Terry Moe entertains and appalls in the opening pages of Special Interest with the tale of the New York City School District's infamous “Rubber Room.” As late as 2009, because they could not be fired, teachers were sequestered from students in a room where they could squabble, snooze, read a newspaper, or practice a ballet step as they pleased, all the while drawing full salary and benefits. Except for the expensive Rubber Room option, Moe explains, it is virtually impossible to get bad public school teachers out of the classroom all across the country. Thus, he argues, because of this and other contractually enforced terms of employment, American public education gets a poor grade. Teachers' unions, he claims, are at the heart of the matter. Because of their power, the “basic requirements cannot be met” for educational success (p. 342). Specifically, unions stand in the way of systematic free choice of charter schools and accountability of teachers (for their students' test scores), the author's sine qua non of reform.

Moe's jeremiad against teachers' unions should not convince anyone except those already converted to his favorite reforms. The problem starts with his neglect of history. Consider New York City again—back in 1893: “In New York city teachers are rarely discharged, even for the grossest negligence and incompetency.” So wrote turn-of-the-century reformer J. M. Rice.Footnote ¹ At least in New York, removing teachers for incompetency was extremely difficult even before passage of formal tenure protections. New Jersey came first with a tenure law in 1909. By 1931, 13 nonsouthern states, New Orleans, and the District of Columbia had them.Footnote ² By the 1940s, roughly 70% of public school teachers across the entire country were already enjoying strong dismissal protections, partly to shut down patronage-based hiring and firing.Footnote ³ In other words, tenure predated unionism, which appeared first in the 1960s.

It would be no simple matter to prove that tenure protections today are dramatically stronger than before unionism. In 1932, New Jersey school officials reported extreme aggravation with “undesirables” they could not dislodge. One New Jersey observer noted that “the means of removing a teacher for anything but immoral conduct is so difficult that a superintendent will permit mediocre teachers to remain undisturbed year after year rather than to take steps toward removal.”Footnote ⁴ A 1934 multistate study found that only nine “educational workers” (possibly including nonteachers) had been dismissed in Chicago in the 12 years up through 1931, and rarely for incompetency alone. In New Jersey, Trenton did not fire a single teacher and Newark dismissed only one during the five-year period ending 1931. In Minnesota, Duluth reported no removals in the handful of years following passage of the state dismissal law in 1927, and Minneapolis only four. In Oregon, from 1913 to 1927, only nine teachers were permanently removed.Footnote ⁵

Reasons for the introduction of tenure in the first place pose the question about their net effects, a matter that Moe does not address. Is there a baby in the bathwater? Early proponents argued for tenure protection for the same reason they demanded better salaries and pensions. From the standpoint of labor recruitment, these reasons may still be valid. Moe himself finds that teachers attach a large dollar value to tenure, and is fully aware that teaching, despite this perk, does not attract anything near the best and brightest. Typical, probably, was a 1905 claim that the shortage of qualified teachers could only be remedied by “adequate inducements to men and women of energy and intellectual capacity.” Men, especially, were diverted into other professions with better and “more assured prospects.”Footnote ⁶ Tenure, according to one observer, became “one of the magnets which now draw men and women into the Public School system from all over the country, and communities, which do not provide it, will always be below the mark.” In states without tenure, he said, because of low pay and the politics of teacher recruitment, “multitudes of young people look on the expression ‘the Teaching Profession’ as something of a joke.”Footnote ⁷

By neglecting history, Moe allows unsuspecting readers to walk away from his book thinking that unions were present and responsible at the origins of America's public school problems. A related evasion, his refusal to delve into the question of unionism's share of the overall blame, may lead the unsuspecting reader to think that unions are the only problem.Footnote ⁸ Here, a comparative as well as historical perspective was needed. Key findings of the 2009 Program for International Student Assessment (PISA) study of 34 highly developed and unionized countries of the Organization for Economic Co-operation and Development (OECD) suggest that American unions may be a secondary problem relative to other factors that Moe barely mentions. Overall, according to the PISA study, U.S. 15-year-olds perform at about average on various measures. (On the “reflect and evaluate” dimension of reading literacy, they even perform distinctly better than those in 23 OECD countries, and worse than those in only five countries.) In science literacy, American students tested about average, and only in mathematics were they distinctly below.Footnote ⁹ By no means do these averages say that we have nothing to worry about, but they indicate that we cannot point an accusing finger at unions without comparative analysis of how collective bargaining and school legislation works in other countries.

