 (47.59)
% Suburban	64.78 (25.63)	44.05 (45.78)	52.23 (45.78)
% Rural	1.52 (8.70)	17.50 (37.69)	8.74 (22.88)
Enrollment (ln)	6.52 (0.90)	5.91 (0.99)	6.29 (0.44)
% Minority	65.37 (27.11)	59.82 (22.26)	68.73 (18.40)
% FRL	53.80 (27.45)	51.54* (23.30)	64.18 (18.04)
Union characteristics
% NEA	66.67%	85.71%	80.95%
% AFT	6.06%	2.38%	9.52%
% Joint AFT/NEA	24.24%	4.76%	9.52%
% Unknown Union	3.03%	9.52%	0.00%
Years certified	9.71*** (4.50)	38.57 (9.122)	35.90 (9.33)
Union members (ln)	3.41*** (1.05)	5.44 (1.64)	6.28 (2.00)
N	33	42	21

.05, **.01, ***.001; Two-sample independent t-tests test the difference in means in continuous school and union characteristics compared to Traditional Public School Units; Standard deviations in parentheses.

The remaining 44% (n = 33) of charter school bargaining units are formed from independent charter schools, charter school networks, or from district-sponsored charter schools that opted to form their own bargaining unit for charter school teachers. We refer to the resulting contracts from these bargaining units as independent charter bargaining units (I-CU). By comparison to the national percentage, 52% of unionized charter schools are covered by independent charter contracts (National Alliance for Public Charter Schools, 2020).

For the 33 I-CUs, we also collected the CBAs from the nearest neighbor traditional public school bargaining unit (TPS-U school), which were generally from the school district with which the charter school would be affiliated because of their proximal location if they did not run independently). Prior research on traditional public school districts has suggested that neighboring school district contracts share similar provisions (Strunk et al., 2022). The 33 I-CUs span 21 nearest neighbor school districts.

Generating Measures of CBA Restrictiveness

We modeled our coding of CBAs after the process that Strunk and Reardon (2010) pioneered and is used extensively to study bargaining agreements in traditional public school districts in California and elsewhere (e.g., Goldhaber et al., 2013; Marianno & Strunk, 2018; Strunk, 2011; Strunk & Reardon, 2010). Strunk and Reardon’s (2010) coding protocol assesses California CBAs’ restrictiveness by asking a series of yes/no questions about the types of provisions in the agreements. We adopted this protocol wholesale, given that we suspected that charter school CBAs in California would be somewhat similar to their traditional public school counterparts and given our express purpose to compare the two types of contracts. While coding the charter school contracts, we kept a list of provisions that were unique to their agreements, but they did not meet Strunk and Reardon’s (2010) frequency threshold (i.e., appearing in 20% of contracts) for inclusion in the final coding instrument. Provisions that appear infrequently in the agreements or are unique to a single charter school provide very little information about the restrictiveness of the contract overall (Strunk & Reardon, 2010).

We coded 531 contract items in each contract across 11 common areas in California teacher contracts (union rights, grievance procedures, class sizes, workload, non-teaching duties, evaluation, transfers and vacancies, compensation, health benefits, leaves, and retirement incentives). The common subareas of the CBA are based upon the groupings from prior research on traditional California school district CBAs’ restrictiveness and are based more generally upon a typology of working conditions that Johnson (1990) proposed to categorize the different types of teacher working conditions (political, economic, physical, organizational, psychological, cultural, and sociological). Figure 1 shows the way that our CBA subareas fit into the different categories of teacher working conditions that Johnson (1990) noted.

Figure 1.

Framework for Understanding Teacher Working Conditions in CBAs.

To create each restrictiveness measure, we used a Partial Independence Item Response Model (PIIR) described in detail in Strunk and Reardon (2010) and Goldhaber et al. (2013). The PIIR model operates on the premise that each area of the CBA has a latent restrictiveness that can be inferred from the presence or absence of various provisions that constrain district operations in that area of the contract, much as a correct or incorrect answer on a test can be used to infer a student’s latent understanding of a subject. The PIIR model allows for the fact that some provisions will be interdependent, such that a more restrictive provision can be present only if a less restrictive gate provision is also present (e.g., the teacher workday cannot be limited to 7 hr or less unless it is also restricted to 7.25 hr or less). Thus, while each contract provision is measured dichotomously based upon its presence or absence, more and less restrictive provisions that govern the same district behavior can be incorporated as well by modeling their interdependence.

We began by performing an exploratory Cronbach’s alpha analysis on the 531 coded contract items. The purpose of this analysis is to improve the reliability and usability of the measures of contract restrictiveness by reducing the number of items (removing items that have a low item-total correlation statistic of less than .25) to those that are correlated most with the latent trait of restrictiveness. For example, after the Cronbach’s alpha analysis, 150 items remained in the overall measure of contract restrictiveness. We followed the same process to generate measures for the 11 subareas of the contract mentioned above. The items in the restrictiveness measure overall are provided in Supplemental Appendix Table 2.

To estimate the PIIR model, we define Y_ik as equal to one if restrictive item k is present in contract i, and zero otherwise, in which h_ik indicates similarly whether item k’s gate item is present. Then, $φ_{ik}$ equals the probability that $Y_{ik}$ equals one conditional on $h_{ik}$ as shown in Equation 1.

φ_{ik} = \Pr (Y_{ik} = 1 | h_{ik} = 1)

(1)

We estimate the structural model as follows:

\log [\frac{φ_{ik}}{1 - φ_{ik}}] = θ_{i} + \sum_{j = 1}^{k} γ_{j} D_{ij}

(2)

The PIIR model is estimated formally as a hierarchical random effects model that takes into account the contracts’ conditional (or gate) structure, as mentioned above. Following Strunk and Reardon (2010), we model the probability that a contract i contains a given item k as a function of the contract’s latent restrictiveness $θ_{i}$ and a vector of dummy variables, $D_{ij}$ , that captures which item is considered, in which the coefficient on the dummy variable, $γ_{j}$ , represents the provision’s conditional severity (i.e., each item’s restrictiveness). Again, $θ_{i}$ is a random effect for each contract that captures that contract’s latent restrictiveness, and is the main term of interest we use as the dependent variable in our analytic models below. To help interpret the measure further, we standardize it, such that differences in the contracts’ restrictiveness can be interpreted with respect to standard deviations.

