Teachers Unions -- Part 1

Their influence on America’s public schools may surprise you.

I started the previous essay series with this observation about unions in general:

What is a labor union?  A labor union is a legal entity that is granted by law the exclusive right to bargain with the employer of the union’s members for worker compensation of other benefits.  Because no one besides the “exclusive bargaining unit” of an existing union can legally negotiate with employers, union officials have a special legal duty to focus exclusively on benefits to their own members, to the exclusion of all the other concerns of everyone else in society.

Indeed, as I’ve also written previously, teachers unions, for example, are legally prohibited from “putting children first”:

The Labor-Management Reporting and Disclosure Act … makes clear in section 501(a) that “The officers, agents, shop stewards, and other representatives of a labor organization occupy positions of trust in relation to such organization and its members as a group.  It is, therefore, the duty of each such person, taking into account the special problems and functions of a labor organization, to hold its money and property solely for the benefit of the organization and its members …”  Such union officials also have the duty “to refrain from dealing with such organization [the union] as an adverse party or in behalf of an adverse party in any matter connected with his duties.” Each of these obligations is considered a “fiduciary duty,” and if any senior union official violated them, they could be sued.  What this means is that if a teachers union official ever actually did put the interests of children and children’s education before the union members’ interests in better wages and work benefits, that teachers union official could be sued, and be legally stripped of their position as a result.

So unions are required by law to focus solely on the best interests of their members, to the exclusion of the interests of all others.  When it comes to teachers unions and their role in the American public education system, this fundamental fact has vast yet underappreciated effects and consequences.  Those effects and consequences are the subject of this series of essays.

This series of essays will draw on several excellent books that discuss the history and influence of teachers unions in America: Terry Moe’s Special Interest: Teachers Unions and America’s Public Schools; Mark Zupan’s Inside Job: How Government Insiders Subvert the Public Interest; Michael Hartney’s How Policies Make Interest Groups: Governments, Unions, and American Education; and Philip Howard’s Not Accountable: Rethinking the Constitutionality of Public Employee Unions.

Terry Moe makes the point that “I am not saying—and do not think—that the teachers unions are solely responsible for the nation's education problems. I am saying that the teachers unions are at the heart of these problems and, therefore, that the unions themselves and the various roles they play in collective bargaining and politics need to be much better studied and understood.”

As Moe explains:

On the surface, it might seem that the teachers unions would play a limited role in public education: fighting for better pay and working conditions for their members, but otherwise having little impact on the structure and performance of the public schools more generally. Yet nothing could be further from the truth. The teachers unions have more influence on the public schools than any other group in American society. Their influence takes two forms. They shape the schools from the bottom up, through collective bargaining activities so broad in scope that virtually every aspect of school organization bears the distinctive imprint of union design. They also shape the schools from the top down, through political activities that give them unrivaled influence over the laws and regulations imposed on public education by government, and that allow them to block or weaken governmental reforms they find threatening. In combining bottom-up and top-down influence, and in combining them as potently as they do, the teachers unions are unique among all actors in the educational arena. It is difficult to overstate how extensive a role they play in making America's schools what they are—and in preventing them from being something different … Like all special interest groups, they try to put the best face on their activities. They say that what is good for teachers is good for kids. And as a matter of public relations, they need to say that. But the simple fact—and it is indeed just a straightforward fact—is that they are not in the business of representing the interests of children. They are unions. They represent the job-related interests of their members, and these interests are simply not the same as the interests of children.

As explored in the previous essay series on public sector unions, union leaders are particularly interested in supporting their members by insisting on rules that make their members less accountable, and consequently more secure, in their jobs.  That legally-required goal of unions (to oppose accountability among their members) isn’t likely to resonate with the public, and so teachers unions typically couch their demands in ways that make it appear their ultimate goal is to improve the well-being of children.

Moe writes of survey questions that asked teachers a general question on accountability: “Do you think individual teachers should be held accountable for how much their own students learn during the school year?”  As More writes, “56 percent of all teachers expressed opposition to even this conceptual version of accountability. Within unions, opposition is still greater, at 60 percent … Outside the unions, things are somewhat different: some 52 percent of teachers said they support accountability, with even Democrats slightly favoring it—suggesting, again, that union framing may be playing a role here.”