Breakdown of the PISA figures points at something else—the gross and increasing inequalities in American society. For example, students in public schools in which less than 25% of them were poor, as measured by eligibility for free or reduced-price lunches, performed substantially above the OECD average.Footnote ¹⁰ Moe's previous research should have prompted him to examine this question in great detail. For support of his lengthy indictment of unions, he cites his own valuable and interesting article, published elsewhere, linking contract restrictiveness and changes in student performance over a small number of years in California. In the book he claims his study showed that the effect of tight contractual control “appears to be huge” (p. 213, original emphasis). However, in his article, the results were not so eyebrow raising. There he reported not huge but “very negative” effects—quite a different thing. And they were not even very negative across the board. In fact, to quote the article (my italics):

Moe's own findings, ignored in his book, suggest that the problem is an interaction between only some aspects of union contracts and already very problematic school districts. What matters most in these districts, it seems, are seniority-linked provisions giving senior teachers priority over more desirable openings in their districts, and the right to reject transfer to other jobs where superintendents and principals think they are needed or better suited. Tenure, strangely, despite all of Moe's attention to it in his book, actually emerged as one of the least important factors, and, even more strangely, it was distinctly outweighed in its effects by “guarantees of teacher preparation time.” Puzzlingly, even limits on class size worked against student performance more than did tenure.

Neither Moe's book nor his article examines why seniority-based transfer provisions could matter so much in large districts dominated by blacks and Latinos, but not in other California districts. In the problematic districts, teacher and probably student turnover rates are very high. About a third of teachers leave the profession within three years, and about half do so in five years.Footnote ¹² Certainly the figure is far higher in the poor and diverse California districts. Maybe what happens is a rapid sorting process in which more senior teachers, including most of the good ones, migrate into the more stable and overall less stressful schools with more well-adjusted, disciplined, and native-English–speaking students, while the more difficult students are stuck with young, inexperienced, and very transient teachers—often quick burnouts. Keeping teachers, not getting rid of them, seems to be the big problem. In any case, the problem is not unions and collective contracts per se but the interaction of certain provisions in the contracts with unstable conditions prevalent in school districts already suffering from the pathologies of poverty, linguistic deficiencies, inadequate pay for the emotionally bruising conditions, and more.

Moe's exclusive focus on the rise of unions and collective bargaining after the 1960s through the 1980s will lead the unsophisticated reader to think that the growing problems in American public education during that period (exposed in A Nation at Risk in 1983) were clearly caused by unions. The truth, of course, is that the socioeconomic and related educational problems, especially in the cities, were already there. Furthermore, there was the economic calamity of deindustrialization and lower-class ghettoization, analyzed by William Julius Wilson, which hit urban industrial America in the 1970s and onward, causing damage that unions had nothing to do with.Footnote ¹³ Even so—interestingly enough—some educational improvements occurred during the supposed assault of unionism. Moe does not ask why, if unions were such a menace, the black–white achievement gap actually narrowed in the 1970s and 1980s before rising somewhat again in the 1990s.Footnote ¹⁴

Again, all of this starkly poses the question that Moe studiously evades—unions' relative share of the blame. A very large and scholarly book on teachers' unions cries for at least a few pages out of a total of 386 on their interaction with socioeconomic factors and trends, and serious debate with scholarship that points fingers in different directions. Why he failed to do this is more than puzzling. In the same article cited previously, he wrote that “socioeconomic status, ethnicity, language problems … are the key determinants of achievement.”Footnote ¹⁵ In his book, however, the tune changes completely. There, he begrudgingly concedes that there is “some truth” in the arguments about social factors (p. 12), only to ignore them the rest of the time. For ignoring them, he excuses himself by declaring that “the task is not to capture everything of any relevance.” Instead, “It is to get to the heart of the matter” (p. 5). But if socioeconomic factors are “key determinants” of achievement—as he once believed—then they are the heart of the matter.

According to a perhaps imperfect count Moe mentions socioenvironmental conditions only one more time. Here, opportunistically, he relies on them to dismiss the unions' efforts to defend themselves with data showing that performance is worse in southern and border states with less collective bargaining, or that schoolchildren in Finland perform well where strong unions and nationwide collective bargaining prevail. Only here does Moe finally say, in reference to Finland, that along with ethnic homogeneity and “virtual absence of poverty,” “investment in education, recruitment of teachers, family structure … and so on—are simply being ignored” (p. 208). So unions are at fault for ignoring the socioeconomic factors, not Moe? Pot calls kettle black.

Strange it is indeed, regarding Finland, that mention of investment in education and recruitment of teachers accompanies discussions in which Moe declares, mostly following the research of Hoover Institution economist Eric Hanushek, that “inputs” like class size matter little, and that teachers seem more than adequately paid if working hours, health and pension benefits, and tenure are taken into account. Some of Hanushek's methods and findings have been forcefully challenged by economist Alan Krueger, but Moe does not cite, much less argue with, him.Footnote ¹⁶

In fact, deep in the endnotes (p. 454), Moe stuns the reader with the announcement that he chose to relegate his limited discussion about achievement impacts, especially of reforms, to the notes. His reason: “the data and existing studies rarely allow for definitive conclusions, and as a result these discussions threaten to distract from the main points I want to make in this book” (my emphasis). So here we learn that reform impacts on achievement are not Moe's “main point.” This comes as a huge surprise to the reader of a lengthy book that sums up with the confident conclusion that “as long as the teachers unions remain powerful,” and therefore block charter schools and test-based personnel decisions, the “basic requirements” of educational achievement “cannot be met” (p. 342).

