Unions and the Public Interest

Is collective bargaining for teachers good for students?

Three years after Barack Obama’s election signaled a seeming resurgence for America’s unions, the landscape looks very different. Republican governors in Wisconsin, Indiana, and Ohio have limited the reach of collective bargaining for public employees. The moves, especially in Wisconsin, set off a national furor that has all but obscured the underlying debate as it relates to schooling: Should public-employee collective bargaining be reined in or expanded in education? Is the public interest served by public-sector collective bargaining? If so, how and in what ways? Arguing in this forum for more expansive collective bargaining for teachers is Richard D. Kahlenberg, senior fellow at The Century Foundation and author of Tough Liberal: Albert Shanker and the Battles over Schools, Unions, Race and Democracy. Responding that public-employee collective bargaining is destructive to schooling and needs to be reined in is Jay P. Greene, chair of the Department of Education Reform at the University of Arkansas and author of Education Myths.


Richard D. Kahlenberg: Wisconsin governor Scott Walker’s campaign earlier this year to significantly curtail the scope of bargaining for the state’s public employees, including teachers, set off a national debate over whether their long-established right to collectively bargain should be reined in, or even eliminated.

If you’re a Republican who wants to win elections, going after teachers unions makes parochial sense. According to Terry Moe, the National Education Association (NEA) and the American Federation of Teachers (AFT) gave 95 percent of contributions to Democrats in federal elections between 1989 and 2010. “Collective bargaining is the bedrock of union well-being,” Moe notes, so to constrain collective bargaining is to weaken union power. The partisan nature of Walker’s campaign was revealed when he exempted two public-employee unions that supported him politically: those representing police and firefighters.

But polls suggest that Americans don’t want to see teachers and other public employees stripped of collective bargaining rights. A USA Today/Gallup poll found that by a margin of 61 to 33 percent, Americans oppose ending collective bargaining for public employees. A Wall Street Journal/NBC poll discovered that while Americans want public employees to pay more for retirement benefits and health care, 77 percent said unionized state and municipal employees should have the same rights as union members who work in the private sector. Is the public wrong in supporting the rights of teachers and other public employees to collectively bargain? I don’t think so.

Richard D. Kahlenberg

The NEA has existed since 1857 and the AFT since 1916, but teachers didn’t have real influence until they began bargaining collectively in the 1960s. Before that, as Albert Shanker, one of the founding fathers of modern teachers unions, noted, teachers engaged in “collective begging.” Educators were very poorly compensated; in New York City, they were paid less than those washing cars for a living. Teachers were subject to the whims of often autocratic principals and could be fired for joining a union.

Some teachers objected to the idea of collective bargaining. They saw unions as organizations for blue-collar workers, not for college-educated professionals. But Shanker and others insisted that teachers needed collective bargaining in order to be compensated sufficiently and treated as professionals.

Democratic societies throughout the world recognize the basic right of employees to band together to pursue their interests and secure a decent standard of living. Article 23 of the 1948 Universal Declaration of Human Rights provides not only that workers should be shielded from discrimination, but also that “everyone has the right to form and to join trade unions for the protection of his interests.”

Collective bargaining is important, not only to advance individual interests but to give unions the power to serve as a countervailing force against big business and big government. Citing the struggle of Polish workers against the Communist regime, Ronald Reagan declared in a Labor Day speech in 1980, “where free unions and collective bargaining are forbidden, freedom is lost.”

The majority of Americans believe that citizens don’t give up the basic right to collective bargaining just because they work for the government. In free societies across the globe, from Finland to Japan, public school teachers have the right to form unions and engage in collective bargaining.

In the United States, only seven states outlaw collective bargaining for teachers. Thirty-four states and the District of Columbia authorize collective bargaining for such employees, and another nine permit it. It is no accident that the seven states that prohibit collective bargaining for teachers are mostly in the Deep South, the region of the country historically most hostile to extending democratic citizenship to all Americans.

Terry Moe finds that collective bargaining for teachers has strong support among candidates for school boards. He writes, “the vast majority of school board candidates, 66 percent, have positive overall attitudes toward collective bargaining. Even among Republicans—indeed, even among Republicans who are not endorsed by the unions—the majority take a positive approach to this most crucial of union concerns.”

