I am delighted that Perspectives on Politics has decided to devote a symposium to Terry Moe's book. The current debate on the state of American education reform has turned its attention to the role of teachers' unions.Footnote 1 Given Moe's trajectory on the topic, Special Interest is a timely contribution to the debate. Moe makes a frontal critique of teachers' unions based on a comprehensive analysis of these organizations and an original survey on teacher public opinion that illuminates the preferences of a crucial actor in the debate over education reform.
Moe argues that American teachers' unions—the American Federation of Teachers (AFT) and the National Education Association (NEA)—resist education reforms that facilitate school choice (e.g., vouchers, charter schools), link teacher pay to student performance on test scores, strengthen principal managerial discretion at the expenses of seniority and tenure, and so on. The cause of teachers' union resistance, he argues, is that union leaders are representative of the median teacher, and if they were to support these reforms, they would be voted out of office—and, indeed, he provides anecdotal evidence supporting such a prediction. The ability of teachers' unions to block these reforms, which would modify the conditions in their collective bargaining contracts, is enhanced by American political institutions, where multiple veto points empower organized interests with a high stake in the process, at a minimum, to defend the status quo.
As a student of comparative politics who has analyzed the reaction of Latin American teachers' unions to the type of reforms proposed by Moe, I do not find his main argument surprising. The incentive structure of teachers' unions should lead us to expect such an outcome. However, while Moe argues that his study shows that the main problem with American education is collective bargaining, I conclude from his study that the American experience suggests a) the need to focus on the incentives of teachers rather than those of teachers' unions to make reforms work, and b) the need to take advantage of the incrementalism and diversity of reform experiences created by federalism and democracy to generate systematic evaluations of the effect produced by both reforms and collective bargaining on childrens' performance in order to learn how to improve American schools.
Union Democracy
Since the classic studies of Robert Michels and Seymour Martin Lipset, Martin Trow, and James S. Coleman, union democracy has been considered a challenge given the structure of labor organizations.Footnote 2 However, Moe shows that both the AFT and the NEA are representative of their members. Teachers' unions are bad for education, he argues, because they are representative of teachers: Teachers' unions “represented the job-related interests of the members, and these interests are simply not the same as the interests of children” (p. 7). His evidence comes from a national survey of 3,000 American teachers, which provides the richest source of information on teacher preferences I have ever seen. The bottom line is that according to Moe, teachers' unions are representative not only of their median member but also of the median teacher.
Moe's data shows that American teachers have surprisingly homogenous views, even when controlling for union membership, partisanship, and their location in districts with limits on teachers' collective bargaining. An overwhelming majority of teachers join unions voluntarily and think that collective bargaining produces reasonable rules that help promote learning (pp. 89, 93). A majority of teacher respondents are afraid of principals' abuse of their discretion if such rules did not exist (p. 81), while they support the right to strike, teacher tenure, and higher taxes to pay for education. Teachers, as Moe argues, do not like the main tools he proposes for reforming the quality of American education. Over 75% of surveyed teachers oppose vouchers and tests to measure achievement, whereas slim majorities reject charter schools (58%) and “holding teachers accountable” (56%) (p. 101). Hence, the author persuasively shows that American teachers do not like the prescribed education reforms, which are therefore resisted by teachers' unions who follow their members' mandate.
The implication of Moe's public opinion analysis seems to be that the problem for American education lies less with teachers' unions, as he argues, than with teachers themselves. His analysis of public opinion and his description of the structure of professional rewards suggest that teachers have self-selected into this career for its working conditions, rather than for vocational reasons associated with student learning. Even though the rate of attrition is high (40% leave the profession in the first three years), Moe's analysis seems to establish that teachers are not mainly driven by pedagogical vocation. Whereas a career choice that follows work conditions and rewards would be taken for granted in many other occupations, we seem to be holding teachers to a higher standard.
The incentives of teachers rather than teachers' unions are, therefore, crucial for understanding the outcome of education reform, not only because teachers' unions are representative of their members but also because teachers cannot be continuously monitored in the classroom, while ex post rewards and punishments would not compensate for the damage done to children's education by the time their shirking is caught by some performance measure. Teacher cooperation is crucial to the successful implementation of any type of education reform that would improve children's performance, and the incentive structure of the profession should be a crucial piece of school reform.