Analysis

To answer our research question, “How do the policies contained in charter school CBAs compare to those in traditional public school districts?,” we estimate a series of Ordinary Least Square (OLS) regression models of the following form:

θ_{i} = β_{o} + β_{1} I - C U_{i} + β_{2} TPS - C U_{i} + B_{i} \partial + ε_{i}

(3)

$θ_{i}$ is our outcome variable, the subarea or contract restrictiveness overall for a bargaining unit contract, i. I-CU_i is an indicator variable that equals 1 if a bargaining unit contract derives from an independent charter bargaining unit and zero otherwise. Similarly, TPS-CU_i is an indicator variable that equals 1 if a bargaining unit contract derives from a traditional public school charter bargaining unit and zero otherwise. TPS-U contracts serve as the reference category, such that $β_{1}$ captures the differential restrictiveness between I-CU contracts and TPS-U contracts, and $β_{2}$ captures the differential restrictiveness between TPS-CU contracts and TPS-U contracts. $B_{i}$ is a vector of bargaining unit controls, including the percentage of schools in the bargaining unit that are in urban locations, mean school enrollment (natural logged), the school’s mean percentage of minorities, the school’s mean percentage of students who qualify for a free-and-reduced price lunch (FRL), the bargaining unit’s certifying union (National Education Association, NEA, Joint NEA/American Federation of Teachers, AFT, Unknown, in which AFT serves as the reference category), the number of years the bargaining unit has been certified, and the size of the union membership (natural logged). Table 1 shows these variables’ summary statistics by bargaining unit type.

We also estimate a variation of equation three to explore differences in the presence and absence of the most restrictive contract items based upon the value of $γ_{k}$ in the PIIR model above. We take the top three most restrictive items from each subarea of the contract and run a series of linear probability models using equation three to predict the probability that these items appear in a bargaining unit’s CBA based upon the bargaining unit’s type and characteristics.

Finally, we run a version of equation three with the restrictiveness of the I-CU’s nearest-neighbor CBA from the TPS-U as a control variable.¹ A positive association on the beta coefficient on the TPS-U nearest-neighbor restrictiveness measure would provide suggestive evidence that IC-Us agreements are modeled after their closest TPS-U counterpart.

Results

Before exploring the differences in the restrictiveness of I-CU, TPS-CU, and TPS-U agreements, we compared these units first with respect to important school and union characteristics that prior research has shown predict CBA restrictiveness in TPS-Us (e.g., Strunk, 2011, 2012).

As shown in Table 1, we found little evidence of statistically significant differences in school and union characteristics between I-CUs and TPS-CUs relative to TPS-Us, as determined through a two-sample independent t-test for differences in means. A similar percentage of schools in all three bargaining units were in urban locations (34%, 38%, and 39%, respectively). TPS-CUs have fewer schools in suburban locations (44%) than TPS-Us (52%) and more schools in rural locations (18%) than TPS-Us (9%); however, these differences were not statistically significant. Further, schools in IC-Us and TPS-CUs had similar enrollments and percentages of minority students relative to TPS-Us. However, TPS-CUs had a lower percentage of FRL students (52%) compared to TPS-Us (64%). With respect to union characteristics, I-CUs were more likely to be represented by a joint AFT/NEA collaboration (24%) than TPS-U bargaining units (10%). Finally, and unsurprisingly, I-CU bargaining units have not existed as long as TPS-U bargaining units (a mean of 10 years relative to 36 years), and tended to have fewer union members.

Table 2 shows the unconditional mean restrictiveness of IC-U, TPS-CU, and TPS-U by subarea and contracts overall and tests again for statistically significant differences between IC-Us and TPS-CUs and TPS-Us using a two-sample independent t-test. Specifically, the table provides a first look at the areas in which IC-U and TPS-CU contracts provide more flexibility to administrators relative to their TPS-U counterparts, as well as areas in which the two agreements match TPS-U agreements’ restrictiveness. We found that I-CU agreements are much less restrictive than TPS-U agreements by 1.555 SD. To put this difference in overall restrictiveness in context, we can take advantage of the conditional severities from the PIIR model to generate the predicted probability that a given restrictive item will appear in the contract (see Supplemental Appendix Table 2) at the mean level of restrictiveness (shown in the center column labeled “0”) and at one SD above (+σ) and below (–σ) the mean level. One of the most restrictive items in the restrictiveness measure overall based upon the item severity is a provision that allows teachers less frequent evaluations in the case of prior satisfactory evaluations (evht). The mean CBA with respect to restrictiveness has a 46% probability of containing this provision. A contract that is one SD more restrictive than the mean CBA in the sample has a 70% probability of having this item.

Table 2.

Average Restrictiveness of ICS-U, TPS-CU, and TPS-U Bargaining Units.

	Independent Charter Units (I-CU)	Traditional Public School Charter Units (TPS-CU)	Traditional Public School Units (TPS-U)
Overall restrictiveness	−0.851*** (0.817)	0.354 (0.632)	0.700 (0.997)
Economic working conditions
Compensation	−0.437* (0.788)	0.257 (1.111)	0.113 (0.886)
Health benefits	−0.067* (1.190)	−0.169** (0.903)	0.487 (0.696)
Retirement incentives	−0.581*** (0.758)	0.220 (0.969)	0.502 (0.986)
Political working conditions
Union rights	−0.402** (0.693)	0.179 (1.117)	0.356 (1.023)
Physical working conditions
Grievance procedures	−0.260 (1.030)	0.060 (0.825)	0.332 (1.188)
Leaves	−0.444*** (0.803)	0.030* (0.863)	0.695 (1.175)
Organizational working conditions
Workload	−0.172 (1.027)	0.147 (0.988)	0.011 (0.972)
Non-teaching duties	−0.445** (1.027)	0.137 (0.915)	0.439 (0.856)
Transfers	−0.878*** (0.885)	0.452 (0.568)	0.532 (0.927)
Class sizes	−0.281 (0.754)	0.168 (1.079)	0.123 (1.097)
Evaluation	−0.303 (1.239)	0.127 (0.641)	0.233 (1.078)
N	33	42	21

Note.*.05, **.01, ***.001; Two-sample independent t-tests test the difference in means compared to Traditional Public School Units; Standard deviations in parentheses.