Nonetheless, some (including Moe) would prefer that collective bargaining for teachers be severely curtailed, or even outlawed. Ironically, one argument advanced by critics is that collective bargaining is undemocratic. The other major argument is that teacher collective bargaining is bad for education. Both claims are without basis.

Those who argue that collective bargaining for teachers is stacked, even undemocratic, say that, unlike in the private sector, where management and labor go head-to-head with clearly distinct interests, in the case of teachers, powerful unions are actively involved in electing school board members, essentially helping to pick the management team. Moreover, when collective bargaining covers education policy areas, such as class size or discipline codes, the public is shut out of the negotiations, some assert. Along the way, they conclude, the interests of adults in the system are served but not the interests of children.

But these arguments fail to recognize that in a democracy, school boards are ultimately accountable to all voters, not just teachers, who often live and vote outside the district in which they teach, and in any event represent a small share of total voters. Union endorsements matter in school board elections, but so do the interests of general taxpayers and parents and everyone else who makes up the community. If school board members toe a teachers union line that is unpopular with voters, those officials can be thrown out in the next election.

Indeed, one could make a strong argument that any outsized influence that teachers unions exercise in school board elections provides a nice enhancement of democratic decisionmaking on education policy because teachers, as much as any other group in society, can serve as powerful advocates for those Americans who cannot vote: schoolchildren. The interests of teachers and their unions don’t always coincide with those of students, but on the really big issues, such as overall investment in education, the convergence of interests is strong. Certainly, the interests of teachers in ensuring adequate educational investment are far stronger than they are for most voters, who don’t have children in the school system and may be more concerned about holding down taxes than investing in the education of other people’s kids.

American society consistently underinvests in children compared with other leading democratic societies. According to the Organisation for Economic Co-operation and Development (OECD), the child poverty rate in the United States is 21.6 percent, the fifth-highest among its 40 member nations. Only Turkey, Romania, Mexico, and Israel have higher child-poverty rates. Put differently, we’re in the bottom one-eighth in preventing child poverty. By contrast, when the interests of children are connected with the interests of teachers, as they are on the question of public education spending, the U.S. ranks close to the top one-third. Among 39 OECD nations, the U.S. ranks 14th in spending on primary and secondary education as a percentage of gross domestic product.

Some critics argue that strong teachers unions make for inefficient spending and bad education policies in the instances when teacher and student interests diverge. For example, it is frequently claimed that teachers unions, through collective-bargaining agreements, protect incompetent members and prevent good teachers from being paid more.

This sometimes occurs, and when it does, it is troublesome. But a number of reform union leaders, going back to Al Shanker, have embraced “peer review” plans, which weed out bad teachers in Toledo, Ohio; Montgomery County, Maryland; and elsewhere. These reform plans put the lie to the notion that the average teacher has an interest in her union protecting incompetent colleagues. To the contrary, dead wood on the faculty makes every other teacher’s job more difficult. Likewise, numerous local unions have adopted pay-for-performance plans, when the measurement of performance is valid and incentives are in place to encourage good teachers to share innovative teaching techniques rather than hoarding them.

Moreover, many of the things that teachers collectively bargain for are good for kids. The majority of students benefit when teachers can more easily discipline unruly students, for example. (Principals, by contrast, often want to take a softer line so the school’s suspension rates don’t look bad.) Higher compensation packages attract higher-quality teacher candidates and reduce disruptive teacher turnover.

If collective bargaining were really a terrible practice for education, we should see stellar results where it does not occur: in the American South and in the charter school arena, for example. Why, then, aren’t the seven states that forbid collective bargaining for teachers (Arizona, Georgia, Mississippi, North Carolina, South Carolina, Texas, and Virginia) at the top of the educational heap? Why do charter schools, 88 percent of which are nonunion, only outperform regular public schools 17 percent of the time, as a 2009 Stanford University study found? Why, instead, do we see states like Massachusetts, and countries like Finland, both with strong teachers unions, leading the pack?