One can debate to what extent standardized tests measure any particular teacher’s performance as a teacher, but it’s hard to argue that standardized tests don’t test the knowledge of a student.  As Moe writes:

Test scores are concrete, objective information on how much students are learning; they are the best information social science has been able to provide; and the American public certainly believes that they are good measures that need to be systematically employed. For unions, however, they are a threat—because low scores provide objective grounds for removing mediocre teachers from the classroom, and job security is the unions' number one concern. It is clearly high on the list for teachers as well: low scores would (or could) threaten their jobs and bring pressure on them to perform. On this issue, teachers are united: test scores are bad measures of student learning, and they should not be used to hold teachers accountable. Among all teachers, 78 percent reject the use of standardized tests as measures of student learning … The ideas behind school accountability have obvious merit. If the school system is to promote academic excellence, it must have clear standards defining what students need to know. It must engage in testing to measure how well the standards are being met. And it must hold students, teachers, and administrators accountable for results—and give them incentives to do their very best—by attaching consequences to outcomes. Writ large, these are simply the principles of effective management that business leaders live by every day: setting goals, measuring performance, attaching consequences, and creating strong incentives.

As Moe asks, “Without that measure [of student knowledge through test scores], what is left of accountability? Not much. And that, really, is the point.”

Moe moves next to the issue of “pay for performance”:

Consider another example: pay for performance. The notion here is that teachers should be paid, at least in part, on the basis of their classroom performance. On the surface, teachers seem to favor pay for performance at a conceptual level, perhaps because (like accountability and many other issues) it has a motherhood appeal in the abstract. In the Public Agenda survey, for instance, teachers were asked whether they would support giving “financial incentives to … teachers who consistently receive outstanding evaluations from their principals,” and they said that they favor it by two to one. These same teachers, however, also agreed by two to one that principals would “play favorites” rather than rewarding those who are truly meritorious; and they agreed by two to one that merit pay would lead to “unhealthy competition and jealousy among teachers” rather than better incentives. They also said, by wide margins, that they do not want teachers rewarded based on how their kids score on standardized tests. What’s wrong with this picture? They “support” pay for performance, but they don’t trust principals to evaluate merit, they don’t want test scores to be used, and they fear the internal divisions that would result among members. How, then, is their performance to be evaluated? These ambiguities arise for a simple reason: there is no clear connection between pay for performance and the occupational interests of each individual teacher … [T]he unions have long taken stands against pay for performance … The point of this reform [pay for performance] was to hold schools—and thus teachers—accountable for seeing to it that children learn what they are supposed to learn. And this they did not want. Historically, teachers had been granted substantial autonomy, and their pay and jobs had been almost totally secure regardless of what they did in their classrooms and regardless of how much their students learned.

As Moe explains:

The science of student testing is by far the best developed, most sophisticated component of the academic field of education. In general, teachers unions disparage standardized tests as inadequate measures of student performance and call instead for broader criteria—course grades, portfolios of student work, graduation rates, parental involvement, and more—that would make assessments far more flexible, complicated, and subjective, and much less dependent on hard, specific measures of how much students are actually learning. From the standpoint of union interests, the problem with standardized tests is that they don't just measure the performance of students. They also provide concrete evidence on the performance of schools—and teachers—and threaten to generate all sorts of problems that the unions are keen to avoid. If the scores show that kids aren't learning, the publicity will inevitably give rise to public complaints and pressures to improve, and the accountability system may require consequences of various kinds. Test scores are especially threatening, moreover, because any system that puts them to rigorous use would quickly reveal (after appropriate controls for student background characteristics and the like) that some teachers are much better than others and some are very bad. Indeed, that is precisely what the research literature does reveal. Were such information routinely collected and readily available, it would be much more difficult for policymakers to embrace the myth that somehow all teachers are the same, that they all have a right to be in the classroom, and that they should all be paid equally. There would be objective grounds for removing bad teachers from classrooms. There would be objective grounds for giving better teachers higher pay. Accountability would begin to have real teeth.

Moe provides a telling example from New York City’s recent history with attempting to impose more teacher accountability:

The New York City example speaks volumes. Recall that Mayor Michael Bloomberg and Chancellor Joel Klein, both strong proponents of accountability, attempted to improve teacher quality by reforming the tenure process. Their idea was to use student test scores as one measure of teacher performance in the classroom and then to bring this information to bear—along with much other relevant information—when new teachers were being evaluated for tenure. The idea that data on student learning should be relevant to teacher tenure seems woefully obvious. Who could disagree? The United Federation of Teachers could, and did. The union took its case to the state legislature and succeeded in getting a new law passed that prohibited any district in the state from using student test scores in the tenure evaluations of teachers. The information was available on how much students were actually learning in their classes. But in New York, the teachers unions—with a little help from their allies in the state legislature—made it illegal to take the information into account … [Teachers unions] have pressured policymakers not to authorize teacher identifiers that can be linked to student identifiers, arguing that teacher and student data sets need to be kept entirely separate. Until Arne Duncan's Race to the Top intervened to change the landscape, only eighteen states even had systems that were capable of linking the student data to teacher data.