Figure 2 shows the differences in restrictiveness visually, in which the solid red line represents the distribution of I-CU contract restrictiveness, and the dotted red line represents the mean of I-CU contract restrictiveness. The blue solid and dotted lines represent the same for TPS-CU agreements, and the black solid and dotted lines represent the same for TPS-U contracts. Notably, some I-CU contracts are more restrictive than some TPS-CU and TPS-U contracts, as indicated by the intersection of the solid red line with the solid blue and black lines.

Figure 2.

Kernel Density Plot of Independent Charter Unit, Traditional Public School Charter Unit, and Traditional Public School Unit Contract Restrictiveness Overall.

Table 3 shows the results from equation three, which tested whether the difference in I-CU and TPS-U contract restrictiveness holds, conditional on bargaining unit controls. We found that even after school and union bargaining unit characteristics are controlled, I-CU agreements are 1.217 SD less restrictive than TPS-U agreements (p < .01). We found that TPS-CU agreements’ restrictiveness does not differ significantly compared to TPS-U agreements, conditional on controls. After mean differences across the bargaining unit categories (TPS-U, TPS-CU, and I-CU) were controlled, the only bargaining unit characteristic that had a statistically significant association with contract restrictiveness is union membership. A 10% increase in union membership is associated with a 0.006 SD ((0.160 * log (1.1)) increase in contract restrictiveness (p < .05).

Table 3.

OLS Regressions Predicting Bargaining Unit Agreement Restrictiveness Overall and by Working Condition Subareas.

	Overall	Economic Working Conditions			Political Working Conditions	Physical Working Conditions
	Overall	Compensation	Health Benefits	Retirement	Union Rights	Grievances	Leaves
I-CU	−1.217** (0.393)	−0.721 (0.525)	−0.700 (0.508)	−1.327** (0.472)	−0.650 (0.513)	−0.492 (0.502)	−0.182 (0.434)
TPS-CU	−0.227 (0.219)	0.354 (0.293)	−0.528 (0.284)	−0.319 (0.264)	0.014 (0.287)	−0.051 (0.281)	−0.605* (0.243)
% Urban	0.001 (0.002)	0.000 (0.003)	0.003 (0.003)	0.006 (0.003)	−0.004 (0.003)	−0.002 (0.003)	0.007* (0.003)
Enrollment (ln)	−0.042 (0.104)	−0.086 (0.139)	0.047 (0.134)	0.074 (0.125)	0.155 (0.136)	−0.103 (0.133)	−0.219 (0.115)
% Minority	0.008 (0.007)	0.004 (0.009)	0.007 (0.008)	0.011 (0.008)	0.002 (0.009)	0.009 (0.008)	0.002 (0.007)
% FRL	−0.010 (0.006)	−0.001 (0.008)	−0.008 (0.008)	−0.012 (0.007)	0.002 (0.008)	−0.013 (0.008)	−0.005 (0.007)
NEA	−0.009 (0.331)	−0.281 (0.443)	0.403 (0.429)	0.545 (0.398)	0.670 (0.433)	−0.403 (0.424)	−0.137 (0.367)
Joint AFT/NEA	−0.679 (0.418)	−0.424 (0.559)	0.074 (0.542)	0.048 (0.503)	0.573 (0.546)	−0.627 (0.535)	−0.309 (0.463)
Unknown Union	−0.558 (0.444)	−0.649 (0.594)	0.339 (0.576)	−0.066 (0.534)	0.321 (0.580)	−1.420* (0.568)	0.010 (0.492)
Years Certified	−0.007 (0.010)	−0.025 (0.014)	−0.021 (0.013)	−0.005 (0.013)	−0.010 (0.014)	−0.018 (0.013)	0.005 (0.012)
Union Membership (ln)	0.160* (0.064)	0.110 (0.085)	0.160 (0.083)	−0.026 (0.077)	0.134 (0.083)	0.181* (0.082)	0.240** (0.071)
Constant	0.282 (0.725)	0.911 (0.969)	−0.480 (0.939)	−0.276 (0.871)	−1.781 (0.947)	1.121 (0.927)	0.483 (0.802)
N	91	91	91	91	91	91	91
R-Sq	.524	.144	.197	.316	.201	.219	.402

Note. *.05, **.01, ***.001; standard error in parentheses.

Table 1 also summarizes the unconditional difference in CBA subarea restrictiveness estimates for each bargaining unit/contract type. Here we interpret the conditional differences generated from Equation 2 and shown in Tables 3 and 4. As Table 3 shows, once bargaining unit characteristics were controlled, we found no statistically significant difference between I-CU and TPS-U agreements in economic conditions (i.e., compensation, and health benefits); however, we note that these and most other estimates of the differences between I-CU and TPS-U agreements are still relatively large in magnitude (0.721 SD for compensation and 0.700 SD for health benefits) but are simply measured imprecisely. I-CU agreements are significantly less restrictive with respect to retirement provisions (1.327 SD). We found no statistically significant differences between I-CU agreements and TPS-U agreements in political and physical working conditions. As shown in Table 4, I-CU agreements are significantly less restrictive than TPS-U agreements in several organizational working condition areas, including non-teaching duties (−0.965 SD, p < .05), transfers (−1.398 SD, p < .001), and evaluation (−0.995 SD, p < .05), but their restrictiveness does not differ significantly from that of TPS-U agreements in workload and class sizes. With respect to TPS-CU contracts, we find that these agreements’ restrictiveness does not differ significantly from that of TPS agreements in all working condition areas except for teacher leave provisions, where they are −0.605 SD less restrictive (p < .05).