Democratic Status Quo Bias
Moe laments that teachers' unions take advantage of the status quo bias generated by the multiple veto points of American democracy to slow down the pace of education reform. Here, a comparative perspective might be useful. In the 1970s, a pro–school choice education reform (including decentralization of the municipal levels and vouchers to select either private or public schools) was swiftly adopted by the authoritarian government of General Augusto Pinochet in Chile. Three decades later, student upheaval and political mobilization against the inequality generated by those reforms has destroyed the popularity of center-right president Sebastian Piñeira—his approval rate declined from 44% to 26% in a year, according to the Centro de Estudios Publicos (CEP).
Democratic status quo bias hinders radical reforms, but also prevents mistakes and facilitates learning by trial and error (it also makes possible freedom of association and social and political pluralism). Hence, the incremental and decentralized nature of American education reforms provides an opportunity to assess the impact of reform tools before they travel beyond districts that led the school reform process due to the poor quality of their schools. It is pretty clear that the infamous New York rubber rooms (and their replacement by the Absent Teacher Reserve) or the accumulation of teacher absenteeism on Fridays and Mondays have deleterious effects on education. However, there is no clear academic consensus on the impact of many specific measures advocated by Moe for solving the problems of student performance.Footnote 3
The importance of systematic analysis is highlighted by Moe even with reference to the educational impact of teachers' unions. He acknowledges that there are only two quantitative articles in peer-refereed journals analyzing the impact of collective bargaining on education performance across school districts: his own and one by Caroline Hoxby.Footnote 4 Moreover, he adds, state-level comparisons that show a weaker performance in states where teachers' collective bargaining is not allowed are biased due to “endogeneity” problems and the difficulty of controlling for confounding factors: “[S]ome of the factors that explain why unions are weak in these jurisdictions may also explain why the school systems do not perform well; their political cultures, for instance, tend to be characterized by attitudes and laws hostile to unions, but also by low spending on public goods (like schools), low taxing, and less educated populations” (p. 211). Moreover, in discussing other countries with excellent education records and high levels of teacher unionization (for instance, 95% of teachers are unionized in high-performing Finland), Moe decries the lack of data to analyze the effect of collective bargaining as opposed to other factors shaping educational outcomes in those cases (pp. 208–9).Footnote 5
Hence, the delay in a widespread adoption of school reforms can provide an opportunity not only to use the now-available data on school and teacher performance to assess the effect of the different tools generated by reforms, but also to systematically study the effect of collective bargaining on education, given the multiplicity of school districts in the United States and the possibility of cross-national comparisons—thereby engaging the comparative literature on the topic that is not addressed by Moe. Systematic analysis is crucial for policy design, and the democratic process generates not only veto players but also the need to justify to constituencies the effectiveness of adopted policies, especially when voters cannot perceive their consequences immediately.
Teachers' Unions and Education Reform
In both Latin America and the United States, teachers' unions have resisted the sort of reforms proposed by Moe. In Latin America, the resistance of teachers' unions to reforms is not only derived from their members' preferences against these policies. Union leaders are also concerned about the impact of these reforms on increasing the heterogeneity of working conditions, which would make it harder for them to represent their members. Moe attacks collective bargaining, which is the reason for the existence of any labor union. Teachers' unions, in the United States and Latin America, are afraid that ending the rigidity of work conditions can be aimed not only at improving education performance but also at weakening the unions as organizations. Similarly, Latin America teachers' unions typically associate school choice measures, such as vouchers, with the promotion of religious or for-profit education, rather than with the generation of competitive incentives to improve performance.
Reformers, therefore, are more likely to gain some degree of union cooperation when they are trusted to seek these policies for their stated goal of improving education, rather than as part of an ideological attack on unions. Trusted reformers are more likely to persuade teachers' unions to accept some of these reforms and “sell” them to their members. This “Nixon-in-China” effect may be at play to some degree in the current wave of reforms that Secretary of Education Arne Duncan has promoted during the Obama administration, which requires teachers' union “buy-in”—against Moe's advice. Similarly, if teachers' unions are as representative of their members as Moe argues, and teacher cooperation is crucial for reform implementation on a daily basis, these conditions may explain why some reform advocates, such as the Bill and Melinda Gates Foundation, also seek to gain the buy-in of teachers' unions in the reform process (p. 311). Signals that reforms are motivated by the stated goal of improving the educational opportunities of children, rather than by an ideological hostility toward representative unions, can help not only to obtain buy-in from teachers' unions but, more importantly, to win the trust and cooperation of teachers in the classroom.