Opponents of collective bargaining will immediately point out that poverty rates are high in the American South, and low in Finland, which is an entirely valid point. But doesn’t that suggest that the national obsession with weakening teachers unions may be less important than finding ways to reduce childhood poverty?

Moreover, scholarly studies that seek to control for poverty find that collective bargaining is associated with somewhat stronger, not weaker, student outcomes. Sociologist Robert Carini’s 2002 review of 17 studies found that “unionism leads to modestly higher standardized achievement test scores, and possibly enhanced prospects for graduation from high school.” Even Terry Moe, an outspoken opponent of collective bargaining for teachers (see “Seeing the Forest Instead of the Trees,” book reviews, page 77), suggests that research on the impact of collective bargaining on student outcomes “has generated mixed findings (so far) and doesn’t provide definitive answers.”

For a variety of reasons, collective bargaining for teachers should not be constrained, much less eliminated. Indeed, if teachers are to be partners in innovative education reform (see “A Different Role for Teachers Unions?features, page 16), the scope of collective bargaining should be expanded. When the United Federation of Teachers first began to bargain collectively in the early 1960s, Albert Shanker was distressed that the New York City school board was willing to discuss only traditional issues like wages and benefits and rejected the idea of bargaining over broader policies that the union proposed, such as the creation of magnet schools.

Shanker saw that by reducing the scope of collective bargaining, critics created a political trap for unions. Union leaders were told they could only address bread-and-butter issues and then were criticized for caring only about their own selfish concerns rather than student achievement or larger policy issues. Moreover, Shanker believed that teachers had a lot of good ideas that could be incorporated into collective bargaining agreements, such as teacher peer review, suggestions for the types of curricula that work best in the classroom, and what sorts of programs would lure teachers into high-poverty schools. He also knew that reforms that draw on teacher wisdom are more likely to be effectively implemented when the classroom door closes.

In the end, Shanker’s frustration with the traditional constraints of collective bargaining spurred him to propose, in a 1988 speech at the National Press Club, the creation of “charter schools,” where teachers would draw upon a wealth of experience to try innovative ideas. Much to Shanker’s dismay, the charter school movement went in a very different direction, becoming a vehicle for avoiding unions and reducing teacher voice (and thereby increasing teacher turnover). And charters still educate a very small fraction of students.

Expanding collective bargaining for teachers to more states and to more education issues will give educators greater voice, and in so doing, indirectly strengthen the voice of students. Overall, the evidence suggests that Scott Walker has it exactly wrong, and the American public, which overwhelmingly supports the right to collective bargaining, has it right.

Jay P. Greene: Asking if teachers unions are a positive force in education is a bit like asking if the Tobacco Institute is a positive force in health policy or if the sugar lobby is helpful in assessing the merits of corn syrup. The problem is not that teachers unions are hostile to the interests of students and their families, but that teachers unions, like any organized interest group, are specifically designed to promote the interests of their own members and not to safeguard the interests of nonmembers. To the extent that teachers benefit from more generous pay and benefits, less-demanding work conditions, and higher job security, the unions will pursue those goals, even if achieving them comes at the expense of students. That is what interest groups do. Unfortunately, a public education system that guarantees ever-increasing pay and benefits while lowering work demands on teachers, who virtually hold their positions for life regardless of performance, harms students.

Collective bargaining is the primary vehicle through which the unions enact their preferred policies regarding pay, benefits, job security, and work conditions. It is also the mechanism by which unions collect fees from teachers that provide them with the resources to prevail politically. Until the ability of teachers unions to engage in collective bargaining is restrained, we should expect unions to continue to use it to advance the interests of their adult members over those of children, their families, and taxpayers.

Teachers unions only won the privilege of engaging in collective bargaining in the last 50 years, about when student achievement began to stagnate and costs to soar. A return to the pre–collective bargaining era may be the tonic our education system needs to return to growth in achievement and restraint in costs.

Jay P. Greene

The nature and function of organized interest groups is widely known and understood. Of course, there is nothing wrong with people organizing interest groups to advocate for themselves. That is an essential part of the freedom of assembly, protected by the U.S. Constitution. If people dislike what an interest group is advocating, they can organize other interest groups to compete in the marketplace of ideas and advocate for other concerns. The normal process of checks and balances among competing interest groups, however, has failed when it comes to education.