Table 4.

OLS Regressions Predicting Bargaining Unit Agreement Restrictiveness Overall and by Working Condition Subareas.

	Organizational Working Conditions
	Workload	Non-teaching	Transfers	Class sizes	Evaluation
I-CU	−0.455 (0.506)	−0.965* (0.468)	−1.398*** (0.392)	−0.350 (0.529)	−0.995* (0.475)
TPS-CU	−0.073 (0.283)	0.022 (0.262)	0.020 (0.219)	0.096 (0.296)	−0.255 (0.265)
% Urban	0.004 (0.003)	−0.002 (0.003)	0.004 (0.002)	0.006 (0.003)	−0.001 (0.003)
Enrollment (ln)	−0.136 (0.134)	0.197 (0.124)	−0.040 (0.104)	−0.043 (0.140)	−0.086 (0.126)
% Minority	−0.012 (0.008)	−0.002 (0.008)	0.004 (0.007)	−0.009 (0.009)	0.013 (0.008)
% FRL	0.001 (0.008)	0.005 (0.007)	−0.002 (0.006)	0.008 (0.008)	−0.014 (0.007)
NEA	−0.659 (0.427)	−0.085 (0.395)	−0.129 (0.331)	0.077 (0.447)	0.190 (0.401)
Joint AFT/NEA	−0.663 (0.539)	−0.033 (0.499)	−0.514 (0.417)	−0.137 (0.564)	−0.418 (0.506)
Unknown Union	0.232 (0.573)	−0.602 (0.530)	−0.287 (0.443)	−0.314 (0.599)	−0.060 (0.537)
Years Certified	0.000 (0.013)	−0.019 (0.012)	−0.008 (0.010)	−0.001 (0.014)	−0.010 (0.013)
Union Membership (ln)	−0.045 (0.082)	0.127 (0.076)	0.085 (0.064)	0.034 (0.086)	−0.087 (0.077)
Constant	2.363* (0.934)	−0.947 (0.864)	0.401 (0.723)	0.088 (0.977)	1.609 (0.877)
N	91	91	91	91	91
R-Sq	.200	.236	.526	.136	.225

Note. *.05, **.01, ***.001; standard error in parentheses.

We explore what specific provisions may determine the statistically significant differences between I-CU and TPS-U bargaining agreements’ restrictiveness in Table 5. These provisions indicate specific working conditions in which I-CU agreements and TPS-CU agreements provide more flexibility than their TPS-U counterparts. Here we take the top three most restrictive items from each contract subarea and run a series of linear probability models using Equation 2 to predict the probability that these items appear in a bargaining unit’s CBA based upon bargaining unit type and characteristics. First, we interpret the differences in the probability of the presence of specific provisions for those subareas, in which we found that the I-CU and TPS-CU agreements’ restrictiveness were lower relative to TPS-U agreements. With respect to retirement, I-CU agreements are far less likely than TPS-U agreements to include retirement at all as a subject of bargaining (−0.448, p < .05), and if they do discuss it, to specify a minimum number of years of employment required to receive these benefits (−0.583, p < .05) (Table 5, Panel E). With respect to non-teaching duties, we found no clear set of provisions that would explain the difference between I-CU and TPS-U agreements’ restrictiveness. For example, I-CU agreements have a similar probability of providing FRL, compensation for extra workload, and a large amount of preparation time relative to TPS-U agreements (Panel J). With respect to transfer provisions, I-CU agreements are less likely to provide current teachers a choice of new vacancies and make current vacancies known to current teachers compared to I-CU agreements (Panel K). For evaluation, I-CU agreements are also less likely to require pre-observations meetings for evaluation (Panel M). With respect to leave, the only area of the contract in which TPS-CU agreements were less restrictive than TPS-U agreements, the former are less likely to provide unpaid leave than TPS-U agreements (Panel H).

Table 5.

Linear Probability Models Predicting Presence of Most Restrictive Collective Bargaining Items.

	Economic Working Conditions
	Annual National Board Bonus (coa2)	$1000 National Board Bonus (coa2a)	$1000 bonus for PhD/EdD (coc1)	Medical benefits offered (hwq)	Medical benefits to spouse/partner (hwq2)	Dental benefits to spouse/partner (hwr2)
	A. Compensation			B. Health Benefits
I-CU	−0.127 (0.233)	−0.115 (0.200)	−0.127 (0.233)	−0.026 (0.106)	−0.346 (0.242)	−0.366 (0.247)
TPS-CU	0.006 (0.130)	−0.041 (0.112)	0.006 (0.130)	−0.017 (0.059)	−0.351* (0.135)	−0.279* (0.138)
	C. Economic Working Conditions (continued)			D. Political Working Conditions
	Retirement plan specified (reta)	Minimum employment for retirement specified (reta)	Minimum employment 15 years for retirement (reta1a)	Union time off provided (ara)	Union president leave provided (ara5)	Union president leave paid (ara5a)
	E. Retirement			F. Union Rights
I-CU	−0.448* (0.205)	−0.583* (0.240)	−0.172 (0.241)	0.127 (0.154)	−0.288 (0.234)	−0.410 (0.230)
TPS-CU	−0.070 (0.115)	−0.043 (0.134)	−0.084 (0.135)	−0.025 (0.086)	−0.118 (0.131)	−0.255 (0.129)
	Physical Working Conditions
	Arbitration is final level (grc1)	Arbitration decision is binding (grc2)	Board decision is not finding (gra)	Parenting leave provided (lvi)	Return rights after parenting leave (lvi8a)	1 year of unpaid leave (lvi6a)
	G. Grievances			H. Leaves
I-CU	0.234 (0.181)	0.227 (0.198)	0.123 (0.153)	0.066 (0.174)	0.244 (0.239)	−0.412 (0.223)
TPS-CU	0.032 (0.101)	0.123 (0.110)	0.010 (0.086)	0.058 (0.097)	−0.109 (0.134)	−0.375** (0.125)
	Organizational Working Conditions
	Instructional minutes specified (scc)	Moderate instructional minutes provided (scc1)	Total work days provided (sce)	Duty-free lunch provided (ntb)	Compensation for extra workload (ntj)	High amount of prep time provided (nth1a)
	I. Workload			J. Non-Teaching Duties
I-CU	−0.336 (0.247)	−0.150 (0.242)	−0.173 (0.144)	0.267 (0.181)	−0.205 (0.123)	0.106 (0.254)
TPS-CU	0.022 (0.138)	0.001 (0.135)	0.043 (0.081)	−0.022 (0.101)	0.003 (0.069)	0.040 (0.142)
	Current teachers considered first (tvo1)	Restrictions on involuntary transfer (tvj2)	Vacancies made known to current teachers (tvq)	Class size addressed (csa)	Action taken if class size exceeded (csa3)	Actions specified (csa3e)
	K. Transfers			L. Class Sizes
I-CU	−0.778*** (0.226)	−0.145 (0.213)	−0.795*** (0.204)	0.250 (0.148)	−0.009 (0.207)	0.023 (0.253)
TPS-CU	−0.151 (0.126)	0.153 (0.119)	0.031 (0.114)	−0.002 (0.083)	0.011 (0.116)	0.065 (0.141)
	Pre-observation meeting for tenured teachers (evat)	Pre-observation meeting for non-tenured teachers evap	Restrictions on information allowed in evaluation (evw)
	M. Evaluation
I-CU	−0.435* (0.205)	−0.524* (0.213)	−0.115 (0.148)
TPS-CU	−0.079 (0.115)	−0.113 (0.119)	0.071 (0.083)

