Vindication for Scott Walker on Teachers’ Unions 

W hen Wisconsin Republicans led by Governor Scott Walker wanted to reform collective bargaining for state employees in 2011, the public-sector unions said the reforms would destroy public education. I would know. I was in public school in Wisconsin at the time. I heard all the propaganda and saw teachers miss school because they called out “sick” to go protest. 

The unions mobilized a national campaign against Walker, collecting signatures to recall him from office and occupying the state capitol in an effort to prevent the reforms from passing. Fourteen Democratic state senators fled to Illinois to deny a quorum. In the end, the reforms passed (“One Walker Beats 14 Runners,” the bumper stickers said), the budget was balanced, and the sky didn’t fall. Walker won his recall election by a larger margin than in his first election. 

One of the many reasons Walker gave for supporting the reforms is that teachers’ unions protect bad teachers and fail to reward good ones. He gave the example of a teacher who won a statewide award for first-year teachers yet was laid off right after receiving it because of “last in, first out” seniority rules in the union. Reducing unions’ power would increase accountability for teachers, which would help students. 

Act 10, the reform law, made union dues and agency fees voluntary and restricted collective bargaining to wages, meaning unions could no longer bargain for benefits or work rules such as seniority. It also required unions to submit to annual recertification votes. 

This major change in the rules surrounding unions, and the years’ worth of data that has now accumulated since the change, presents a good opportunity for researchers to study unions’ effects. Morgan Foy of the University of California, Berkeley has a new paper that does just that. 

He notes that after Act 10 became law, the teacher unionization rate dropped from over 98 percent to under 50 percent, and about half of Wisconsin’s 400 local teachers’ unions failed their recertification votes. That means today there are roughly equal populations of union and non-union teachers in Wisconsin public schools to compare with each other. 

Foy uses individual-level data on each teacher and links them to their students’ performance. He was able to determine which teachers were union members by consulting campaign-finance data because the teachers’ union automatically contributes part of each members’ dues to its PAC. He constructs an econometric measure of value added for teachers based on their students’ standardized test scores between 2017 and 2019, after Act 10 had been in effect for several years and before Covid. 

Foy finds that “lower value-added teachers are more likely to be union members than higher value-added teachers within the same experience profile, consistent with union representation being more valuable for lower performers.” In other words, once Wisconsin teachers had the choice whether to be union members or not, the worse ones were more likely to stay in than the better ones. 

If the unions’ doomsday predictions from 2011 were correct and the absence of unions would hurt students’ performance, then we would expect the districts where unions were decertified to have worse results than the ones that decided to keep their union. Instead, Foy finds that “decertification of the local teachers’ union led to a significant gain in student test scores.” He also finds that student attendance rates increased in districts that decertified. 

Test scores improved for students of all races in decertified districts, and the increase for black students was larger than for white students. The improvement in attendance rates was “especially strong” for students enrolled in free and reduced-price lunch programs and non-white students. In general, “effects were not more or less pronounced in lower-income or larger districts,” Foy finds. 

Based on estimates from other research, Foy writes that the gain in test scores from decertification was roughly equal to increasing per pupil spending by 8 percent, or about $1,000. There’s no such thing as a free lunch, but this is about as close as you can get to one in real life. Not only did the districts get the results of more spending without having to actually spend more, they were in many cases able to spend less because Act 10 allowed them to shop around for health insurance rather than paying monopoly union prices. 

“I fail to find any evidence of exit or entry differences post-decertification on average,” Foy writes, torpedoing the union claim that teachers would leave Wisconsin schools if forced unionism wasn’t kept in place. He also finds that decertification had no effect on teacher salaries and no effect on per-pupil spending at the district level. 

Though Foy did not find evidence of more pay-for-performance salary structures, he did find evidence that pay was less based on seniority in districts that decertified. That means they “raised the floor of teacher pay for inexperienced teachers at the relative expense of high-tenured teachers.” One hopes the old union model of new, energetic teachers barely scraping by while old, lazy teachers are raking it in will someday be a thing of the past. 

In addition to this data analysis, Foy also sent a survey to Wisconsin public-school teachers earlier this year. “Overwhelmingly, the most cited reason for being a union member is ‘representation in case of conflict’ with about 85 percent selecting the option,” Foy finds. If you’re good at your job, you’re probably less worried about conflict than someone who is bad at his job. Given that union membership is voluntary, the fact that the teachers who chose to stay in overwhelmingly named representation in case of conflict as their reason why further suggests that the worse teachers stayed in and the better teachers left, on average. 

The survey also found that there was no difference between teachers in districts with certified unions and those in districts that decertified on satisfaction with work environment, salary, or job security. The only difference is that teachers in decertified districts were more likely to say they had greater control over setting their curriculum than those in districts with certified unions, which “may accord with the idea that the districts became more flexible organizations after the union decertified,” Foy writes. 

There are some details in Foy’s paper that are illustrative of public-sector unions’ shenanigans. For example, he notes that before 2016, PACs in Wisconsin were not required to separately report individual donations of less than $20. Guess how much the state NEA chapter automatically deducted from each member’s dues to contribute to its PAC? Yep: $19.99. 

Of course, this is one paper, and others could find different results. But give Foy credit for being willing, in an academic setting, to thoroughly examine the effects of teachers’ unions on education. It is very difficult to get straight research on this question because education schools are dominated by teachers’ unions. Foy is at the business school at Berkeley, so he doesn’t have to deal with that influence as much. 

The bigger benefits of Act 10 were to the Wisconsin state budget and pension system, which are in substantially better shape than they were before the reforms. But students in districts with decertified teachers’ unions had slightly better test scores and attendance rates, contrary to the unions’ scaremongering and vindicating Walker’s brave stand against them.