There are three factors that have contributed to the failure of other groups to check the power of teachers unions. First, there is an asymmetry in the ability of groups to organize in education, significantly favoring the teachers unions. Teachers unions have a huge advantage in organizing and advocating for their interests. Employees of the public school system are physically concentrated in school buildings, making it easier for them to organize. And because current employees are in a good position to know how they can benefit from the system, they can be mobilized relatively easily to advocate for those policies. Parents, taxpayers, and members of the general public are geographically dispersed, making it harder for them to organize. And because they are not immersed in education matters, they cannot easily envision how policy changes might help or hurt, making it harder to mobilize them on those issues. It is hardly unique to education that concentrated interests have an advantage over diffuse interests, but this is one factor contributing to teachers union dominance.

Second, teachers unions have fooled a large section of the general public and elites into thinking of them as something other than a regular interest group advocating for their own concerns.

The teachers unions have worked hard to convince people that they are a collection of educators who love our children almost as much as the parents do. They’re like the favorite aunt or uncle who dotes on our children. This image of the teachers unions as part of our family is facilitated by the fact that virtually every college-educated household (the households with the greatest political influence) has at least one current or former public school teacher sitting at the dining table when they gather for Thanksgiving. This impression is also fostered by ad campaigns featuring teachers buying school supplies out of their own pockets and movie portraits of heroic teachers believing in students, even as their parents have abandoned them.

Of course, some teachers really do buy school supplies with their own money (which should make people wonder what kind of education system would make that necessary after spending an average of more than $12,000 per student each year). And some teachers really are like the doting aunt or uncle who sticks with kids, even when the parents have given up. But loving children and being part of the family is certainly not what teachers unions are about. They are about accumulating the power necessary to advocate for the interests of their members. In a moment of candor, Bob Chanin, former general counsel of the National Education Association, explained the key to the union’s effectiveness: “Despite what some among us would like to believe, it is NOT because of our creative ideas. It is NOT because of the merit of our positions. It is NOT because we care about children, and it is NOT because we have a vision of a great public school for every child. NEA and its affiliates are effective advocates because we have power.”

The disarming image of teachers unions as Mary Poppins has begun to morph into that of a burly autoworker, as teachers union advocacy has become more militant in recent years. As states attempt to trim very generous benefit packages for teachers, the unions have organized large demonstrations, occupied state capitols, and chanted angry slogans. The public image of teachers unions fighting like autoworkers for the benefit to retire at 55 with full medical coverage and 66 percent of their peak salary while the economy is in shambles and the quality of their industry stagnates has done much to undermine the doting aunt or uncle meme. The angry slogans emanating from Diane Ravitch’s and Valerie Strauss’s Twitter feeds may soothe disgruntled teachers, but they are eroding the public perception that teachers unions are somehow different from other interest groups. Media and policy elites are increasingly treating teachers union claims with the same skepticism that they used to apply only to other interest groups.

A third factor is that unions have significant influence over who is elected or appointed to negotiate with them over pay, benefits, and work conditions. In the private sector, the power of unions is constrained by the competing organized interests of management. When they sit down to negotiate pay, benefits, and work conditions, members of management are inclined to represent the interests of shareholders, not those of employees. But in education, as in other public-sector collective bargaining, the interests of employees are represented on both sides of the table. The employees, as citizens, can organize, finance, and vote for elected officials who favor the union’s interests. It is precisely for this reason that public employees historically did not have collective bargaining rights.

But didn’t the lack of collective bargaining rights sometimes leave teachers vulnerable to arbitrary and discriminatory treatment by school administrators? Yes, but unionization and collective bargaining were neither necessary nor efficient means of correcting those abuses. We can look to other public employees, such as members of the armed forces, who still do not have collective bargaining rights, to see how progress could have occurred without unionization. The military, like public schools, was once racially segregated. African American servicemen and servicewomen were treated horribly. And sometimes officers treated all soldiers in an arbitrary and unfair manner. These abuses were not corrected by unionization and collective bargaining in the military. They were corrected by executive orders and changing legislation governing those public employees. The same path could have been taken with public school employees without the political distortions that public employee unions introduce by virtue of having their interests represented on both sides of the bargaining table.