Note. *.05, **.01, ***.001; standard error in parentheses.

Specific provisions that have a similar probability of appearing in I-CU agreements relative to TPS-U agreements are also worth noting. These provisions illustrate where I-CUs have secured restrictive working conditions similar to their longer established TPS-U counterparts. For example, with respect to economic working conditions, I-CU agreements are similarly likely to pay an annual board certification bonus and bonuses for PhD/EdD degrees (Panel A). Further, they are equally likely to pay medical benefits and provide these benefits to spouses/partners (Panel B). When political working conditions are examined, I-CU agreements are just as likely as TPS-U agreements to provide time off for union leaders (Panel F). I-CU agreements provide arbitration rights for employee grievances at similar levels as TPS-U contracts (Panel G) and are just as likely as TPS-U contracts to specify instructional minutes and the number of workdays (Panel I). Finally, compared to TPS-U agreements, I-CU contracts are just as likely to indicate class sizes, take action if class sizes are exceeded, and specify which actions should be taken (Panel L).

Finally, Table 6 provides the results from Equation 3 that estimate the relation between nearest-neighbor TPS-Us’ contract restrictiveness and I-CUs’ contract restrictiveness. A positive relation could indicate that I-CUs are borrowing restrictive contract provisions from their nearest TPS-U. However, we found no statistically significant relation between I-CUs’ contract restrictiveness and their nearest-neighbor TPS-Us’ restrictiveness overall or across most subareas of the CBA. We did find a positive and statistically significant association between I-CUs’ class size restrictiveness and nearest-neighbor TPS-Us’ contract restrictiveness. A one standard deviation increase in nearest-neighbor TPS-Us’ class size restrictiveness is associated with a 0.537 SD increase in I-CUs’ class size restrictiveness (p < .001).

Table 6.

OLS Regressions Predicting I-CU Bargaining Unit Agreement Restrictiveness Using TPS-U Nearest-Neighbor Restrictiveness.

	Overall	Economic Working Conditions			Political Working Conditions	Physical Working Conditions
	Overall	Compensation	Health Benefits	Retirement	Union Rights	Grievances	Leaves
TPS-U Nearest Neighbor	0.040 (0.151)	−0.151 (0.154)	0.613 (0.444)	0.266 (0.217)	0.094 (0.119)	0.029 (0.187)	0.037 (0.228)
N	28	28	28	28	28	28	28
R-Sq	.422	.640	.326	.503	.606	.492	.149
	Organizational Working Conditions
	Workload	Non-Teaching	Transfers	Class Sizes	Evaluation
TPS-U Nearest Neighbor	−0.086 (0.298)	−0.308 (0.208)	0.103 (0.213)	0.537*** (0.115)	0.061 (0.152)
N	28	28	28	28	28
R-Sq	.515	.445	.192	.638	.618

Note.*.05, **.01, ***.001; standard error in parentheses.

Taken together, we find that California charter schools in general and independent charter schools specifically tend to negotiate contracts that are measurably less restrictive than their TPS-U counterparts overall and in important subareas of the CBA, including retirement, non-teaching duties, transfers, and performance evaluations. These bargaining units have not been in force as long and are smaller in size than those of TPS-Us. In contrast, agreements between charter schools affiliated with traditional public school bargaining units do not appear to differ significantly from TPS-U agreements.

Limitations

This paper provides novel descriptive data on charter CBAs in comparison to their traditional public school district counterparts. However, we note some limitations. First, while are sample is larger than what is found in other published work on CBAs in charter schools, we still have a relatively small sample with which to investigate differences in contracts across charter and TPS unions. Although many of our estimates are imprecise as a result, we still find several ways in which I-CU and TPS-CU contracts differ from their nearest neighbor TPS-U counterparts.

Second, although a sizeable percentage of unionized charter schools in the United States are in California and these schools have been a key focus of existing research on charter school unionization (Hart & Sojourner, 2015; Matsudaira & Patterson, 2017), our study has limited ability to be generalized to contexts outside of California where charter schools have also unionized.

Finally, this study does not investigate the association between California charter school contracts’ restrictiveness and performance and thus, we cannot say whether these agreements determine some of the performance differences between unionized and non-unionized charters observed in prior work. Lastly, the data in our study cannot explain what aspects of the bargaining process shape CBAs’ restrictiveness.