The experience in California provides another example:

Consider what happened in California. There, the teachers unions defeated a 2005 bill that would simply have added teacher identifiers to the state data system—a unique number for each teacher, allowing for the collection of data over time. The California Federation of Teachers testified about what it called the “hidden danger in creating a teacher identifier”—namely, that “it could easily be linked to student databases and thus student performance.” How much clearer could the conflict of interest be? To the state, school districts, and citizens at-large, this database offered a wealth of information for the management and improvement of the public schools. To the teachers union, it was a “danger.” In 2006 a teacher identifier provision did pass the state legislature. But the unions were able to win language in the new law that sharply limited how the data could be used. Here is what the law said: “Data in the system may not be used…for the purposes of pay, promotion, sanction, or personnel evaluation of an individual teacher or groups of teachers, or of any other employment decisions related to individual teachers … In 2010 reporters from the Los Angeles Times obtained local test score data from the Los Angeles school district under the state's Public Records Act: data the district had collected but never revealed, analyzed, or put to any educational purpose. With expert help, the reporters created seven years' worth of value-added scores to arrive at measures of classroom effectiveness for each of some 6,000 school district elementary teachers. They then—amid great controversy, not just locally but nationwide, with the unions fervently protesting—published the ratings of individual teachers, by name, in the newspaper, along with a searchable database that ultimately got many hundreds of thousands of hits from parents and others who wanted to know the effectiveness of local teachers. For the first time in any district, anywhere, the public was being provided with information about the effectiveness of local teachers and being shown what researchers have long known: that some teachers are consistently much better than others, some are consistently much worse, and many children are being short-changed … [T]his event needs to be understood in context: information on teacher effectiveness has long been suppressed, and the unions have played a big role in that suppression. The Los Angeles Times was attempting to do something about that by taking action on its own. It published its ratings, it said, “because they bear on the performance of public employees who provide an important public service, and in the belief that parents and the public have a right to the information.” Had it not taken action, “this information might have remained uselessly locked away.” The Los Angeles Times also explicitly recognized the role of United Teachers Los Angeles (UTLA) in keeping the data under cover. “The UTLA's position,” it argued, “is understandable; it exists to protect teachers, including the bad ones. Fortunately for parents and the public, a newspaper exists to give them information that would otherwise be withheld.” … Secretary of Education Arne Duncan, pressed by the media to comment on the Los Angeles Times articles, said, “In other fields, we talk about success constantly, with statistics and other measures to prove it. Why, in education, are we scared to talk about what success look like? What is there to hide?”

Moe then points out the convenience of what teachers unions do support regarding the consequences for poor school performance:

The unions are actually fine with having consequences for poor performance—as long as they are entirely positive consequences. Low-performing schools should get more money, more assistance, more support. As the unions argue it, these are the keys to turning such schools around. This logic of positivity is fundamental to the unions' “support” for accountability and pervades the way they talk about how it should operate in turning around problem schools. Here, for instance, is a typical statement of official NEA [the National Education Association union] policy: “The paradigm must change from labeling and punishing to investing in proven programs and interventions … Hard-to-staff schools, especially those … that have consistently struggled to meet student achievement targets, need significant supports and resources, including additional targeted funding to attract and retain quality teachers, and induction programs with intensive mentoring components that will help teachers become successful.” Here and elsewhere, union positions are couched in terms of school improvement and doing what works, but the focus is exclusively on positive inducements and thus, ultimately, on spending more money—and spending it on the existing staff, all of whose jobs are totally safe. This is what “good” accountability comes down to, in the final analysis, and this is the kind of accountability they support. What they oppose is true accountability—which would, among other things, identify poorly performing teachers and get them out of the classroom and off the payroll.

And of course:

The teachers unions are not the only interest groups that fight against school accountability. They have allies among (some) school administrators, who see accountability—accurately—as a threat to their traditional autonomy and a source of considerable new performance pressures.

In the next essay in this series, we’ll examine the nature and effect of “collective bargaining agreements” that embody the rules teachers unions impose on local school systems.