It may have taken longer than many would like to integrate the military, expand the roles of women in the armed forces, and end “don’t ask, don’t tell,” but we were able to achieve all of those through an open, public process of changing laws and regulations. Unionized collective bargaining might also have addressed those issues, but it would have been done mostly behind closed doors and would have been accompanied by provisions to protect the narrow interests of the unions at the expense of the public interest. Perhaps the use of drones would have been restricted because it displaces jobs for Air Force pilots; perhaps there would be caps on the hours soldiers could engage in combat. Who knows what else a unionized military might have produced? The point is we rightly restrict the ability of members of the armed forces from unionizing and engaging in collective bargaining, just as we once did and could again for teachers. The claim that public employees have a “right” to unionize and collectively bargain and that exercising this “right” necessarily advances the public interest is obviously false.

The proper mechanism for improving compensation and work conditions in the public sector is through changes in law and regulation. The salary, benefits, job security, and work conditions of public employees are just as much a matter of public policy as the work that those employees are supposed to do. We don’t allow smoky backroom deals arrived at in collective bargaining to dictate the goals, structure, or existence of the public education system, so neither should we use that process to determine compensation and work condition policies.

What evidence is there that teachers unions have actually had negative effects on students and the education system? The research literature generally finds that unionization is associated with higher per-pupil costs and lower student achievement, but those findings are not very large and are sometimes inconsistent. A 1996 article by Caroline Hoxby in the Quarterly Journal of Economics is widely considered the most methodologically rigorous analysis of the issue. Claremont Graduate University professor Charles Kerchner described Hoxby’s study in a literature review prepared for the National Education Association as “the most sophisticated of the econometric attempts to isolate a union impact on the student results and school operations …” Hoxby finds that unionization is associated with higher student dropout rates as well as higher spending.

But the reality is that it is very hard to produce rigorous research on the effects of teachers unions on education. For one thing, teachers unions are powerful and active almost everywhere. Even in states without collective bargaining, the unions push state legislatures to put into law what is normally put into collective bargaining agreements. This is less than ideal for the unions, because they don’t collect dues in exchange for pushing through legislation like they can for representing members to achieve the same ends through collective bargaining. Unions operate these money-losing operations in right-to-work states to make sure that there is no meaningful policy variation on their key issues. They’d rather that we not discover that the world does not end without a mandatory step-and-ladder pay scale, fair dismissal procedures, and favorable work rules. The lack of policy variation hinders researchers, because outcomes are not likely to be very different where the policies are not very different.

But we don’t need a wealth of evidence on teachers unions specifically as long as we know about the effects of interest groups and recognize that teachers unions are indeed interest groups. Seeking to produce evidence on the effects of each interest group separately, especially when there are empirical challenges to doing so, is a bit like trying to prove that gravity operates in every room of a house. We could drop a bowling ball in each room to see if it hits the floor, but sometimes there are tables, couches, or beds in the way. If we don’t get the result we expected, it doesn’t mean that gravity only applies in certain places; it just means that research constraints prevent us from seeing in a particular situation what we know to be true in general.

In general, we know that interest groups advocate for the benefits of their members, even if it comes at the expense of others. We know that teachers unions are interest groups. And we know that the interests of teachers unions are not entirely consistent with the needs of students and taxpayers. Thus, teachers unions are likely to be negative forces for the education system and certainly should not be seen as helpful. The most rigorous research that does exist bears this out, but we also know this from our more general knowledge of how interest groups affect policy.

It is not currently practical to forbid the unionization of teachers, as we forbid the unionization of members of the armed forces. But if we want to limit the ability of teachers unions to advance their own interests at the expense of children, their families, and taxpayers, we need to consider ways of restricting their ability to engage in collective bargaining. Restricting collective bargaining would force teachers unions to pursue their interests through the legislative process, where competing interests might have a better chance to check their power. And forcing unions to operate through legislation rather than backroom collective-bargaining negotiations would improve transparency, which could also place a check on the unions’ ability to satisfy their own interests at the expense of others.