Conclusion

While prior studies have explored the strength of CBAs in traditional public schools, little research has been conducted on charter schools’ CBAs. California has the largest number of unionized charter schools and is an ideal state in which to explore whether, and to what extent, CBAs in charter schools differ from their traditional public school counterparts. Our study is the first of which we are aware to conduct a systematic comparison between CBAs negotiated by charter school bargaining units (I-CUs and TPS-CUs) and their traditional public school counterparts (TPS-Us). Overall, we find that I-CU CBAs are much less restrictive than TPS-CU and TPS-U CBAs. However, I-CU contracts are comparably restrictive in certain areas of the CBA, specifically compensation, health benefits, union rights, and leave policies. We find that I-CU agreements provide more flexibility to administrators than TPS-U agreements in retirement, non-teaching duties, transfers, and evaluation. Compared to TPS contracts, TPS-CU agreements are similarly restrictive in most contract areas except for leave. We find that these differences between I-CU, TPS-CU, and TPS-U agreements exist even after bargaining unit characteristics are controlled. Further, we find little evidence that I-CU CBAs’ restrictiveness is associated with their nearest-neighbor TPS-U agreements’ restrictiveness.

Our findings provide evidence that current charter school CBAs fit both a bureaucratic and decentralized frame. As might be expected, TPS-CU agreements fit a bureaucratic frame to a greater extent, as very few provisions in TPS-CU contracts are unique to charter school teachers. In TPS-CUs, unions negotiate these agreements on behalf of both charter school and TPS teachers. Some of the similarities in CBAs in the two bargaining units may be attributable to the way that unions approach bargaining more generally. In negotiations, teachers’ unions make an effort to treat all bargaining unit members similarly, as doing otherwise undermines collective action by creating inequity among members (Johnson et al., 2009). This could lead to negotiations and CBAs that provide very similar working conditions to charter and TPS teachers, as observed.

However, I-CU CBAs appear to follow a more decentralized frame, particularly with respect to non-teaching duties, evaluations, and transfers. Core provisions that have been the hallmark of restrictive TPS-U agreements appear to be much more flexible or completely absent in I-CU agreements. Although I-CU agreements appear to diverge from TPS-U agreements in certain key domains, some provisions, particularly economic provisions, are just as restrictive in I-CU agreements as TPS-U agreements. In addition, we expected that the underlying competition for students between ICs and nearest neighbor TPSs would engender greater isomorphic pressure on IC CBAs. In short, if I-CUs are borrowing provisions from neighboring TPS-Us, then we would observe a positive relation between I-CU contracts’ restrictiveness and that of their neighbor TPS-U. However, with the exception of class size, we found little evidence that I-CU CBAs’ restrictiveness was associated with that of nearby TPS-U schools’ agreements, further suggesting that I-CUs follow a decentralized frame.

Taken together, we conclude that California charter schools’ unionization has not yet led to the widespread adoption of traditional public school rules within the charter school sector, as evidenced by the comparison of CBAs. In short, even independent charter schools with unions and CBAs retain autonomy, consistent with Huerta and Zuckerman’s (2009) decentralized frame. One explanation for this result is that, although charter school teachers unionize for many of the same reasons that traditional public school teachers do (e.g., Jabbar et al., 2020; Lavery & Jochim, 2022), charter teachers demand different union protections. In short, the differences between I-CU and TPS-U agreements may reflect charter school teachers’ different preferences. Moreover, if the differences between I-CU and TPS-U agreements’ restrictiveness arise from dissimilar teacher preferences, another explanation may be that teachers’ unions in I-CUs are not as strong as those in TPS-Us, and thus, are unable to achieve CBAs with working conditions equal to those of TPS-U CBAs.

However, as the unionized charter sector matures and expands, I-CU CBAs may still ultimately converge to the restrictiveness of TPS-U CBAs. Obtaining CBAs with working conditions similar to those of their TPS-U counterparts is one avenue for charter school teachers to signal their legitimacy and professionalization. If isomorphism and the pursuit of legitimacy are determining forces of charter school bargaining, then we may see a trend in charter school bargaining agreement restrictiveness that approximates that of TPS contracts further. In addition, I-CU unions’ increasing strength may be another force that leads to agreements that converge to the restrictiveness of those in TPS-Us. However, if the differences in restrictiveness between I-CU and TPS-U agreements arise because of fundamental differences in the bargaining unit members’ preferences, then these differences may instead be fairly durable over time. Future research should explore trends in charter bargaining agreements to determine the resilience of the differences observed in this paper over time.

Supplemental Material

sj-docx-1-epx-10.1177_08959048231178024 – Supplemental material for Collective Bargaining Agreement Restrictiveness in Unionized Charter Schools

Supplemental material, sj-docx-1-epx-10.1177_08959048231178024 for Collective Bargaining Agreement Restrictiveness in Unionized Charter Schools by Bradley D. Marianno, David S. Woo and Kate Kennedy in Educational Policy

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Bradley D. Marianno

David S. Woo

Kate Kennedy

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Bradley D. Marianno is an Associate Professor of educational policy and leadership and faculty director of the Center for Research, Evaluation, and Assessment at the University of Nevada, Las Vegas.

David S. Woo is an Assistant Professor of educational leadership and policy at the University of Utah.

Kate Kennedy is an Associate Policy Researcher at RAND Corporation.

References

Ableidinger

Hassel

B. C.

(2010). Free to lead: Autonomy in highly successful charter schools. Issue brief, National Alliance for Public Charter Schools. https://files.eric.ed.gov/fulltext/ED535213.pdf

Budde

(1988). Education by charter: Restructuring school districts. Regional Laboratory for Educational Improvement of the Northeast and Islands.