More persuasive, but actually unsupportive of that particular summation, are two of Moe's other main points. The first, based on primary research elaborated at length, is that teachers' unions reflexively defend tenure, seniority, and other contract and legal restrictions because that is what their members want, even members who identify themselves as Republicans. Leaders, he says, are tightly constrained, because if they caved in to reformers, they would lose their jobs. The other finding, based largely on accounts in the press, is that these union leaders have strenuously and very effectively obstructed the “accountability” (testing, performance pay, and dismissal for bad performance) and “choice” (vouchers and nonunion charter schools) reform efforts of the last two decades, which Moe fervently advocates. On these two issues he offers a good account.

Some readers may object to the often sarcastic tone of Moe's discussion about unions' resistance to reform in the collective bargaining and political arenas. When unions make accountability and choice concessions, they are only being “bought off” (p. 253) with “pots of gold” (p. 254) and other “incredibly expensive” (p. 239) things like “whopping” salary increases (p. 255). Money is just “shoveled at teachers” (p. 236), and reforms that do not introduce accountability and choice “amount to little more than tons of additional money … pumped … into high-poverty districts” (p. 350).

Moe's gratuitous, ideologically tinged sentiments harmonize tellingly with his characterization of the public sector union AFSCME (American Federation of State County and Municipal Employees) as a “monster political force” (p. 295) or unionized auto workers as spoiled with “sky-high wages and platinum benefit packages” (p. 373). Interestingly—one is inclined to say predictably—Moe favors doctors, a high-income group, with the flattering and generally inaccurate observation that their economic success depends on “[doing] a good job” (p. 205). The truth is that because of fee-for-service (piece-rate) payment, ferociously defended in the past by the American Medical Association, our expensive doctors (and hospitals) gain more from the quantity of specialized, high-tech care than from quality-of-health outcomes.Footnote ¹⁷

These discussions slant away from Moe's posture of objectivity in his declaration that he has no decided view on whether teachers are adequately paid or on the sincerity of reform unionists' declarations of support for accountability and choice. Unabashedly, Moe dismisses reform unionism as “completely wrong-headed” (p. 242) and declares that the idea that some union leaders are open to change is “false at its core” (p. 263). Nonunion education reformers, “mesmerized” (p. 271) by reform unionists' “sweet songs” (p. 272), naively “[jump] … out of their skins with excitement” (p. 273) and “happily link arms” with them to “walk down the garden path” (p. 274). The sarcasm about reform unionism shoulders aside the staid rational-choice institutionalism that dominates at the beginning of his book.

Today's enemies of teacher and other public sector unions may—to recycle an expression—jump out of their skins with excitement about Special Interest, but not because of its scholarly merits. To the extent that Moe, as Frederick Hess of the American Enterprise Institute puts it, did “meticulous research,” it applies only to his efforts to show that unions are responsible for blocking his favored reforms, and in doing so, that their leaders are doing members' self-interested bidding. This is a useful and interesting but rather modest accomplishment. But the book fails drastically in demonstrating that unions are the heart of our public education problems, for three reasons: first, because unions are not the original and exclusive cause of these problems; second, because Moe consciously refuses to assess unions' relative contribution to the problems; and third, because he absolutely fails to demonstrate—because he cannot—that the reforms that unions block are sufficient and necessary solutions to our educational predicament, regardless of who and what caused them in the first place.

In sum, only true believers in choice and accountability reforms will be propelled by Moe's findings about union leaders' political behavior to the conclusion that unions are the central problem of public education in America. Other readers, even those like this one, who are no fans of the tenure system for public school teachers, and who are open to persuasion about well-designed choice and accountability reforms—not those that result in “teaching to the test” and other corruptions—will have to remain undecided.Footnote ¹⁸ There are good reasons for hesitation about market-inspired reform, given how often information-deficient choice settings produce perverse results. The evidence in favor of charter schools is, so far, not strong. And recent findings suggest that information technology, the rainbow on the horizon in Moe's exuberant view—it will even, he says, “loosen the unions' iron grip” (p. 385)—is not helping. Profit-driven, non-union, internet-based “virtual schools” may be even worse than the old brick-and-mortar versions, especially for poor students. But not surprisingly, Wall Street visionaries of free-market school reform are heavily promoting this technological fix, considering the huge pot of taxpayer gold at the rainbow's end.Footnote ¹⁹

A persuasive but also more difficult book to write on teachers' unions would have to put America's relative educational performance in a more sophisticated and ideally comparative context. It would need to evaluate student achievement with all key variables, including socioeconomic factors, along with teacher training, pay, and other recruitment factors. My guess is that the results would show that unionism is only one of many vital organs of the matter, and thus contradict this splenetic attack.