RDK: Jay Greene’s opening line, comparing teachers unions to the Tobacco Institute, is very telling about his overall analysis. He’s right, of course, that both are “interest groups,” but does he not see a massive difference between an entity that is devoted to getting more kids addicted to deadly cigarettes so they’ll be lifelong clients and a group representing rank-and-file teachers whose life’s work is educating children?

Greene complains that teachers unions have become “more militant in recent years.” But teacher strikes, which were quite common in the 1960s and 1970s, dropped 90 percent by the mid-1980s and are now, as one education report noted, essentially “relics of the past.” To the extent that teachers have rallied, it’s in response to unprecedented attacks on them in places like Wisconsin, where a half century of labor law was radically rewritten. Astonishingly, Greene would go further than Wisconsin Republicans and “return to the pre–collective bargaining era.”

Greene says providing teachers with better pay and benefits is bad for kids, but where is his evidence? Don’t better compensation packages attract brighter talent, or are the laws of supply and demand suddenly suspended when it comes to teachers?

Finally, Greene is correct to suggest that teacher and student interests are not perfectly aligned, but who are the selfless adults who better represent the interests of kids? The hedge fund managers who support charter schools and also want their income taxed at lower rates than regular earned income, thereby squeezing education budgets? Superintendents who sometimes junk promising initiatives for which they cannot take credit? I’d rather place my faith in the democratically elected representatives of educators who work with kids day in and day out.

JPG: Richard Kahlenberg places his faith in “democratically elected representatives of educators,” that is, the teachers unions, to safeguard the interests of children. Note that he does not say the democratically elected representatives of the people, or the voters. Kahlenberg is perfectly comfortable with a school system whose policies and practices are dominated by its employees, not by the citizens who pay for it or by the families whose children are compelled to attend it. Rather than seeing a system controlled by its employees as one characterized by self-interested adults maximizing their benefits at the expense of children, Kahlenberg sees it as the ideal.

In my ideal vision, we would put our faith in parents, not teachers unions, to represent the interests of children. If we had a robust system of parental school choice, I would have no problem with teachers unions and collective bargaining. In the private sector, if unions ask for too much, at least they experience the natural consequences of destroying their own companies or industries (to wit, the auto industry). But in the public sector, unions are almost entirely insulated from the consequences of making unreasonable demands, since governments never go out of business. Public sector unions can drive total revenue for their industry higher without any improvements in productivity simply by getting public officials to increase taxes.

Unfortunately, we lack a robust system of school choice and instead have to rely on democratic institutions, like school boards and state legislatures, to determine most school policies and practices. But unless we also restrict the collective bargaining rights of school employees, teachers unions will dominate the decisions of those democratic institutions, given their advantages in funding and organization, to distort elections and policy decisions.

Teachers unions almost certainly raise salaries and benefits, as Kahlenberg suggests, but that doesn’t necessarily attract better teachers if the salary schedule does nothing to reward excellence. Similarly, union-imposed dismissal procedures make it virtually impossible to fire ineffective teachers. The alignment that Kahlenberg sees between teachers unions’ desire to increase education spending and the interests of students would only be a real concordance if the unions facilitated the use of those funds in ways that actually improved outcomes.

This article appeared in the Winter 2012 issue of Education Next. Suggested citation format:

Kahlenberg, R.D., and Greene, J.P. (2012). Unions and the Public Interest: Is collective bargaining for teachers good for students? Education Next, 12(1), 60-68.

Last Updated


Notify Me When Education Next

Posts a Big Story

Business + Editorial Office

Program on Education Policy and Governance
Harvard Kennedy School
79 JFK Street, Cambridge, MA 02138
Phone (617) 496-5488
Fax (617) 496-4428
Email Education_Next@hks.harvard.edu

For subscription service to the printed journal
Phone (617) 496-5488
Email subscriptions@educationnext.org

Copyright © 2024 President & Fellows of Harvard College