Busser

C. F.

(2017). NEA adopts charter school policy statement. National Education Association. https://www.nea.org/about-nea/press/press-releases/nea-adopts-charter-school-policy-statement-0

Camera

(2017). Teachers union adopts new, anti-charter school policy. US News & World Report. https://www.usnews.com/news/education-news/articles/2017-07-05/teachers-union-adopts-new-anti-charter-school-policy

Cannata

(2007). Teacher community in elementary charter schools. Education Policy Analysis Archives, 15(11), 1–31.

Cannata

(2011). Charter schools and the teacher job search. Journal of School Choice, 5(1), 111–133. https://doi.org/10.1080/15582159.2011.548256

Chubb

J. E.

Moe

T. M.

(1988). Politics, markets, and the organization of schools. American Political Science Review, 82(4), 1065–1087.

Cook

Lavertu

Miller

(2021). Rent-seeking through collective bargaining: Teachers unions and education production. Economics of Education Review, 85, 102193.

Cowen

J. M.

Fowles

(2013). Same contract, different day? An analysis of teacher bargaining agreements in Louisville since 1979. Teachers College Record, 115(5), 1–30.

10.

David

Hesla

(2018). Estimated public charter school enrollment, 2017-2018. National Alliance for Public Charter Schools. https://www.publiccharters.org/sites/default/files/documents/2018-03/FINAL%20Estimated%20Public%20Charter%20School%20Enrollment,%202017-18.pdf

11.

Eberts

R. W.

(1983). How unions affect management decisions: Evidence from public schools. Journal of Labor Research, 4(3), 239–247.

12.

Eberts

R. W.

Stone

J. A.

(1984). Unions and public schools: The effect of collective bargaining on American education. Lexington Books.

13.

Eberts

R. W.

Stone

J. A.

(1987). Teacher unions and the productivity of public schools. ILR Review, 40(3), 354–363. https://doi.org/10.1177/001979398704000303

14.

Freeman

R. B.

Medoff

J. L.

(1984). What do unions do. Industrial and Labor Relations Review, 38, 244.

15.

Goldhaber

Lavery

Theobald

(2014). My end of the bargain: Are there cross- district effects in teacher contract provisions? ILR Review, 67(4), 1274–1305.

16.

Goldhaber

Lavery

Theobald

D’Entremont

Fang

(2013). Teacher collective bargaining: Assessing the internal validity of partial independence item response measures of contract restrictiveness. Sage Open, 3(2).

17.

Hart

C. M. D.

Sojourner

A. J.

(2015). Unionization and productivity: Evidence from charter schools. Industrial Relations A Journal of Economy and Society, 54(3), 422–448. https://doi.org/10.1111/irel.12097

18.

Henig

Holyoke

Brown

Lacireno-Paquet

(2005). The influence of founder type on charter school structures and operations. American Journal of Education, 111(4), 487–588. https://doi.org/10.1086/431181

19.

Hess

Loup

(2008). The leadership limbo: Teacher labor agreements in America’s fifty largest school districts. Thomas B. Fordham Institute. http://eric.ed.gov/?id=ED500316

20.

Hess

F. M.

Kelly

A. P.

(2006). Scapegoat, albatross, or what? The status quo in teacher collective bargaining. In Hannaway

Rotherham

(Eds.), Collective bargaining in education: Negotiating change in today’s schools (pp. 53–88). Harvard Education Press.

21.

Huerta

L. A.

Zuckerman

(2009). An institutional theory analysis of charter schools: Addressing institutional challenges to scale. Peabody Journal of Education, 84(3), 414–431.

22.

Hussar

Zhang

Hein

Wang

Roberts

Cui

Dilig

(2020). The condition of Education 2020. NCES 2020-144. National Center for Education Statistics. https://files.eric.ed.gov/fulltext/ED605216.pdf

23.

Jabbar

(2018). Recruiting “talent”: School choice and teacher hiring in New Orleans. Educational Administration Quarterly, 54(1), 115–151.

24.

Jabbar

Chanin

Haynes

Slaughter

(2020). Teacher power and the politics of union organizing in the charter sector. Educational Policy, 34(1), 211–238.

25.

Jochim

Lavery

(2021). Worth the bargain? Collective bargaining agreements in unionized charter schools. Journal of School Choice, 15(2), 248–267.

26.

Johnson

S. M.

(1990). Teachers at work: Achieving success in our schools. Basic Books.

27.

Johnson

S. M.

Donaldson

M. L.

Munger

M. S.

Papay

J. P.

Qazilbash

E. K.

(2009). Leading the local: Teachers union presidents chart their own course. Peabody Journal of Education, 84(3), 374–393.

28.

Johnson

S. M.

Kardos

S. M.

(2000). Reform bargaining and its promise for school improvement. In Loveless

(Ed.), Conflicting missions? Teachers unions and educational reform (pp. 7–46). Brookings Institution Press.

29.

Kahlenberg

R. D.

(2007). Tough liberal: Albert Shanker and the battles over schools, unions, race, and democracy. Columbia University Press.

30.

Kahlenberg

R. D.

Potter

(2015). Restoring Shanker’s vision for charter schools. American Education, 38(4), 4.

31.

Lavery

Jochim

(2022). Why charter teachers unionize. Educational Policy, 37(5), 1217–1240. https://doi.org/10.1177/08959048221087215

32.

Lubienski

(2003). Innovation in education markets: Theory and evidence on the impact of competition and choice in charter schools. American Educational Research Journal, 40(2), 395–443.

33.

Lubienski

Lee

(2016). Competitive incentives and the education market: How charter schools define themselves in metropolitan Detroit. Peabody Journal of Education, 91(1), 64–80. https://doi.org/10.1080/0161956x.2016.1119582

34.

Malloy

C. L.

Wohlstetter

(2003). Working conditions in charter schools: What’s the appeal for teachers? Education and Urban Society, 35(2), 219–241.

35.

Marianno

B. D.

(2021). A negotiated disadvantage? California collective bargaining agreements and achievement gaps. Educational Researcher, 50(7), 451–462.

36.

Marianno

B. D.

Bruno

Strunk

K. O.

(2021). The effect of teachers’ union contracts on school district efficiency: Longitudinal evidence from California. Sage Open, 11(1), 2158244020988684.

37.

Marianno

B. D.

Kilbride

Theobald

Strunk

K. O.

Cowen

J. M.

Goldhaber

(2018). Cut from the same cloth? Comparing urban district CBAs within states and across the United States. Educational Policy, 32(2), 334–359.

38.

Marianno

B. D.

Strunk

K. O.

(2018). The bad end of the bargain? Revisiting the relationship between collective bargaining agreements and student achievement. Economics of Education Review, 65, 93–106.

39.

Matsudaira

J. D.

Patterson

R. W.

(2017). Teachers’ unions and school performance: Evidence from California charter schools. Economics of Education Review, 61, 35–50.

40.

McDonnell

L. M.

Pascal

(1988). Teacher unions and educational reform. Publications Department, The Rand Corp.

41.

Meyer

J. W.

Rowan

(1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363.

42.

Meyer

J. W.

Rowan

(1978). The structure of educational organizations. In Meyer

M. W.

Scott

W. R.

Rowan

Deal

T. E.

(Eds.), Environments and organizations (pp. 78–109). Jossey-Bass.

43.

Murphy

(1992). Blackboard unions: The AFT and the NEA, 1900-1980. Cornell University Press.

44.

National Alliance for Public Charter Schools. (2020). Unionized charter schools: Data from 2018-19. https://www.publiccharters.org/sites/default/files/documents/2021-01/napcs_unionization_rd3.pdf

45.

National Alliance for Public Charter Schools. (2022). Automatic Collective Bargaining Exemption. https://www.publiccharters.org/our-work/charter-law-database/components/14

46.

National Center for Education Statistics. (2022). Public charter school enrollment. https://nces.ed.gov/programs/coe/indicator/cgb/public-charter-enrollment

47.

(2012). Teacher working conditions in charter schools and traditional public schools: A comparative study. Teachers College Record, 114(3), 1–26.

48.

(2017). Teacher working conditions, teacher commitment, and charter schools. Teachers College Record, 119(6), 1–38.

49.

Peterson

P. E.

(2010). No, Al Shanker did not invent the charter school. Education Next. https://www.educationnext.org/no-al-shanker-did-not-invent-the-charter-school/

50.

Price

(2011). Are charter school unions worth the bargain?Center on Reinventing Public Education. https://files.eric.ed.gov/fulltext/ED532632.pdf

51.

Renzulli

L. A.

Barr

A. B.

Paino

(2015). Innovative education? A test of specialist mimicry or generalist assimilation in trends in charter school specialization over time. Sociology of Education, 88(1), 83–102. https://doi.org/10.1177/0038040714561866

52.

Strunk

K. O.

(2011). Are teachers’ unions really to blame? Collective bargaining agreements and their relationships with district resource allocation and student performance in California. Education Finance and Policy, 6(3), 354–398.

53.

Strunk

K. O.

(2012). Policy poison or promise: Exploring the dual nature of California school district collective bargaining agreements. Educational Administration Quarterly, 48(3), 506–547.

54.

Strunk

K. O.

Cowen

Goldhaber

Marianno

B. D.

Theobald

Kilbride

(2022). Public school teacher contracts and state-level reforms: Assessing changes to collective bargaining restrictiveness across three states. American Educational Research Journal, 59(3), 538–573.

55.

Strunk

K. O.

Cowen

J. M.

Goldhaber

Marianno

B. D.

Kilbride

Theobald

(2018). It is in the contract: How the policies set in teachers’ unions’ collective bargaining agreements vary across states and districts. Educational Policy, 32(2), 280–312.

56.

Strunk

K. O.

McEachin

(2011). Accountability under constraint: The relationship between collective bargaining agreements and California schools’ and districts’ performance under No Child Left behind. American Educational Research Journal, 48(4), 871–903.

57.

Strunk

K. O.

Reardon

S. F.

(2010). Measuring the strength of teachers’ unions: An empirical application of the partial independence item response approach. Journal of Educational and Behavioral Statistics, 35(6), 629–670.

58.

Stuit

D. A.

Smith

T. M.

(2012). Explaining the gap in charter and traditional public school teacher turnover rates. Economics of Education Review, 31(2), 268–279.

59.

Torres

A. C.

(2014). Are we architects or construction workers?” Re-examining teacher autonomy and turnover in charter schools. Education Policy Analysis Archives, 22(124), 124.

60.

Urban

W. J.

(2004). Teacher politics. In Henderson

R. D.

Urban

Wolman

(Eds.), Teacher unions and education policy: Retrenchment of reform (pp. 103–123). Emerald Group Publishing Limited.

61.

Vergari

(2007). The politics of charter schools. Educational Policy, 21(1), 15–39.

62.

White

Snydman

(2020). How many charter schools and students are there?National Alliance for Public Charter Schools. https://data.publiccharters.org/digest/charter-school-data-digest/how-many-charter-schools-and-students-are-there/

63.

Wiley

(2019). CTA spent $1 million a month to restrict charter growth. Sacramento Bee. https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwiqocf6mtT8AhVeLUQIHdebC_AQFnoECAcQAQ&url=https%3A%2F%2Fwww.sacbee.com%2Fnews%2Fpolitics-government%2Fcapitol-alert%2Farticle233393557.html&usg=AOvVaw3hAa6CzgORG5dmy5dMyE1G

64.

Wohlstetter

Smith

Farrell

C. C.

(2013). Choices and challenges: Charter school performance in perspective. Harvard Education Press.

65.

Wohlstetter

Wenning

Briggs

K. L.

(1995). Charter schools in the United States: The question of autonomy. Educational Policy, 9(4), 331–358.

66.

Woo

D. S.

Marianno

B. D.

(2021). After the vote: Unionization’s effect on Chicago’s charter schools. In D’Amico

(Ed.), Walkout: Teacher militancy, activism, and school reform (pp. 175–198). Information Age Publishing, Inc.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB